Modulators of cell surface protein interactions, and methods and compositions related thereto

Abstract

To provide: methods for identifying modulators of activity and cell surface protein interacting with PD-L1 protein; and methods of identifying an individual who may benefit from a treatment based thereon.SOLUTION: A method comprises determining, in a sample from an individual, expression levels of a first member and a second member of at least one pair of genes in a specific gene group, where an expression level of the first member of the gene pair that is above a first reference expression level and an expression level of the second member of the gene pair that is above a second reference expression level identify the individual as an individual who may benefit from a treatment including a PD-L1 axis binding antagonist.SELECTED DRAWING: None

Description

[Technical Field]

[0001] Sequence Listing This application contains a Sequence Listing that has been submitted electronically in ASCII format, the entire contents of which are incorporated herein by reference. The ASCII copy was created on March 27, 2020, is named 50474-182WO3_Sequence_Listing_3.27.20_ST25, and is 24,492 bytes in size.

[0002] FIELD OF THE INVENTION Provided herein are modulators of cell surface protein interactions and activities, and methods for identifying modulators of cell surface protein interactions and activities. [Background technology]

[0003] background Recent annotations predict that over 5,000 genes in the human genome encode proteins expressed or secreted at the plasma membrane, i.e., proteins that function at the cell surface. Classes of proteins present on the cell surface include cell surface receptors (e.g., single-transmembrane (STM) receptors) and immunoglobulin superfamily (IgSF) proteins. Many of these proteins have been implicated in cancer and other diseases and are therefore prime targets for therapeutic development. However, the binding partners of many plasma membrane proteins (e.g., STM and IgSF proteins) remain uncharacterized. This discrepancy is due to inconsistencies in currently used proteomic techniques to study plasma membrane-expressed proteins, such as yeast two-hybrid assays and affinity purification / mass spectrometry (AP / MS). High-throughput screening of extracellular protein-protein interactions has recently been developed. However, therapeutic agents that modulate the majority of these interactions have yet to be identified.

[0004] Thus, there is an unmet need for methods and compositions for the identification of interactions between cell surface proteins, as well as for novel modulators of such interactions and methods of using the same. Summary of the Invention

[0005] Summary of the Invention The present invention provides modulators of cell surface protein interactions and activities, and methods for identifying modulators of cell surface protein interactions and activities.

[0006] In one aspect, the disclosure features a method of identifying an individual having cancer who may benefit from a treatment comprising a PD-L1 axis-binding antagonist, the method comprising determining in a sample from the individual the expression levels of a first member and a second member of at least one gene pair of Table 15, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual as one who may benefit from a treatment comprising a PD-L1 axis-binding antagonist.

[0007] In another aspect, the disclosure features a method of selecting a treatment for an individual having cancer, the method comprising determining in a sample from the individual the expression levels of a first member and a second member of at least one gene pair of Table 15, wherein an expression level of the first member of the gene pair greater than a first reference expression level and an expression level of the second member of the gene pair greater than a second reference expression level identifies the individual as one who is likely to benefit from treatment comprising a PD-L1 axis-binding antagonist.

[0008] In some embodiments, the individual has an expression level of a first member of the gene pair that is greater than a first reference expression level and an expression level of a second member of the gene pair that is greater than a second reference expression level, and the method further comprises administering to the individual an effective amount of a PD-L1 axis-binding antagonist.

[0009] In another aspect, the disclosure features a method of treating an individual having cancer, the method including: (a) determining the expression levels of a first member and a second member of at least one gene pair of Table 15 in a sample from the individual, wherein the expression level of the first member of the gene pair is greater than a first reference expression level, and the expression level of the second member of the gene pair is greater than the second reference expression level; and (b) administering to the individual an effective amount of a PD-L1 axis-binding antagonist.

[0010] In another aspect, the disclosure features a method of treating an individual having cancer, the method including administering to the individual determined to have an expression level of a first member of a gene pair in Table 15 that is greater than a first reference expression level and an expression level of a second member of the gene pair that is greater than a second reference expression level a PD-L1 axis-binding antagonist.

[0011] In another aspect, the disclosure features a method of identifying an individual having cancer who may benefit from a treatment other than, or in addition to, a PD-L1 axis-binding antagonist, the method comprising determining the expression levels of the first member and the second member of at least one gene pair of Table 16 in a sample from the individual, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual as one who may benefit from a treatment other than, or in addition to, a PD-L1 axis-binding antagonist.

[0012] In another aspect, the disclosure features a method of selecting a treatment for an individual having cancer, the method comprising determining the expression levels of a first member and a second member of at least one gene pair of Table 16 in a sample from the individual, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual as one that may benefit from treatment other than, or in addition to, a PD-L1 axis-binding antagonist.

[0013] In some aspects, the individual has an expression level of a first member of the gene pair that is greater than a first reference expression level and an expression level of a second member of the gene pair that is greater than a second reference expression level, and the method comprises administering to the individual an effective amount of a therapy other than, or in addition to, a PD-L1 axis-binding antagonist.

[0014] In some embodiments, the first reference expression level is a pre-assigned expression level and the second reference expression level is a pre-assigned reference expression level.

[0015] In some embodiments, the sample from the individual is obtained from the individual prior to administration of the anti-cancer therapy, hi some embodiments, the sample from the individual is obtained from the individual after administration of the anti-cancer therapy.

[0016] In some embodiments, the sample from the individual is a tumor tissue sample or a tumor fluid sample.

[0017] In some embodiments, the sample is a formalin-fixed, paraffin-embedded (FFPE) sample, an archived sample, a fresh sample, or a frozen sample. In some embodiments, the tumor tissue sample is an FFPE sample.

[0018] In some embodiments, the expression levels of the first and second members of the gene pair in the sample are protein expression levels; or the expression levels of the first and second members of the gene pair in the sample are mRNA expression levels. In some embodiments, the expression levels of the first and second members of the gene pair in the sample are mRNA expression levels of the first and second members of the gene pair, respectively.

[0019] In some embodiments, the mRNA expression levels of the first and second members of the gene pair are determined by in situ hybridization (ISH), RNA-seq, RT-qPCR, qPCR, multiplex qPCR or RT-qPCR, microarray analysis, SAGE, MassARRAY technology, FISH, or a combination thereof. In some embodiments, the first reference expression level is between about 0.25 and about 0.5 counts per million (CPM), and the second reference expression level is between about 0.25 and about 0.5 CPM. In some embodiments, the first reference expression level is 0.25 CPM, and the second reference expression level is 0.25 CPM.

[0020] In some embodiments, the first reference expression level and the second reference expression level are the expression levels of the first member and the second member of the gene pair, respectively, in a reference population of individuals with cancer. In some embodiments, the cancer is a urinary tract cancer, e.g., a urinary tract cancer, e.g., locally advanced urothelial carcinoma or metastatic urothelial carcinoma (mUC).

[0021] In some embodiments, the benefit comprises an increase in overall survival (OS) of the individual compared to treatment without the PD-L1 axis-binding antagonist.

[0022] In some embodiments, the first member of the gene pair is SIGLEC6 and the second member of the gene pair is NCR1.

[0023] In some embodiments, the first member of the gene pair is BTN3A1 and the second member of the gene pair is LRRC4B.

[0024] In some embodiments, the first member of the gene pair is CD80 and the second member of the gene pair is CTLA4.

[0025] In some embodiments, the first member of the gene pair is BTN3A3 and the second member of the gene pair is LRRC4B.

[0026] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is TRHDE.

[0027] In some embodiments, the first member of the gene pair is CTLA4 and the second member of the gene pair is PCDHGB4.

[0028] In some embodiments, the first member of the gene pair is CTLA4 and the second member of the gene pair is FAM200A.

[0029] In some embodiments, the first member of the gene pair is CA12 and the second member of the gene pair is SIGLEC6.

[0030] In some embodiments, the first member of the gene pair is ILDR2 and the second member of the gene pair is CLEC12B.

[0031] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is ITLN1.

[0032] In some embodiments, the first member of the gene pair is CADM1 and the second member of the gene pair is CRTAM.

[0033] In some embodiments, the first member of the gene pair is CD79B and the second member of the gene pair is CD244.

[0034] In some embodiments, the first member of the gene pair is DAG1 and the second member of the gene pair is EFNB1.

[0035] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is EVC2.

[0036] In some embodiments, the first member of the gene pair is GPC4 and the second member of the gene pair is FGFRL1.

[0037] In some embodiments, the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB4.

[0038] In some embodiments, the first member of the gene pair is PTPRD and the second member of the gene pair is LRFN4.

[0039] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is AQPEP.

[0040] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is DSG4.

[0041] In some embodiments, the first member of the gene pair is LDLR and the second member of the gene pair is LILRB5.

[0042] In some embodiments, the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB3.

[0043] In some embodiments, the first member of the gene pair is PLXNB3 and the second member of the gene pair is SEMA4G.

[0044] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is EPHB6.

[0045] In some embodiments, the first member of the gene pair is FLT4 and the second member of the gene pair is FLRT2.

[0046] In some embodiments, the first member of the gene pair is AXL1 and the second member of the gene pair is IL1RL1.

[0047] In some embodiments, the first member of the gene pair is CD320 and the second member of the gene pair is IGSF5.

[0048] In some embodiments, the first member of the gene pair is CD59 and the second member of the gene pair is STAB1.

[0049] In some embodiments, the first member of the gene pair is CNTN3 and the second member of the gene pair is PTPRG.

[0050] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is EPHA3.

[0051] In some embodiments, the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB2.

[0052] In some embodiments, the first member of the gene pair is EGF and the second member of the gene pair is TNFRSF11B.

[0053] In some embodiments, the first member of the gene pair is ENPEP and the second member of the gene pair is SLITRK1.

[0054] In some embodiments, the first member of the gene pair is FCGR3B and the second member of the gene pair is EDA2R.

[0055] In some embodiments, the first member of the gene pair is IL20RA and the second member of the gene pair is CLEC14A.

[0056] In some embodiments, the first member of the gene pair is IL6R and the second member of the gene pair is BTNL9.

[0057] In some embodiments, the first member of the gene pair is IZUMO1 and the second member of the gene pair is LILRA5.

[0058] In some embodiments, the first member of the gene pair is NGFR and the second member of the gene pair is LRRTM3.

[0059] In some embodiments, the first member of the gene pair is NTM and the second member of the gene pair is AMIGO2.

[0060] In some embodiments, the first member of the gene pair is PCDHB3 and the second member of the gene pair is IGSF11.

[0061] In some embodiments, the first member of the gene pair is PTGFRN and the second member of the gene pair is TMEM59L.

[0062] In some embodiments, the first member of the gene pair is TREM1 and the second member of the gene pair is VSIG8.

[0063] In some embodiments, the PD-L1 axis binding antagonist is selected from the group consisting of a PD-L1 binding antagonist, a PD-1 binding antagonist, and a PD-L2 binding antagonist.

[0064] In some embodiments, the PD-L1 axis binding antagonist is a PD-L1 binding antagonist.

[0065] In some embodiments, the PD-L1 binding antagonist inhibits the binding of PD-L1 to one or more of its ligand binding partners.

[0066] In some embodiments, the PD-L1 binding antagonist inhibits the binding of PD-L1 to PD-1.

[0067] In some embodiments, the PD-L1 binding antagonist inhibits the binding of PD-L1 to B7-1.

[0068] In some embodiments, the PD-L1 binding antagonist inhibits the binding of PD-L1 to both PD-1 and B7-1.

[0069] In some embodiments, the PD-L1 binding antagonist is an antibody or antigen-binding fragment thereof.

[0070] In some aspects, the antibody is selected from the group consisting of atezolizumab, MDX-1105, MEDI4736 (durvalumab), and MSB0010718C (avelumab).

[0071] In some embodiments, the anti-PD-L1 antibody comprises the following hypervariable regions: (a) the HVR-H1 sequence of GFTFSDSWIH (SEQ ID NO: 19); (b) the HVR-H2 sequence of AWISPYGGSTYYADSVKG (SEQ ID NO: 20); (c) the HVR-H3 sequence of RHWPGGFDY (SEQ ID NO: 21); (d) the HVR-L1 sequence of RASQDVSTAVA (SEQ ID NO: 22); (e) the HVR-L2 sequence of SASFLYS (SEQ ID NO: 23); and (f) the HVR-L3 sequence of QQYLYHPAT (SEQ ID NO: 24).

[0072] In some embodiments, the anti-PD-L1 antibody comprises: (a) a heavy chain variable (VH) domain comprising an amino acid sequence that has at least 90% sequence identity to the amino acid sequence of SEQ ID NO: 3; (b) a light chain variable (VL) domain comprising an amino acid sequence that has at least 90% sequence identity to the amino acid sequence of SEQ ID NO: 4; or (c) the VH domain described in (a) and the VL domain described in (b).

[0073] In some embodiments, the anti-PD-L1 antibody comprises: (a) a heavy chain variable (VH) domain comprising an amino acid sequence that has at least 95% sequence identity to the amino acid sequence of SEQ ID NO: 3; (b) a light chain variable (VL) domain comprising an amino acid sequence that has at least 95% sequence identity to the amino acid sequence of SEQ ID NO: 4; or (c) the VH domain described in (a) and the VL domain described in (b).

[0074] In some embodiments, the anti-PD-L1 antibody comprises (a) a heavy chain variable (VH) domain comprising an amino acid sequence that has at least 96% sequence identity to the amino acid sequence of SEQ ID NO: 3; (b) a light chain variable (VL) domain comprising an amino acid sequence that has at least 96% sequence identity to the amino acid sequence of SEQ ID NO: 4; or (c) the VH domain described in (a) and the VL domain described in (b).

[0075] In some embodiments, the anti-PD-L1 antibody comprises: (a) a heavy chain variable (VH) domain comprising an amino acid sequence that has at least 97% sequence identity to the amino acid sequence of SEQ ID NO: 3; (b) a light chain variable (VL) domain comprising an amino acid sequence that has at least 97% sequence identity to the amino acid sequence of SEQ ID NO: 4; or (c) the VH domain described in (a) and the VL domain described in (b).

[0076] In some embodiments, the anti-PD-L1 antibody comprises: (a) a heavy chain variable (VH) domain comprising an amino acid sequence that has at least 98% sequence identity to the amino acid sequence of SEQ ID NO: 3; (b) a light chain variable (VL) domain comprising an amino acid sequence that has at least 98% sequence identity to the amino acid sequence of SEQ ID NO: 4; or (c) the VH domain described in (a) and the VL domain described in (b).

[0077] In some embodiments, the anti-PD-L1 antibody comprises: (a) a heavy chain variable (VH) domain comprising an amino acid sequence that has at least 99% sequence identity to the amino acid sequence of SEQ ID NO: 3; (b) a light chain variable (VL) domain comprising an amino acid sequence that has at least 99% sequence identity to the amino acid sequence of SEQ ID NO: 4; or (c) the VH domain described in (a) and the VL domain described in (b).

[0078] In some embodiments, the anti-PD-L1 antibody comprises (a) a VH domain comprising the amino acid sequence of SEQ ID NO:3; and (b) a VL domain comprising the amino acid sequence of SEQ ID NO:4.

[0079] In some embodiments, the anti-PD-L1 antibody is atezolizumab (MPDL3280A).

[0080] In some embodiments, the PD-L1 axis binding antagonist is a PD-1 binding antagonist, hi some embodiments, the PD-1 binding antagonist inhibits the binding of PD-1 to one or more of its ligand binding partners.

[0081] In some embodiments, the PD-1 binding antagonist inhibits the binding of PD-1 to PD-L1.

[0082] In some embodiments, the PD-1 binding antagonist inhibits the binding of PD-1 to PD-L2.

[0083] In some embodiments, the PD-1 binding antagonist inhibits the binding of PD-1 to both PD-L1 and PD-L2.

[0084] In some embodiments, the PD-1 binding antagonist is an antibody or antigen-binding fragment thereof.

[0085] In some aspects, the antibody is selected from the group consisting of MDX-1106 (nivolumab), MK-3475 (pembrolizumab), MEDI-0680 (AMP-514), PDR001, REGN2810, and BGB-108.

[0086] In some embodiments, the PD-1 binding antagonist is an Fc fusion protein.

[0087] In another aspect, the disclosure features a method of treating an individual having cancer, including administering to the individual an effective amount of an agonist of CD177 activity.

[0088] In another aspect, the disclosure features a method for identifying an individual having cancer who may benefit from a treatment comprising an agonist of CD177 activity, the method comprising determining an expression level of podoplanin (PDPN) in a sample from the individual, wherein an expression level of PDPN in the sample that exceeds a reference PDPN expression level identifies the individual as an individual who may benefit from a treatment comprising an agonist of CD177 activity.

[0089] In another aspect, the disclosure features a method for selecting a therapy for an individual having cancer, comprising determining an expression level of PDPN in a sample from the individual, wherein an expression level of PDPN in the sample that exceeds a reference PDPN expression level identifies the individual as one who may benefit from a therapy comprising an agonist of CD177 activity.

[0090] In some embodiments, the individual has an expression level of PDPN in the sample that is greater than a reference PDPN expression level, and the method further comprises administering to the individual an effective amount of an agonist of CD177 activity.

[0091] In another aspect, the disclosure features a method of treating an individual having cancer, the method including: (a) determining an expression level of PDPN in a sample from the individual, wherein the expression level of PDPN in the sample is greater than a reference PDPN expression level; and (b) administering to the individual an effective amount of an agonist of CD177 activity.

[0092] In another aspect, the disclosure features a method of treating an individual having cancer, the method comprising administering to the individual an effective amount of an agonist of CD177 activity, wherein the expression level of PDPN in a sample from the individual has been determined to be greater than a reference PDPN expression level.

[0093] In some embodiments, the CD177 activity is inhibition of PDPN.

[0094] In some embodiments, the sample from the individual is a tumor tissue sample or a tumor fluid sample. In some embodiments, the tumor tissue sample is a formalin-fixed, paraffin-embedded (FFPE) sample, an archived sample, a fresh sample, or a frozen sample.

[0095] In some embodiments, the expression level of PDPN in the sample is the protein expression level of PDPN or the RNA expression level of PDPN. In some embodiments, the expression level of PDPN in the sample is the RNA expression level of PDPN. In some embodiments, the RNA expression level of PDPN is determined by in situ hybridization (ISH), RNA-seq, RT-qPCR, qPCR, multiplex qPCR or RT-qPCR, microarray analysis, SAGE, MassARRAY technology, FISH, or a combination thereof.

[0096] In some embodiments, the reference PDPN expression level is the expression level of PDPN in a population of individuals with cancer, hi some embodiments, the cancer is colorectal cancer (CRC), squamous cell carcinoma of the head and neck, or glioma.

[0097] In some embodiments, the reference PDPN expression level is the 50th percentile of expression levels in a population.

[0098] In some embodiments, the reference PDPN expression level is the 66th percentile of expression levels in a population.

[0099] In some embodiments, the reference PDPN expression level is a pre-assigned PDPN expression level.

[0100] In some embodiments, the cancer is CRC, squamous cell carcinoma of the head and neck, or glioma. In some embodiments, the cancer is CRC, e.g., stage II CRC or stage IV CRC.

[0101] In some embodiments, the benefit comprises an increase in recurrence-free survival (RFS) of an individual compared to treatment without an agonist of CD177 activity.

[0102] In some embodiments, an agonist of CD177 activity results in increased binding of PDPN to CD177 compared to binding of the two proteins in the absence of the agonist.

[0103] In some embodiments, the agonist of CD177 activity results in a change in a downstream activity of PDPN compared to the downstream activity in the absence of the agonist of CD177 activity.

[0104] In some aspects, the alteration of a downstream activity is a decrease in tumor growth.

[0105] In some embodiments, the alteration in downstream activity is a decrease in cancer-associated fibroblast (CAF) contractility.

[0106] In some embodiments, the agonist of CD177 activity is a small molecule, an antibody or antigen-binding fragment thereof, a peptide, or a mimetic.

[0107] In some embodiments, the agonist of CD177 activity is a peptide. In some embodiments, the peptide is a CD177 peptide. In some embodiments, the CD177 peptide is the extracellular domain of CD177. In some embodiments, the peptide is multimerized, e.g., tetramerized, e.g., tetramerized using streptavidin.

[0108] In some embodiments, the agonist of CD177 activity is an antibody or antigen-binding fragment thereof.

[0109] In some embodiments, the antibody or antigen-binding fragment thereof binds to PDPN. In some embodiments, the antibody or antigen-binding fragment thereof is an antagonist antibody or antigen-binding fragment thereof.

[0110] In some embodiments, the antibody or antigen-binding fragment thereof binds to CD 177. In some embodiments, the antibody or antigen-binding fragment thereof is an agonist antibody or antigen-binding fragment thereof.

[0111] In some embodiments, the antigen-binding fragment is a bis-Fab, Fv, Fab, Fab'-SH, F(ab')2, diabody, linear antibody, scFv, scFab, VH domain, or VHH domain.

[0112] In some embodiments, the individual is a human.

[0113] In another aspect, the disclosure features a collection of polypeptides, where each polypeptide comprises an extracellular domain, a tag, and an anchor, and the collection of polypeptides comprises at least 81% of the extracellular domains of the proteins in Table 7.

[0114] In some embodiments, the collection of polypeptides comprises the extracellular domains of at least 85% of the proteins in Table 7. In some embodiments, the collection of polypeptides comprises at least 90% of the proteins in Table 7. In some embodiments, the collection of polypeptides comprises the extracellular domains of at least 95% of the proteins in Table 7. In some embodiments, the collection of polypeptides comprises the extracellular domains of all of the proteins in Table 7.

[0115] In some embodiments, the anchor is capable of tethering the extracellular domain to the surface of the plasma membrane of a cell. In some embodiments, the anchor is a glycosylphosphatidylinositol (GPI) polypeptide.

[0116] In some embodiments, the tag can be visualized directly or indirectly. In some embodiments, the tag comprises a moiety that can be detected using an antibody or antibody fragment. In some embodiments, the tag is a glycoprotein D (gD) polypeptide. In some embodiments, the tag comprises a fluorescent protein.

[0117] In some aspects, the extracellular domain has a native conformation. In some aspects, the extracellular domain includes a native post-translational modification.

[0118] In some embodiments, the cell is a mammalian cell, hi some embodiments, the cell is a COS7 cell.

[0119] In some embodiments, the cells are transiently transfected with a plasmid encoding the polypeptide.

[0120] In another aspect, the disclosure features a collection of vectors encoding the collection of polypeptides of any one of the above aspects.

[0121] In another aspect, the disclosure features a collection of cells comprising the collection of vectors of the above aspects. In some aspects, a plurality of cells can express at least one polypeptide of any one of the above aspects, and optionally, different cells express different polypeptides.

[0122] In some embodiments, one or more of said polypeptides are each immobilized at a distinct location on one or more solid surfaces.

[0123] In another aspect, the disclosure features a method for identifying protein-protein interactions, the method including: providing a collection of polypeptides of any one of the above aspects, optionally immobilized on one or more solid surfaces; contacting the collection of step (a) with a multimerized query protein under conditions that allow binding of the multimerized query protein to at least one extracellular domain of the polypeptides; and detecting an interaction between the multimerized query protein and at least one extracellular domain, thereby identifying a protein-protein interaction. In some aspects, one or more of the polypeptides are each immobilized at a distinct location on the one or more solid surfaces. In some aspects, the distinct locations comprise cells that display the polypeptides.

[0124] In some embodiments, the cell is a mammalian cell.

[0125] In some embodiments, the contacting step is semi-automated.

[0126] In some embodiments, detecting the interaction comprises detecting a signal, optionally a fluorescent signal, above a threshold level at a location on the solid surface. In some embodiments, the detection is automated.

[0127] In some embodiments, the interaction is a transient interaction.

[0128] In some embodiments, the interaction is a low affinity interaction. In some embodiments, the low affinity interaction is a micromolar affinity interaction.

[0129] In some aspects, the multimerized query protein is a dimerized, trimerized, tetramerized, or pentamerized query protein. In some aspects, the multimerized query protein is a tetramerized query protein. In some aspects, the multimerized query protein comprises an isolated extracellular domain of a query protein. In some aspects, the isolated extracellular domain of a query protein is biotinylated and conjugated to fluorescent streptavidin to tetramerize the query protein.

[0130] In another aspect, the disclosure features a method of identifying a modulator of an interaction between a protein of Table 1 and a protein of Table 2, the method including: (a) providing a candidate modulator; (b) contacting the protein of Table 1 with a protein of Table 2 in the presence or absence of the candidate modulator under conditions that allow binding of the protein of Table 1 to the protein of Table 2, wherein the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (c) measuring binding of the protein of Table 1 to the protein of Table 2, wherein an increase or decrease in binding in the presence of the candidate modulator compared to binding in the absence of the candidate modulator identifies the candidate modulator as a modulator of the interaction between the protein of Table 1 and the protein of Table 2.

[0131] In another aspect, the disclosure features a method of identifying a modulator of a downstream activity of a protein of Table 1, the method including: (a) providing a candidate modulator; (b) contacting the protein of Table 1 with a protein of Table 2 in the presence or absence of the candidate modulator under conditions that allow binding of the protein of Table 1 to the protein of Table 2, wherein the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (c) measuring the downstream activity of the protein of Table 1, wherein a change in downstream activity in the presence of the candidate modulator compared to the downstream activity in the absence of the candidate modulator identifies the candidate modulator as a modulator of the downstream activity of the protein of Table 1.

[0132] In another aspect, the disclosure features a method of identifying a modulator of a downstream activity of a protein of Table 2, the method including: (a) providing a candidate modulator; (b) contacting the protein of Table 2 with a protein of Table 1 in the presence or absence of the candidate modulator under conditions that allow binding of the protein of Table 2 to the protein of Table 1, wherein the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (c) measuring the downstream activity of the protein of Table 2, wherein a change in downstream activity in the presence of the candidate modulator compared to the downstream activity in the absence of the candidate modulator identifies the candidate modulator as a modulator of the downstream activity of the protein of Table 2.

[0133] In some embodiments, the increase or decrease in binding is at least 70% as measured by surface plasmon resonance, biolayer interferometry, or enzyme-linked immunosorbent assay (ELISA).

[0134] In some embodiments, the modulator is an inhibitor of a downstream activity of a protein of Table 1 or Table 2. In some embodiments, the modulator is an activator of a downstream activity of a protein of Table 1 or Table 2.

[0135] In some aspects, the change in downstream activity is a decrease in the amount, intensity, or duration of the downstream activity. In some aspects, the change in downstream activity is an increase in the amount, intensity, or duration of the downstream activity.

[0136] In some embodiments, the modulator is a small molecule, an antibody or antigen-binding fragment thereof, a peptide, a mimetic, an antisense oligonucleotide, or a small interfering RNA (siRNA). In some embodiments, the antigen-binding fragment is a bis-Fab, Fv, Fab, Fab'-SH, F(ab')2, diabody, linear antibody, scFv, scFab, VH domain, or VHH domain. In some embodiments, the antibody or antigen-binding fragment thereof binds to a protein in Table 1. In some embodiments, the antibody or antigen-binding fragment thereof binds to a protein in Table 2.

[0137] In some embodiments, the protein in Table 1 is podoplanin (PDPN). In some embodiments, the protein in Table 2 is CD177.

[0138] In some embodiments, the downstream activity is cancer-associated fibroblast (CAF) contractility. In some embodiments, CAF contractility is decreased in the presence of the modulator. In some embodiments, CAF contractility is decreased by at least 20% as measured by a gel contraction assay. In some embodiments, CAF contractility is decreased by at least 20% as measured by a 3D gel elongation assay.

[0139] In some embodiments, the downstream activity is tumor growth. In some embodiments, tumor growth is reduced in the presence of the modulator. In some embodiments, tumor growth is reduced by at least 20% as measured in a tumor growth assay.

[0140] In some embodiments, the modulator is an antibody or antigen-binding fragment thereof that targets PDPN. In some embodiments, the modulator is an antibody or antigen-binding fragment thereof that targets CD177.

[0141] In some embodiments, the protein in Table 1 is PD-L1 (CD274). In some embodiments, the protein in Table 2 is EPHA3.

[0142] In some embodiments, the protein in Table 1 is PD-L2 (PDCD1LG2). In some embodiments, the protein in Table 2 is CEACAM4, ICAM5, NECTIN3, PSG9, or TNFRSF11A. In some embodiments, the protein in Table 2 is CEACAM4.

[0143] In some embodiments, the downstream activity is immune checkpoint inhibition. In some embodiments, immune checkpoint inhibition is reduced in the presence of the modulator. In some embodiments, immune checkpoint inhibition is reduced by at least 30% as measured in a cell-based assay.

[0144] In some embodiments, the protein in Table 1 is PTPRD. In some embodiments, PTPRD comprises G203E and K204E; R232C and R233C; P249L; G285E; E406K; S431L; R561Q; P666S; E755K; V892I; S912F; R995C; or R1088C amino acid substitution mutations or a ΔG61ΔE106 amino acid deletion mutation. In some embodiments, the protein in Table 2 is BMP5, CEACAM3, IL1RAP, IL1RAPL2, LECT1, LRFN5, SIRPG, SLITRK3, SLITRK4, SLITRK6, or TGFA.

[0145] In some embodiments, the downstream activity is inhibition of cell proliferation. In some embodiments, the inhibition of cell proliferation is increased in the presence of the modulator. In some embodiments, the inhibition of cell proliferation is increased by at least 30% as measured in a colony formation assay.

[0146] In some embodiments, the downstream activity is inhibition of STAT3 phosphorylation. In some embodiments, the inhibition of STAT3 phosphorylation is increased in the presence of the modulator. In some embodiments, the inhibition of STAT3 phosphorylation is increased by at least 30% as measured by Western blot of phosphorylated STAT3.

[0147] In some embodiments, the protein in Table 1 is a PTPRS. In some embodiments, the protein in Table 2 is C6orf25, IL1RAP, IL1RAPL1, IL1RAPL2, LRFN1, LRFN5, LRRC4B, NCAM1, SLITRK1, SLITRK2, SLITRK3, SLITRK4, or SLITRK6.

[0148] In some embodiments, the protein in Table 1 is PTPRF. In some embodiments, the protein in Table 2 is CD177, IL1RAP, or LRFN5.

[0149] In some embodiments, the downstream activity is inhibition of cell migration. In some embodiments, the inhibition of cell migration is increased in the presence of the modulator. In some embodiments, the inhibition of cell migration is increased by at least 20%.

[0150] In some embodiments, the downstream activity is phosphorylation of EGFR. In some embodiments, phosphorylation of EGFR is decreased in the presence of the modulator. In some embodiments, phosphorylation of EGFR is decreased by at least 30% as measured in an assay of EGFR phosphorylation.

[0151] In some embodiments, the protein in Table 1 is CHL1. In some embodiments, the protein in Table 2 is CNTN1, CNTN5, SIRPA, L1CAM, or TMEM132A.

[0152] In some embodiments, the downstream activity is inhibition of tumorigenesis. In some embodiments, inhibition of tumorigenesis is increased in the presence of the modulator. In some embodiments, inhibition of tumorigenesis is increased by at least 20%.

[0153] In some embodiments, the protein in Table 1 is CNTN1. In some embodiments, the protein in Table 2 is CDH6, CHL1, FCGRT, PCDHB7, or SGCG.

[0154] In some embodiments, the downstream activity is cell proliferation or cell invasion. In some embodiments, cell proliferation or cell invasion is decreased in the presence of the modulator. In some embodiments, cell proliferation is decreased by at least 20% as measured in a colony formation assay. In some embodiments, cell invasion is decreased by at least 20% as measured in a gel invasion assay.

[0155] In some embodiments, the protein in Table 1 is LILRB1. In some embodiments, the protein in Table 2 is CLEC6A, CXADR, EDAR, FLT4, IL6R, ILDR1, or LILRA5.

[0156] In some embodiments, the downstream activity is inhibition of phagocytosis. In some embodiments, the inhibition of phagocytosis is decreased in the presence of the modulator. In some embodiments, the inhibition of phagocytosis is increased by at least 20%.

[0157] In some embodiments, the protein in Table 1 is LILRB2. In some embodiments, the protein in Table 2 is IGSF8 or MOG.

[0158] In some embodiments, the protein in Table 1 is LILRB3. In some embodiments, the protein in Table 2 is LRRC15 or LY6G6F.

[0159] In some embodiments, the protein in Table 1 is LILRB4. In some embodiments, the protein in Table 2 is CNTFR.

[0160] In some embodiments, the protein in Table 1 is LILRB5. In some embodiments, the protein in Table 2 is APLP2, CD177, CLEC10A, CLECSF13, LDLR, PILRA, or UNC5C. In some embodiments, the protein in Table 2 is LDLR.

[0161] In some embodiments, the downstream activity is osteoclast differentiation. In some embodiments, osteoclast differentiation is reduced by at least 20% in the presence of the modulator. In some embodiments, osteoclast differentiation is measured in an assay of TRAP+ multinucleated cells.

[0162] In some embodiments, the protein in Table 1 is AXL. In some embodiments, the protein in Table 2 is IL1RL1 or VSIG10L.

[0163] In some embodiments, the downstream activity is activation of the RAS / RAF / MAPK / ERK1 / 2 pathway, activation of the JAK / STAT pathway, activation of the PI3K signaling pathway, cell migration, filopodia formation, or phosphorylation of AXL. In some embodiments, cell migration is reduced by at least 20% as measured by a gel invasion assay.

[0164] In some embodiments, the protein in Table 1 is LRRC4B. In some embodiments, the protein in Table 2 is BTN3A1 or BTN3A3.

[0165] In some embodiments, expression of a protein of Table 1 or a protein of Table 2 is upregulated or downregulated in tumor tissue compared to healthy tissue.

[0166] In another aspect, the disclosure features an isolated modulator of an interaction between a protein of Table 1 and a protein of Table 2, where (a) the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (b) the modulator causes an increase or decrease in binding of the protein of Table 1 to the protein of Table 2 compared to binding in the absence of the modulator.

[0167] In another aspect, the disclosure features an isolated modulator of a downstream activity of a protein of Table 1 or a protein of Table 2, where (a) the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (b) the modulator causes a change in the downstream activity of the protein of Table 1 or the protein of Table 2 compared to the downstream activity in the absence of the modulator.

[0168] In some embodiments, the increase or decrease in binding is at least 70% as measured by surface plasmon resonance, biolayer interferometry, or enzyme-linked immunosorbent assay (ELISA).

[0169] In some embodiments, the modulator is an inhibitor of a downstream activity of a protein of Table 1 or Table 2. In some embodiments, the modulator is an activator of a downstream activity of a protein of Table 1 or Table 2.

[0170] In some aspects, the change in downstream activity is a decrease in the amount, intensity, or duration of the downstream activity. In some aspects, the change in downstream activity is an increase in the amount, intensity, or duration of the downstream activity.

[0171] In some embodiments, the modulator is a small molecule, an antibody or antigen-binding fragment thereof, a peptide, or a mimetic. In some embodiments, the antigen-binding fragment is a bis-Fab, Fv, Fab, Fab'-SH, F(ab')2, diabody, linear antibody, scFv, scFab, VH domain, or VHH domain. In some embodiments, the antibody or antigen-binding fragment thereof binds to a protein in Table 1. In some embodiments, the antibody or antigen-binding fragment thereof binds to a protein in Table 2. [Brief explanation of the drawings]

[0172] [Figure 1A]Figure 1A is a pair of graphs showing the occurrence of major protein domains and motifs according to UniProt annotation in the set of immunoglobulin superfamily (IgSF) members and other non-IgSF proteins tested for pairwise interactions in the study. EGF = epidermal growth factor, TNF = tumor necrosis factor. [Figure 1B] Figure 1B is a schematic diagram illustrating an automated high-throughput technique for extracellular interactome discovery (extracellular interaction screening). (I) shows a library of 1,364 human STM receptors (prey proteins) expressed as receptor extracellular domain (ECD)-Fc fusion proteins for expression in the conditioned medium of transfected cells. (II) shows a library of IgSF priority members (query proteins) tested for binding to the collection of STM receptors. IgSF proteins were expressed as pentameric constructs fused to beta-lactamase enzyme for increased avidity and sensitive detection of binding partners. (III) illustrates the automated extracellular interaction screening. Each IgSF query protein was screened for binding to all STM receptors and processed in a computational pipeline to minimize false positives, incorporate gene expression data, and enable further analysis. (IV) shows the IgSF interactome, containing over 800 high-confidence protein-protein interactions. (V) illustrates the method for validation of selected interactions. Validation methods include surface plasmon resonance and immunofluorescence for detection of cell surface binding partners. [Figure 1C] Figure 1C shows scatter plots of two independent replicates of the extracellular interaction screen for PD-1 (PDCD1) and PD-L2 (PDCD1LG2), respectively. Known binding partners from the literature are shown in blue; these are PDCD1LG2 and CD274 for PD-1, and PDCD1 for PD-L2. [Figure 2A]Figure 2A shows a network representation of all predicted receptor-interacting pairs identified by the IgSF interactome (the "IgSF interactome"). Nodes represent IgSF query and STM prey proteins, and edges represent interactions between them. Node size corresponds to the number of network neighbors (node ​​degree), indicating network hubs. Edges representing known interactions (e.g., interactions previously predicted in the Bioplex, Biogrid, or STRING databases) are shown in red. Reconstructed subnetworks that map to well-studied immunoglobulin (Ig) families are marked as follows: (1) the semaphorin-plexin subnetwork; (2) the ephrin receptor tyrosine kinase subnetwork; (3) the PD-1 / PD-L1 immunoregulatory axis; and (4) the PVR / TIGIT subnetwork. [Figure 2B] Figure 2B is a ridge plot showing the separation of the specificity score distributions for the non-specific, positive, and negative classes in the training set. [Figure 2C] Figure 2C is a regression plot showing that the topology coefficients within the IgSF interactome network follow a power law, characteristic of scale-free networks. [Figure 2D] Figure 2D is a Venn diagram showing the overlap between interactions identified in the extracellular interaction screen and known interactions in the Bioplex, Biogrid, and STRING databases. [Figure 2E] Figure 2E is a bar graph showing that the percentage of identified interactions between two extracellular proteins (Bioplex; 1%) according to the cellular localization associations in the Human Protein Atlas is substantially underrepresented compared to the estimated percentage of extracellular proteins in the human proteome (18%). [Figure 2F]Figure 2F is a bar graph sorting the number of reported unique interaction pairs (577) by directional subset: 114 pairs (red) where a prey in the STM library was included in the query set and the reported interaction was mutually confirmed; 124 pairs (orange) where a prey in the STM library was included in the query set and the reported interaction was not mutually confirmed; and 463 pairs where a prey in the STM library was not included in the query set. [Figure 3A] Figure 3A shows a network representation of the IgSF interactome analyzed by Markov clustering (MCL) clustering based on network connectivity and healthy tissue gene expression profiles from genotype-tissue expression (GTEx). Edges connecting nodes within a single cluster are annotated in black; edges connecting nodes in different clusters are colored light gray. All clusters are annotated with the most common statistically significantly enriched biological process Gene Ontology (GO) terms, which correspond to the numbered legends below. Network nodes with prior annotations of enriched GO terms are displayed in different colors corresponding to those terms. Nodes with unknown annotations or different annotations to the terms assigned to the cluster are shown as diamonds. [Figure 3B] Figure 3B is a set of violin plots showing that the average correlation (Pearson's r) of mRNA expression measured across all GTEx tissues is significantly higher (p<1.2x10-20) for all pairs of reported interacting proteins (left plot, yellow) compared to the complement of all possible non-interacting pairs of network nodes (right plot, blue). The dotted line indicates the 95th percentile of the correlation distribution, above which selected novel interactions validated in this study are shown. [Figure 3C] Figure 3C is a scatter plot comparing normalized mRNA expression (log2 nRPKM) of the reported binding partners Nectin1 and Nectin4 in esophageal GTEx tissue subsets. Expression patterns were significantly correlated (q < 0.05) with the overlaid regression model (red). [Figure 3D] Figure 3D is a scatter plot comparing normalized mRNA expression (log2 nRPKM) of the reported binding partners CEACAM5 and CEACAM7 in colonic GTEx tissue subsets. Expression patterns were significantly correlated (q<0.05) with the overlaid regression model (red). [Figure 3E] Figure 3E is a scatter plot comparing normalized mRNA expression (log2 nRPKM) of reported binding partners LILRA5 and LILRB1 in blood GTEx tissue subsets. Expression patterns were significantly correlated (q<0.05) with the overlaid regression model (red). [Figure 3F] FIG. 3F is a scatter plot showing a comparison of normalized mRNA expression (log2 nRPKM) of reported binding partners PTPRZ1 and CNTN1 in brain GTEx tissue subsets. [Figure 3G] Figure 3G is a set of scatter plots showing a comparison of normalized mRNA expression (log2 nRPKM) of reported binding partners L1CAM and CHL1 in specific GTEx tissue subsets (left to right: colon, small intestine, nerve, and stomach); expression patterns were significantly correlated (q < 0.05) with the overlaid regression model (red). [Figure 3H] Figure 3H is a schematic diagram showing the design of the cell surface interaction assay. The binding partner (BP) was expressed on cells, and the binding of the target protein expressed as a recombinant purified ECD to the cell surface of untransfected (UT) and receptor-expressing cells was analyzed using immunofluorescence. The recombinant protein was biotinylated and multimerized with fluorescent streptavidin to increase binding avidity and detect transient interactions. [Figure 3I] Figure 3I is a set of micrographs showing the results of a cell surface interaction assay of soluble query NCR1 and binding partners SIGLEC6, SIGLEC7, and SIGLEC8 compared to control UT cells. "Bound" shows only the query protein in gray. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 3J] Figure 3J is a set of micrographs showing the results of a cell surface interaction assay of soluble query CHL1 and its binding partner L1CAM compared to control UT cells. "Bound" shows only the query protein in gray. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 3K] Figure 3K is a set of micrographs showing the results of a cell surface interaction assay of soluble query CNTN1 and its binding partner PTPRZ1 compared to control UT cells. "Bound" shows only the query protein in gray. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 3L] FIG. 3L is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of NCR1-FLAG in cells co-expressing SIGLEC7-HA or vector control. [Figure 3M] FIG. 3M is a set of immunoblots showing the results of a co-IP assay of NCR1-FLAG in cells co-expressing SIGLEC8-HA or vector control. [Figure 3N] FIG. 3N is a set of immunoblots showing the results of a co-IP assay of NCR1-FLAG in cells co-expressing CD4-HA or vector control. [Figure 4A] Figure 4A is an IgSF interactome network representation showing the PD-L1 (CD274) and PD-L2 (PDCD1LG2) immune regulatory clusters (light red shading). Edges corresponding to CD274-EPHA3 and CEACAM4-PDCD1LG2 interactions are shown in bold and highlighted in red. Nodes are color-coded according to their primarily assigned GO categories, as shown in Figure 3. [Figure 4B]Figure 4B is a sensorgram showing the binding of PD-1, PD-L1, and PD-L2 (expressed as ECD-Fc fusion proteins) to CEACAM4 as analyzed by SPR. Recombinant CEACAM4 (expressed as ECD-Fc fusion proteins) was immobilized on a sensor chip, and the indicated proteins were injected at 250 nM concentrations. An irrelevant Fc-tagged protein was used as a control. The sensorgram shown is representative of at least three independent runs. [Figure 4C] Figure 4C shows sensorgrams depicting the binding of PD-L1, EPHA3, and EPHA5 (expressed as ECD-Fc fusion proteins) to LILRA3 as analyzed by SPR. Recombinant LILRA3 (expressed as ECD-Fc fusion proteins) was immobilized on a sensor chip, and the indicated proteins were injected at 250 nM concentrations. An irrelevant Fc-tagged protein was used as a control. The sensorgrams shown are representative of at least three independent runs. [Figure 4D] Figure 4D is a graph showing the binding of PD-L1 to its partners PD-1 and EPHA3 in the presence of increasing concentrations of the anti-PD-1 / PD-L1 antibody atezolizumab. Response units were measured at the end of the infusion. Bars show the mean ± standard deviation. Experiments are representative of two independent assays. [Figure 4E] Figure 4E is a sensorgram showing the binding of PD-1, PD-L1, and PD-L2 (expressed as ECD-Fc fusion proteins) to EPHA3, as analyzed by SPR. Recombinant EPHA3 (expressed as ECD-Fc fusion proteins) was immobilized on a sensor chip, and the indicated proteins were injected at 250 nM concentrations. An irrelevant Fc-tagged protein was used as a control. The sensorgram shown is representative of at least three independent runs. [Figure 4F]Figure 4F is a set of micrographs showing the results of cell surface interaction assays for soluble query PD-L1 and PD-L2 and the binding partners PD-1, CD80, EPHA3, EPHB1, PD-L2, PD-L1, CEACAM4, and CEACAM5. Query proteins are shown in red, and DAPI-stained nuclei are shown in blue. Scale bar = 50 μm. [Figure 4G] FIG. 4G is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of LILRB1-FLAG in cells co-expressing EDAR-HA or a vector control. [Figure 4H] Figure 4H is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of LILRB1-FLAG in cells co-expressing IL6R-HA or vector control. [Figure 4I] FIG. 4I is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of LILRB4-FLAG in cells co-expressing CNTFR-HA or vector control. [Figure 5A] Figure 5A shows a simplified network representation of interactions in the nervous system-associated community centered on individual binding partners of the PTPR family (Figure 3A, cluster 1) within the SLITRK, NTRK, LFRN, IL1RAP, and LRRC families. Red edges represent interactions validated using cell surface interaction assays, as shown in Figures 5B-5E. Dotted edges represent interactions with weak binding. [Figure 5B] Figure 5B is a set of micrographs showing the results of a cell surface interaction assay for soluble query SLITRK2 (top) and SLITRK3 (bottom) with binding partners PTPRD and PTPRS compared to control untransfected (UT) cells. "Bound" shows only the query protein in gray. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 5C]Figure 5C is a set of micrographs showing the results of a cell surface interaction assay of soluble query LFRN5 with its binding partners PTPRD, PTPRS, and PTPRF compared to control UT cells. "Bound" shows only the query protein in gray. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 5D] Figure 5D is a set of micrographs showing the results of a cell surface interaction assay of soluble query IL1RAP with binding partners PTPRD, PTPRS, and PTPRF compared to control UT cells. "Bound" shows only the query protein in gray. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 5E] Figure 5E is a set of micrographs showing the results of a cell surface interaction assay of soluble queries BTN3A1, BTN3A3, and BTN2A2 with their binding partner LRRC4B compared to control UT cells. "Bound" shows only the query protein in gray. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 5F] Figure 5F is a set of sensorgrams showing the binding of ILRAP, ILRAPL1, SLITRK1, SLITRK4, LRFN1, and LRFN4 to PTPRD as analyzed by SPR. The indicated proteins (expressed as ECD-Fc fusion proteins) were immobilized on a sensor chip, and PTPRD (expressed as the ectodomain fused to an Fc tag) was injected at the indicated concentrations. [Figure 5G] FIG. 5G is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of FAM-FLAG in cells co-expressing BTN2A1-HA or vector control. [Figure 5H] FIG. 5H is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of LRRC4B-FLAG in cells co-expressing BTN3A2-HA or vector control. [Figure 5I]FIG. 5I is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of LRRC4B-FLAG in cells co-expressing BTN3A3-HA or vector control. [Figure 5J] FIG. 5J is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of LRRC4B-FLAG in cells co-expressing EDAR-HA or vector control. [Figure 5K] Figure 5K is a schematic diagram showing cancer-associated amino acid substitution mutations in PTPRD. [Figure 5L] Figure 5L is a heatmap showing row-clustered log2 ratios of normalized absorbance per binding partner, color-coded from white (loss of binding) to red (preserved binding). [Figure 5M] Figure 5M is a network representation showing the interactions between LILR family members and their previously uncharacterized binding partners, demonstrating coordinated upregulation of LILR proteins in kidney renal clear cell carcinoma (KIRC). [Figure 5N] Figure 5N is a set of micrographs showing the results of cell surface interaction assays of soluble query proteins EDAR, LDLR, LILRA5, and CNTFR with their binding partners LILRB1, LILRB2, LILRB3, LILRB4, and LILRB5 compared to control UT cells. Query proteins are shown in red, and DAPI-stained nuclei are shown in blue. [Figure 6A] Figure 6A shows a network diagram depicting the IgSF interactome, represented as a network community, augmented with the results of differential expression analysis between tumor and adjacent normal tissue for each TCGA indication. Node color and size indicate the number of TCGA indications in which a gene was found to be significantly dysregulated (|log2 fold change|>1 and p<0.05). [Figure 6B]Figure 6B shows a bar graph showing the number of edges connecting significantly deregulated nodes in descending order for each TCGA indication. LUSC = lung squamous cell carcinoma, KIRC = kidney clear cell carcinoma, COAD = colon adenocarcinoma, KICH = kidney chromophobe, LUAD = lung adenocarcinoma, UCEC = uterine endometrial carcinoma, KIRP = kidney papillary cell carcinoma, BLCA = bladder urothelial carcinoma, LIHC = liver hepatocellular carcinoma, BRCA = breast invasive carcinoma, THCA = thyroid carcinoma, ESCA = esophageal carcinoma, STAD = gastric adenocarcinoma, HNSC = head and neck squamous cell carcinoma, and PRAD = prostate adenocarcinoma. Red bars represent the IgSF interactome. Light red bars represent the reference set of unrelated gene pairs reported in TCGA. [Figure 6C] FIG. 6C is a network diagram showing up- or down-regulated genes in the immune regulatory community (defined in FIG. 3B) in kidney renal clear cell carcinoma (KIRC). [Figure 6D] FIG. 6D is a network diagram showing up- or down-regulated genes in the immune regulatory community (defined in FIG. 3B) in head and neck squamous cell carcinoma (HNSC). [Figure 6E] Figure 6E is a set of violin plots showing that the average correlation (Pearson's r) of mRNA expression measured across all Cancer Cell Line Encyclopedia (CCLE) cell lines is significantly higher (p<4x10-2) for all pairs of reported interacting proteins (left plot, yellow) compared to the complement of all possible non-interacting pairs of network nodes (right plot, blue). The dotted line indicates the 95th percentile of the correlation distribution, above which are shown selected novel interactions validated in this study. [Figure 6F] Figure 6F is a scatter plot comparing the relative protein expression of the reported binding partners CEACAM5 and CEACAM6 in colonic CCLE tissue subsets. Expression patterns were significantly correlated (q < 0.2) with the overlaid regression model (red). [Figure 6G]Figure 6G is a scatter plot comparing the relative protein expression of reported binding partners CNTN1 and NRCAM in lung CCLE tissue subsets. Expression patterns were significantly correlated (q < 0.2) with the overlaid regression model (red). [Figure 6H] Figure 6H is a scatter plot comparing the relative protein expression of reported binding partners BTN3A1 and LRRC4B in colonic CCLE tissue subsets. Expression patterns were significantly correlated (q < 0.2) with the overlaid regression model (red). [Figure 6I] Figure 6I is a scatter plot comparing the relative protein expression of reported binding partners FLRT3 and UNC5C in breast CCLE tissue subsets. Expression patterns were significantly correlated (q<0.2) with the overlaid regression model (red). [Figure 6J] Figure 6J is a scatter plot comparing the relative protein expression of reported binding partners CNTN1 and PTPRZ1 in CCLE tissue subsets of hematopoietic and lymphoid tissues. Expression patterns were significantly correlated (q < 0.2) with the overlaid regression model (red). [Figure 6K] Figure 6K is a scatter plot comparing the relative protein expression of reported binding partners IGSF3 and PTGFRN in colonic CCLE tissue subsets. Expression patterns were significantly correlated (q<0.2) with the overlaid regression model (red). [Figure 6L] Figure 6L is a scatter plot comparing the relative protein expression of reported binding partners IGSF3 and PTGFRN in breast CCLE tissue subsets. Expression patterns were significantly correlated (q<0.2) with the overlaid regression model (red). [Figure 6M] Figure 6M is a scatter plot comparing the relative protein expression of reported binding partners AXL and VSIG10L in CCLE tissue subsets of hematopoietic and lymphoid tissues. Expression patterns were significantly correlated (q<0.2) with the overlaid regression model (red). [Figure 6N]FIG. 6N is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of IL1RL1-FLAG and VSIG10L-FLAG in cells co-expressing AXL-HA or vector control. [Figure 6O] FIG. 6O is a set of sensorgrams showing the binding of AXL and TYRO3 to IL1RL1 or GAS6 analyzed using biolayer interferometry. [Figure 6P] FIG. 6P is a set of sensorgrams showing the binding of AXL and MER to IL1RL1 or GAS6 analyzed using biolayer interferometry. [Figure 6Q] Figure 6Q is a set of micrographs showing the results of a cell surface interaction assay of soluble query AXL and its binding partner IL1Rl1 compared to control UT cells. "Bound" shows only the query protein in gray. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 7A] Figure 7A is a graph showing the distribution of associated protein domains and motifs according to UniProt annotation for a set of 1129 unique single-pass transmembrane proteins ("prey constructs") in the receptor library: ITIM / ITAM = immunoreceptor tyrosine-based inhibitory motif / immunoreceptor tyrosine-based activation motif, TNFR = tumor necrosis factor receptor, TLR / ILR = Toll-like receptor / interleukin receptor, Ig-like = immunoglobulin-like, EGF = epidermal growth factor. [Figure 7B] Figure 7B is a Venn diagram showing the overlap between unique protein entries present in the Biogrid, Bioplex, and STRING databases (three comprehensive protein-protein interaction resources) and this study. [Figure 7C] Figure 7C is a schematic diagram of a set of 384-well plates showing two independent replicates of an automated cell surface interaction assay for PD-1 (PDCD1) tested against a single-pass transmembrane protein (STM) receptor library. [Figure 7D]Figure 7D is a schematic diagram of a set of 384-well plates showing two independent replicates of an automated cell surface interaction assay for PD-L2 (PDCD1LG2) tested against the STM receptor library. [Figure 7E] Figure 7E is a boxplot showing, from top to bottom, the distribution of raw enzyme absorbance background estimates per plate; the maximum enzyme absorbance control per plate; the unscaled enzyme absorbance values ​​across wells; the enzyme absorbance values ​​corrected by the background estimate per plate; and the subsequent "normalized" absorbance values ​​relative to the estimated maximum absorbance per plate. [Figure 8A] Figure 8A is a heat map of clustered normalized absorbance values ​​of 1,364 prey proteins (STM ECD) with 445 screened IgSF query proteins. [Figure 8B] Figure 8B is a set of ridge plots showing the discriminability of predictive features between nonspecific, true-positive, and true-negative interactions in the training set. From top to bottom: normalized absorbance; query Z-score of one query screened against the entire STM library; prey Z-score of a single prey in the STM library across all query proteins screened; and custom specificity score. [Figure 8C] Figure 8C is a plot showing principal component analysis (PCA) of training set negatives, training set nonspecifics, training set positives, predicted nonspecific / negatives, and predicted positive interactions. Mapping the 4-dimensional feature matrix to the first and second principal components shows that the predicted true positive interactions (dark blue) are in good agreement with the true positive interactions from our training set (light blue) and are well separated from the known nonspecific interactions (orange) and sampled true negative interactions (red). [Figure 8D]Figure 8D is a dot plot showing normalized intensities color-coded by predicted class (red: negative; green: positive / specific; blue: nonspecific) for all observed interactions between CD274, PDCD1, and PDCD1LG2, demonstrating excellent reproducibility and specificity across multiple query clones. [Figure 8E] Figure 8E is a dot plot showing normalized intensities color-coded by predicted class (red: negative; green: positive / specific; blue: nonspecific) for all observed interactions of CD274, PDCD1, and PDCD1LG2 with each other, demonstrating excellent reproducibility and specificity across multiple query clones. [Figure 8F] Figure 8F is a bar graph showing the total number of extracellular interactions in the Bioplex ("Bioplex-interacting extracellular proteins (Human Protein Atlas)"; 627), the number of Bioplex interactions between two proteins screened for interactions ("Bioplex-interacting proteins in this study"; 350), and the total number of extracellular interactions reported in this study ("All interactions in this study"; 577). [Figure 8G] Figure 8G is a histogram showing that the distribution of shortest paths in the IgSF interactome network is centered around 6. [Figure 9A] Figure 9A is a clustered heatmap of row-scale (Z-transformed), log2 rpkm counts of all IgSF interactome genes across GTEx tissue detail categories. [Figure 9B] Figure 9B is a violin plot showing that the mean distribution of tissue expression correlations (Pearson's r for GTEx) for all pairs of reported interacting proteins is significantly higher (p<1.2e-20) compared to the distribution of all possible non-interacting pairs (screening library complement). In the clustered network, the intra-cluster correlations are also significantly higher (p<0.05) compared to the distribution of all inter-cluster interactions. [Figure 9C]Figure 9C is a scatter plot of binding versus CEACAM5 and CEACAM7, and mRNA expression across all GTEx tissues was highly correlated. [Figure 9D] Figure 9D is a scatter plot of binding pairs of LILRB1 and LILRA5, and mRNA expression across all GTEx tissues was highly correlated. [Figure 9E] Figure 9E is a scatter plot of the binding pairs SIGLEC7 and NCR1, and mRNA expression across all GTEx tissues was highly correlated. [Figure 9F] FIG. 9F is a set of faceted scatter plots showing L1CAM and CHL1 mRNA expression across different brain regions, demonstrating constitutively high expression of CHL1 in the cerebellar region and strongly correlated expression patterns in all other regions. [Figure 9G] Figure 9G is a set of micrographs showing the results of a cell surface interaction assay of soluble query CHL1 and cell surface-expressed binding partners BTLA and CNTN5 compared to control, untransfected (UT) cells. In the merged images, the query protein bound to the cell surface is shown in red, and DAPI-stained nuclei are shown in blue. Scale bar = 50 μm. [Figure 9H] FIG. 9H is a network diagram showing new interactions identified using independent assays for the CNTN1 and CHL1 clusters. [Figure 9I] Figure 9I is a set of graphs showing the analysis of binding of NRCAM, NFASC, MCAM, CHL1, and control proteins to CNTN1 using SPR. Recombinant NRCAM, NFASC, MCAM, CHL1, and control proteins were immobilized on a sensor chip, and CNTN1 (expressed as recombinant ECD-Fc) was injected at a concentration of 250 nM. [Figure 9J] Figure 9J is a clustered heatmap of row-scaled expression values ​​for a representative network cluster (cluster 1). Network clusters often contain distinct tissue expression subgroups, exemplified by brain and whole blood, spleen, lung, and small intestine subgroups. [Figure 9K] Figure 9K is a clustered heatmap showing the representation of simplified GO categories (rows) for all network clusters (columns) with more than two members. Cell values ​​are color-coded according to the odds ratio for each category (upper limit: odds ratio > 50). Network clusters repeatedly contain genes with multiple biological annotations. [Figure 10A] Figure 10A shows a clustered tissue expression heatmap for members of the PD-L1 / CD274 and PDCD1LG2 / PD-L2 immunoregulatory clusters (green) along with members of the ephrin (purple) and CEACAM (olive) families, highlighting evidence of co-expression with CEACAM4 and PD-L2, and differences in co-expression between EPHA3 and the ephrin cluster. [Figure 10B] Figure 10B is a set of box plots showing the expression of PD-L1 (CD274) and EPHA3 in normal tissues based on GTEx data. [Figure 10C] Figure 10C is a set of box plots showing the expression of CEACAM4 and PD-L2 in normal tissues based on GTEx data. [Figure 10D] Figure 10D shows a sensorgram representing a representative SPR experiment demonstrating the binding of EPHA3 to PD-L1. PD-L1 was immobilized on a sensor chip, and EPHA3 expressed as a recombinant His-tagged ECD was injected at concentrations of 0, 50, 10, 20, 50, and 100 nM. The binding rate was calculated at equilibrium. [Figure 10E] Figure 10E shows a representative SPR sensorgram demonstrating the binding of CEACAM4 to PD-L2. PD-L2 was immobilized on a sensor chip, and CEACAM4 expressed as a recombinant His-tagged ECD was injected at concentrations of 0, 10, 20, 50, 100, and 200 nM. The binding rate was calculated at equilibrium. [Figure 10F]Figure 10F is a set of micrographs showing the results of a cell surface interaction assay of soluble query MDGA1 and cell surface-expressed binding partners NLGN3 and NLGN4X compared to control, untransfected (UT) cells. In the merged images, the query protein bound to the cell surface is shown in red, and DAPI-stained nuclei are shown in blue. Scale bar = 50 μm. [Figure 10G] Figure 10G is a set of micrographs showing the results of a cell surface interaction assay of soluble query TREML2 and cell surface-expressed binding partner ANTRX1 compared to control, untransfected (UT) cells. In the merged image, the query protein bound to the cell surface is shown in red, and DAPI-stained nuclei are shown in blue. Scale bar = 50 μm. [Figure 10H] FIG. 10H is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of IGSF5-FLAG in cells co-expressing CD300A-HA or vector control. [Figure 10I] FIG. 10I is a set of immunoblots showing the results of a co-immunoprecipitation (co-IP) assay of IGSF5-FLAG in cells co-expressing CD300LF-HA or vector control. [Figure 10J] Figure 10J is a set of micrographs showing the results of a cell surface interaction assay of soluble query proteins FLRT1, FLRT2, and FLRT3 and cell surface-expressed binding partners UNC5A, UNC5C, and UNC5D compared to control, untransfected (UT) cells. In the merged images, cell surface-bound query proteins are shown in red, and DAPI-stained nuclei are shown in blue. Scale bar = 50 μm. [Figure 11A] Figure 11A is a clustered heatmap of network nodes by TCGA indication. Cell values ​​represent the log2 mean change between gene expression levels in tumors and adjacent normals. [Figure 11B]Figure 11B is a set of micrographs showing the results of a cell surface interaction assay of soluble query CHL1 and cell surface-expressed binding partners L1CAM, BTLA, and CNTN5 compared to control untransfected (UT) cells. In the merged images, the query protein bound to the cell surface is shown in red, and DAPI-stained nuclei are shown in blue. Scale bar = 50 μm. [Figure 11C] Figure 11C is a subnetwork diagram of the IgSF interactome showing putative interactions identified for the CNTN1 and CHL1 IgSF proteins. Edges colored yellow indicate interactions that have been validated by independent techniques. [Figure 12A] Figure 12A is a set of micrographs showing the results of a cell surface interaction assay of soluble query LFRN5 and binding partners PTPRZ1, PTPRG, PTPRT, PTPRS, PTPRO, PTPRM, PTPRF, and PTPRD compared to control untransfected (UT) cells. "Bound" shows only the query protein in gray. The experiment shown is representative of two independent assays. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 12B] Figure 12B is a set of micrographs showing the results of a cell surface interaction assay of soluble query SLITKR2 and binding partners PTPRZ1, PTPRG, PTPRT, PTPRS, PTPRO, PTPRM, PTPRF, and PTPRD compared to control untransfected (UT) cells. "Bound" shows only the query protein in gray. The experiment shown is representative of two independent assays. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 12C]Figure 12C is a set of micrographs showing the results of a cell surface interaction assay of soluble query IL1RAP and binding partners PTPRZ1, PTPRG, PTPRT, PTPRS, PTPRO, PTPRM, PTPRF, and PTPRD compared to control untransfected (UT) cells. "Bound" shows only the query protein in gray. The experiment shown is representative of two independent assays. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 12D] Figure 12D is a set of micrographs showing the results of a cell surface interaction assay of soluble query CNTN1 and binding partners PTPRZ1, PTPRG, PTPRT, PTPRS, PTPRO, PTPRM, PTPRF, and PTPRD compared to control untransfected (UT) cells. "Bound" shows only the query protein in gray. The experiment shown is representative of two independent assays. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 12E] Figure 12E is a set of micrographs showing the results of a cell surface interaction assay of soluble query BTN2A2 and binding partner LRRC4B compared to control untransfected (UT) cells. "Bound" shows only the query protein in gray. The experiment shown is representative of two independent assays. "Merge" shows the query protein in red and DAPI-stained nuclei in blue. Scale bar = 50 μm. [Figure 12F] Figure 12F is a set of immunoblots showing the results of co-immunoprecipitation (co-IP) assays of LRRC4B-FLAG, LRRC4C-FLAG, TGOLN2-FLAG, VSIG8-FLAG, CDH9-FLAG, and ST14-FLAG in cells co-expressing BTN3A1-HA or vector control. [Figure 12G] FIG. 12G is a bar graph showing the binding specificity of the indicated PTPRD variants to the indicated binding partners compared to wild-type PTPRD. [Figure 13A]Figure 13A is a set of Kaplan-Meier curves and a table showing the survival probability of patients with stage II disease divided into tertiles by level of podoplanin (PDPN) expression (T1, T2, and T3) in the publicly available CRC (stage II) microarray gene expression dataset GSE33113. Logrank p-values ​​(Logrank p) are associated with the Kaplan-Meier curves. Cox proportional hazards (CoxPH) p-values ​​are associated with the univariate model detailed in Table 10. [Figure 13B] Figure 13B is a set of Kaplan-Meier curves and a table showing the survival probability of patients at any stage divided into tertiles by the level of PDPN expression in the publicly available CRC (all stages) microarray gene expression dataset GSE39582. Log-rank p-values ​​are associated with the Kaplan-Meier curves. Cox proportional hazards p-values ​​are associated with the univariate model detailed in Table 10. [Figure 13C] FIG. 13C is a plot showing the correlation between PDPN expression and tumor content (percentage of cancer cells) in The Cancer Genome Atlas (TCGA). [Figure 13D] FIG. 13D is a plot showing the correlation between PDPN and activated fibroblast signature (fibroblast score) in TCGA. [Figure 13E] Figure 13E is a set of Kaplan-Meier curves and a table showing the survival probability of patients with stage II CRC divided into tertiles by expression levels of the activated fibroblast signature (Act. Fib tertiles) in the GSE33113 and GSE39582 datasets. Log-rank p-values ​​are associated with the Kaplan-Meier curves. Cox proportional hazards p-values ​​are associated with the univariate model detailed in Table 10. [Figure 13F]Figure 13F is a set of Kaplan-Meier curves and a table showing the survival probability of patients with any stage of CRC divided into tertiles by expression levels of the activated fibroblast signature (Act. Fib tertiles) in the GSE33113 and GSE39582 datasets. Log-rank p-values ​​are associated with the Kaplan-Meier curves. Cox proportional hazards p-values ​​are associated with the univariate model detailed in Table 10. [Figure 14A] Figure 14A is a set of diagrams and micrographs showing the binding of monomeric or tetrameric PD-L1 to PD-1 or PD-L2. The diagrams (top) show the normalized fluorescence intensity detected for cells expressing PD-1 or PD-L2 on their surface that have been contacted with 5 nM, 10 nM, 50 nM, 200 nM, or 500 nM of monomeric or tetrameric PD-L1. The micrographs (bottom) are representative images of fluorescence for cells expressing PD-1 or PD-L2 on their surface that have been contacted with 500 nM of monomeric or tetrameric PD-L1. [Figure 14B] Figure 14B is a set of diagrams and micrographs showing the binding of monomeric or tetrameric poliovirus receptor (PVR) to CD96, CD226, or TIGIT. The diagram (top) shows the normalized fluorescence intensity of fluorescent streptavidin detected for cells expressing CD96, CD226, or TIGIT on their surface contacted with 5 nM, 10 nM, 50 nM, 200 nM, or 500 nM of monomeric or tetrameric PVR. The micrographs (bottom) are representative images of fluorescence for cells expressing CD96, CD226, or TIGIT on their surface contacted with 500 nM of monomeric or tetrameric PVR. [Figure 14C] Figure 14C is a schematic diagram showing the design of members of the ectodomain-gD-GPI library. Left: untagged full-length single-transmembrane (STM) protein. Right: ectodomain-gD-GPI protein containing the ectodomain of the STM protein, a glycoprotein D (gD) tag, and a glycosylphosphatidylinositol (GPI) linker. [Figure 14D]Figure 14D is a graph showing quantification of surface expression (as fluorescence intensity per cell) of members of the ectodomain-gD-GPI library, measured using an anti-gD antibody. Representative images of surface staining for undetectable (no), low, medium, and high expressers are shown. Dotted lines indicate arbitrary cutoff values ​​for different expression levels. Expression is representative of two independent assays. [Figure 14E] Figure 14E shows a schematic diagram of an automated cell-based platform for receptor discovery (cell surface interaction screening). (1) A library consisting of approximately 1,200 unique single-spanning-membrane (STM) receptors expressed as STM extracellular domains (ECDs) fused to gD-GPI tags is shown. (2) Tetramerization of the query protein podoplanin (PDPN) is shown. PDPN was expressed as the PDPN extracellular domain (ECD) fused to Avidity AVITAG™ (Avi-tag) for site-specific biotinylation, biotinylated, and conjugated to fluorescent streptavidin (SA) to form high-affinity tetramers. (3) Automated receptor-ligand interaction screening (cell surface interaction screening) is shown. Individual single-spanning-membrane (STM) receptors were transfected into mammalian cells and cultured in individual wells on a plate. High-affinity PDPN tetramers were incubated with the cells at 4°C. Upon PDPN interaction with the STM receptor, the fluorescent SA tag was retained on the cell surface. (4) High-content fluorescence imaging of an individual well from the assay in (3). (5) A representation of the fluorescence signal intensity is shown. The background is calculated by averaging the signal intensities of all receptor interactions in the library. (6) A representative surface plasmon resonance (SPR) plot is shown. SPR is used as an orthogonal method to further validate the interaction. [Figure 14F]Figure 14F is a cross plot showing the results of an automated cell surface interaction screen in which the immune receptor B7-H3 (CD276) was tested for interaction with a library of ectodomain-gD-GPI STM proteins. Each circle represents a binding interaction between B7-H3 and an STM receptor. Unique high-scoring hits are shown in red circles. Hits shown in gray circles are empirically determined nonspecific binders. Interleukin-20 receptor subunit alpha (IL20-RA) was identified as an interacting partner. [Figure 14G] Figure 14G is a cross plot showing the results of an automated cell surface interaction screen in which the immune receptor B7-H3 (CD276) was tested for interaction with a library of tag-free, full-length STM proteins. Each circle represents a binding interaction between B7-H3 and an STM receptor. Unique, high-scoring hits are shown in red circles. Hits shown in gray circles are empirically determined nonspecific binders. Interleukin-20 receptor subunit alpha (IL20-RA) was identified as an interacting partner. [Figure 14H] Figure 14H is a cross plot showing the results of an automated cell surface interaction screen in which podoplanin (PDPN) was tested for interaction with a library of STM proteins. Hits, indicated by gray circles, are empirically determined nonspecific binders. Each interaction was tested in duplicate. Each circle represents a binding interaction between PDPN and an STM receptor. CD177 was identified as a novel interacting partner. [Figure 14I] Figure 14I is a sensorgram showing the binding of PDPN to CD177 analyzed by surface plasmon resonance (SPR). Recombinant CD177 (expressed as the ECD) was immobilized on a sensor chip, and recombinant PDPN (expressed as an Fc-tagged ECD) and control proteins were injected at the indicated concentrations. The dissociation constant (KD) value for each interaction, measured using recombinant PDPN expressed as a monomeric ectodomain, is shown. [Figure 14J]Figure 14J is a sensorgram showing the absence of binding between a control ECD and PDPN as analyzed by SPR. The control ECD was immobilized on a sensor chip, and recombinant PDPN (expressed as an Fc-tagged ECD) and control proteins were injected at the indicated concentrations. The dissociation constant (KD) values ​​for each interaction, measured using recombinant PDPN expressed as a monomeric ectodomain, are shown. [Figure 15A] Figure 15A is a pair of graphs showing CD177 expression on neutrophils. The left panel is a representative histogram showing CD177 expression on neutrophils in healthy blood from three different donors and an isotype control. The right panel is a graph showing the percentage of neutrophils that were CD177+ in donor blood samples (n=31). [Figure 15B] Figure 15B is a set of plots showing the gating strategy for identifying neutrophils in healthy volunteer blood, CRC patient blood, adjacent normal colon tissue (adjacent colon), and cancerous colon tissue (CRC colon) using flow cytometry. The blood plot represents data from a blood sample in which red blood cells (RBCs) were lysed. The colon tissue plot represents data from a single-cell suspension of a colon tissue sample in which red blood cells were lysed and the suspension was filtered through a 70-micron mesh. All samples were stained with 7-aminoactinomycin D (7-AAD) to gate out dead cells and then incubated with the indicated antibodies. The shaded area represents the data selected for analysis ("gate"). The gate was selected to include events that were live (indicated by the absence of 7-AAD staining), singlet, and CD45+. Data are representative of 4 to 7 independent donors. [Figure 15C] Figure 15C is a graph showing the percentage of cells that were neutrophils in samples from 4-7 independent donors for healthy volunteer blood (normal blood), CRC patient blood (pt blood), adjacent normal colon tissue (adj colon), and cancerous colon tissue (CRC colon). ***p<0.001, Kruskal-Wallis and Dunn's multiple comparison test. [Figure 15D]Figure 15D is a graph showing the percentage of neutrophils that were CD177+ in samples from 4 to 7 independent donors for blood of healthy volunteers, blood of CRC patients, adjacent normal colon tissue, and cancerous colon tissue. [Figure 15E] Figure 15E is a representative histogram showing neutrophil CD177 expression levels in healthy blood, patient blood, adjacent normal (adj) colon tissue, and cancerous (CRC) colon tissue compared to isotype controls. Data are representative of 4-7 independent donors. [Figure 15F] Figure 15F is a set of plots showing the gating strategy for the identification of stromal cells in adjacent normal colon tissue, cancerous (CRC), and diverticulitis (Div) colon tissue using flow cytometry. The plots represent data from single-cell suspensions of tissue samples. Samples were stained as indicated. Gates were selected to include events that were live (indicated by the absence of 7-AAD staining), singlet, CD45- (i.e., not immune cells), and EpCAM- (i.e., not tumor cells) for the examination of stromal cells. DN = double-negative T cells; BEC = vascular endothelial cells. Data are representative of 3 to 7 independent samples. [Figure 15G] Figure 15G is a graph showing the percentage of cells that are PDPN+ in CRC colon cells that are cancer-associated fibroblast (CAF) cells (EpCAM-, CD45-, CD31-); endothelial cells (EC) (EpCAM-, CD45-, CD31+); tumor cells (TC) (EpCAM+), myeloid cells (CD45+, CD11B+), CD4 T cells (CD45+, CD3+, CD4+); or CD8 T cells (CD45+, CD3+, CD8+). Data are representative of 3–7 independent samples. **p<0.01, ***p<0.001, Kruskal-Wallis and Dunn's multiple comparison test. [Figure 15H]Figure 15H is a graph showing the mean fluorescence intensity (MFI) of PDPN in CRC colon cells, which are cancer-associated fibroblast (CAF) cells (EpCAM-, CD45-, CD31-); endothelial cells (EC) (EpCAM-, CD45-, CD31+); tumor cells (TC) (EpCAM+), myeloid cells (CD45+, CD11B+), CD4 T cells (CD45+, CD3+, CD4+); or CD8 T cells (CD45+, CD3+, CD8+). Data are representative of 3–7 independent samples. **p<0.01, ****p<0.0001, Kruskal-Wallis and Dunn's multiple comparison test. [Figure 15I] Figure 15I is a graph showing the percentage of PDPN+ fibroblasts in adjacent normal (adj), cancerous (CRC), and diverticulitis (div) colon tissue. Data are representative of 3 to 7 independent samples. [Figure 15J] Figure 15J is a graph showing the MFI of PDPN fibroblasts in adjacent normal (adj), cancerous (CRC), and diverticulitis (div) colon tissues. Data are representative of 3 to 7 independent samples. [Figure 16A] Figure 16A is a pair of photomicrographs from a tissue microarray showing adjacent normal colon tissue stained for CD177 (blue) and PDPN (pink). PDPN staining was largely absent from fibroblasts and marked lymphatic vessels. CD177 staining was barely observed. The bottom panel shows a magnified image. [Figure 16B] Figure 16B is a pair of photomicrographs from a tissue microarray showing cancerous (CRC) colon tissue stained for CD177 (blue) and PDPN (pink). PDPN staining was largely absent in fibroblasts, marking lymphatic vessels. CD177 staining was barely observed. This tumor showed strong PDPN staining in the stroma surrounding the tumor bed, but no staining was seen in the epithelial cells themselves. The lower panel shows a magnified image. [Figure 16C]Figure 16C is a set of photomicrographs showing representative images of double immunofluorescence staining for PDPN and CD177 in cancerous (CRC) colon cells. The first panel shows an overview of the tissue with PDPN (green) and CD177 (red) staining. The second, third, and fourth panels show insets of the photomicrograph in the first panel, as indicated by the boxes in the first panel. The second panel shows PDPN and CD77 staining. The third panel shows CD177 staining only. The fourth panel shows PDPN staining only. Scale bar 50 μm. [Figure 16D] Figure 16D is a set of photomicrographs showing representative images of double immunofluorescence staining of myeloperoxidase (MPO; a marker for neutrophils) and CD177 in CRC cells. The first panel shows an overview of the tissue with MPO (green) and CD177 (red) staining. The second, third, and fourth panels show insets of the photomicrograph in the first panel, as indicated by the boxes in the first panel. The second panel shows MPO and CD77 staining. The third panel shows CD177 staining only. The fourth panel shows MPO staining only. Scale bar 50 μm. [Figure 16E] FIG. 16E is a bar graph showing the percentage of normal adjacent colon (normal) and CRC cancer (tumor) cells that were negative (-) or positive (+) for PDPN or CD177 staining. [Figure 17A] Figure 17A is a bar graph showing the morphology index of wild-type cancer-associated fibroblasts (CAFs) seeded in 3D gels and treated with isotype control, recombinant human CLEC-2 (rCLEC-2), or recombinant human CD177 (r-CD177). Dots represent single wells containing more than 50 cells and six independent experiments. Data are plotted relative to the isotype control for each experiment. **p<0.01, Kruskal-Wallis and Dunn's multiple comparison test. [Figure 17B]Figure 17B is a set of photomicrographs showing representative images of cancer-associated fibroblasts in a 3D gel stained for actin and DAPI. The left panel shows cells treated with an isotype control. The middle panel shows cells treated with recombinant human CLEC-2 (rCLEC-2). The right panel shows cells treated with recombinant human CD177 (rCD177). Scale bar: 20 μm. [Figure 17C] Figure 17C is a bar graph showing the morphology index when wild-type cancer-associated fibroblasts (CAFs) were seeded onto 3D gels alone (-) or at a 5:1 ratio with primary neutrophils (neutrophils) or T cells isolated from blood. Dots represent a single well containing more than 50 cells and two to three independent experiments, each representing five donors. **p<0.01, Kruskal-Wallis and Dunn's multiple comparison test. [Figure 17D] Figure 17D is a bar graph showing the morphology index of primary human fibroblasts from healthy human bladder, colon, or ovary (HOF) tissue seeded in 3D gels and treated with isotype control, recombinant human CLEC-2 (r-CLEC2), or recombinant human CD177 (r-CD177). Dots represent a single well containing more than 50 cells and three independent experiments. *p<0.05, Kruskal-Wallis and Dunn's multiple comparison test. [Figure 17E] FIG. 17E is a representative histogram showing staining of PDPN expression on fibroblasts from healthy human bladder, colon, and ovary (HOF) tissue and isotype controls. [Figure 17F] Figure 17F is a bar graph showing contraction (μm) of wild-type cancer-associated fibroblasts (CAFs) treated with isotype control, recombinant human CLEC-2 (rCLEC-2), or recombinant human CD177 (r-CD177) compared to unstimulated cells (un). Dots represent the mean of 2–3 wells per condition. The graph contains data from four independent experiments. *p<0.05, Kruskal-Wallis and Dunn's multiple comparison test. [Figure 18A]Figure 18A shows a schematic diagram of the workflow for multiplexed global proteomics and phosphoproteomics experiments in parallel with tandem mass tagging and fractionation for deep coverage. Cancer-associated fibroblasts (CAFs) were treated in duplicate with DMSO control, CLEC-2, or CD177 for 2 or 30 minutes, followed by lysis. Protein lysates were reduced and digested with LysC and trypsin, followed by peptide normalization. Peptides were labeled with tandem mass tags (TMTs). A portion of the peptides (approximately 0.5 mg) was used for protein profiling. These peptides were fractionated by high-pH reversed-phase fractionation (Hi pH RP). Samples were run on HPLC, and peptides were identified by liquid chromatography-mass spectrometry (LC-MS / MS). The remaining peptides (approximately 6.5 mg) were used for global phosphorylation analysis. Peptides were fractionated by strong cation exchange chromatography (SCX) and analyzed by liquid chromatography-mass spectrometry (LC-MS / MS). [Figure 18B] Figure 18B is a heatmap showing the fold change in protein phosphorylation (compared to untreated controls) for all phosphorylation sites that were significantly altered (|Log2FC|>1 and p<0.05) in CAFs treated with CLEC-2 or CD177 for 2 minutes (2') or 30 minutes (30'). [Figure 18C] FIG. 18C is a volcano plot showing significant changes in protein phosphorylation in CAFs stimulated with CD177 (circles) or CLEC-2 (triangles) compared to untreated cells after 2 minutes. [Figure 18D] FIG. 18D is a volcano plot showing significant changes in protein phosphorylation in CAFs stimulated with CD177 (circles) or CLEC-2 (triangles) compared to untreated cells after 30 minutes. [Figure 18E] Figure 18E is a bar graph showing enriched (q<0.05) Gene Ontology (GO) pathways represented by proteins with significantly altered phosphorylation sites in CLEC-2- or CD177-treated CAFs compared to untreated CAFs. Numbers indicate the phosphorylation site group associated with each curated GO term. [Figure 18F]Figure 18F is a table showing the relative abundance change (compared to unstimulated controls) per treatment condition for selected phosphorylation sites on proteins enriched in biological process pathways. Grouped phosphorylation sites are shown separated by a " / ". [Figure 18G] Figure 18G is a Venn diagram showing approximately 70% overlap (2,912 proteins) between the unique proteins identified in the global proteome (6,309 proteins) and the phosphoproteome (4,122 proteins). Overall, 18,558 unique phosphopeptides with high-quality reporter ion intensities were identified and mapped to 4,122 unique proteins with at least one phosphorylated residue. [Figure 18H] Figure 18H is a Venn diagram summarizing the number of phosphorylation sites significantly altered (p<0.05 and | Log2 fold change|>1) at 30 min by CLEC-2 (190 phosphorylation sites), CD177 (54 phosphorylation sites), or both (32 phosphorylation sites). [Figure 18I] Figure 18I is a triangular plot showing all phosphopeptides detected in the phosphoproteome assay. Significantly altered phosphorylation sites are indicated as upward-pointing triangles (dephosphorylated compared to unstimulated) or downward-pointing triangles (hyperphosphorylated compared to unstimulated) and are color-coded according to the fold change measured between CLEC-2 at 30 minutes (purple) or CD177 at 30 minutes (green). [Figure 19A] Figure 19A is a set of Kaplan-Meier curves and a table showing the survival probability of patients with stage II disease divided into tertiles by level of podoplanin (PDPN) expression (T1, T2, and T3) in the publicly available CRC microarray gene expression dataset GSE39582. Logrank p-values ​​(Logrank p) are associated with the Kaplan-Meier curves. Cox proportional hazards (CoxPH) p-values ​​are associated with the univariate model detailed in Table 10. [Figure 19B]Figure 19B is a set of Kaplan-Meier curves and a table showing the survival probability of patients with stage IV disease divided into tertiles by level of podoplanin (PDPN) expression (T1, T2, and T3) in the publicly available CRC microarray gene expression dataset GSE39582. Logrank p-values ​​(Logrank p) are associated with the Kaplan-Meier curves. Cox proportional hazards (CoxPH) p-values ​​are associated with the univariate model detailed in Table 10. [Figure 19C] Figure 19C is a set of Kaplan-Meier curves and a table showing the survival probability of patients with stage II disease divided into tertiles by the level of FAP+ fibroblast (FAP+ fib) signature expression (T1, T2, and T3) in the publicly available CRC microarray gene expression dataset GSE33113. Logrank p-values ​​(Logrank p) are associated with the Kaplan-Meier curves. Cox proportional hazards (CoxPH) p-values ​​are associated with the univariate model detailed in Table 10. [Figure 19D] Figure 19D is a set of Kaplan-Meier curves and a table showing the survival probability of patients with all stages of disease divided into tertiles by the level of FAP+ fibroblast signature expression (T1, T2, and T3) in the publicly available CRC microarray gene expression dataset GSE39582. Logrank p-values ​​(Logrank p) are associated with the Kaplan-Meier curves. Cox proportional hazards (CoxPH) p-values ​​are associated with the univariate model detailed in Table 10. [Figure 19E] Figure 19E is a set of Kaplan-Meier curves and a table showing the survival probability of patients with stage II disease divided into tertiles by the level of activity (Act.) fibroblast signature expression (T1, T2, and T3) in the publicly available CRC microarray gene expression dataset GSE39582. Logrank p-values ​​(Logrank p) are associated with the Kaplan-Meier curves. Cox proportional hazards (CoxPH) p-values ​​are associated with the univariate model detailed in Table 10. [Figure 19F]Figure 19F is a set of Kaplan-Meier curves and a table showing the survival probability of patients with stage II disease divided into tertiles by the level of FAP+ fibroblast signature expression (T1, T2, and T3) in the publicly available CRC microarray gene expression dataset GSE39582. Logrank p-values ​​(Logrank p) are associated with the Kaplan-Meier curves. Cox proportional hazards (CoxPH) p-values ​​are associated with the univariate model detailed in Table 10. [Figure 20] Figure 20 shows a cross plot of the results of an automated cell surface interaction screen in which podoplanin (PDPN) expressed as a tetramer using fluorescent streptavidin was tested in duplicate for interaction with a library of full-length, untagged STM proteins. The newly identified binding partner, CD177, was not present in this receptor library. Images were acquired from individual wells using a high-content microscope and processed using InCell Developer software to express the amount of PDPN bound to the cell surface as signal to background (S / N ratio). Each circle represents a binding interaction between PDPN and an STM receptor. Green dots represent nonspecific binders observed as hits in an unrelated screen. CLEC-2 was identified as a novel interaction partner. [Figure 21] Figure 21 shows sensorgrams depicting the binding of PDPN and CLEC-2 analyzed by SPR. Recombinant CLEC-2 was immobilized on a sensor chip, and purified PDPN (expressed as an Fc-tagged ECD) and an unrelated Fc-tagged ECD control protein were injected at the indicated concentrations. The dissociation constant (KD) values ​​for each interaction, measured using recombinant PDPN expressed as a monomeric ectodomain, are shown. [Figure 22A] FIG. 22A is a boxplot showing the levels of CD177 RNA expression (log2 nRPKM) in normal colon tissue and CRC tumors in The Cancer Genome Atlas (TCGA) dataset. [Figure 22B]Figure 22B is a boxplot showing the levels of PDPN RNA expression (log2 nRPKM) in normal colon tissues and CRC tumors in The Cancer Genome Atlas (TCGA) dataset. [Figure 22C] Figure 22C is a boxplot showing the levels of PDPN and CD177 RNA expression in CRC tumors from the GSE39582 dataset. [Figure 22D] Figure 22D is a box plot showing the levels of PDPN and CD177 RNA expression in CRC tumors from the GSE33113 dataset. [Figure 23A] Figure 23A is a set of photomicrographs showing serial sections of normal adjacent (adj) colon tissue stained for PDPN (left panel) and CD177 (right panel). PDPN staining was largely absent from fibroblasts and marked lymphatic vessels. CD177 staining was largely absent. [Figure 23B] Figure 23B is a set of photomicrographs showing serial sections of cancerous (CRC) colon tissue stained for PDPN (left panel) and CD177 (right panel). The inset shows a magnified image. The tumor showed strong PDPN staining in the stroma surrounding the tumor bed, but no staining was seen in the epithelial cells themselves. [Figure 23C] Figure 23C is a set of photomicrographs showing representative images of double immunofluorescence staining for PDPN and CD177 in adjacent normal colon tissue. The first panel shows an overview of the tissue with PDPN and CD177 staining. The second, third, and fourth panels show insets of the photomicrograph in the first panel, as indicated by the boxes in the first panel. The second panel shows PDPN and CD77 staining. The third panel shows CD177 staining only. The fourth panel shows PDPN staining only. [Figure 23D]Figure 23D is a set of photomicrographs showing representative images of double immunofluorescence staining for myeloperoxidase (MPO; a marker for neutrophils) and CD177 in adjacent normal colon tissue. The first panel shows an overview of the tissue with MPO and CD177 staining. The second, third, and fourth panels show insets of the photomicrograph in the first panel, as indicated by the boxes in the first panel. The second panel shows PDPN and CD77 staining. The third panel shows CD177 staining only. The fourth panel shows MPO staining only. [Figure 24A] Figure 24A is a set of density plots showing the observed changes in phosphorylated residues (x-axis) and changes in total protein abundance (y-axis) for proteins identified in both proteomic and phosphoproteomic assays in CAFs stimulated with CD177 or CLEC-2 for 2 or 30 minutes. [Figure 24B] Figure 24B is a heat map showing the fold change in protein phosphorylation (compared to untreated controls) for all phosphorylation sites that were significantly altered (|Log2FC|>1 and p<0.05) in CAFs treated with CLEC-2 or CD177 for 2 minutes (2') or 30 minutes (30'), including the identities of phosphoproteins where significant changes were observed, as shown in Figure 18B. [Figure 25] Figure 25 shows a pair of diagrams depicting a model of the interaction between CD177 and PDPN in the tumor microenvironment. The left panel shows a tissue overview. Dense bands of CAFs form between islands in the tumor bed. These fibers generally run parallel to the tumor bed and contain many immune cells. In particular, neutrophils and regulatory T cells (Tregs) can be found in these stroma-rich areas. Although there are still many PDPN+ CAFs that are not in contact with these CD177+ immune cells, some CAFs interact with these cells and likely receive inhibitory signals downstream of PDPN. The right panel shows a model of the molecular interaction between CD177 and PDPN and the downstream consequences in CAFs. CD177 binding alters contraction, motility, extracellular matrix (ECM) remodeling, and metabolism in CAFs. [Figure 26A]FIG. 26A is a histogram showing the expression of podoplanin (PDPN) on wild-type (WT) cancer-associated fibroblasts (CAFs) and Pdpn − / − CAFs compared to isotype control samples. [Figure 26B] Figure 26B is a graph showing the morphology index (perimeter 2 / 4 * area) of WT and Pdpn- / - CAFs seeded in 3D gels. Each dot represents the average of one well containing more than 50 cells, and the plot is representative of three independent experiments. *p<0.05, Mann-Whitney U test. [Figure 26C] FIG. 26C is a pair of photomicrographs showing WT (left) and Pdpn − / − (right) CAFs in a 3D gel stained for actin (red) and nuclei (DAPI; blue). [Figure 26D] Figure 26D is a graph showing the relative contraction of Pdpn − / − cells compared to WT cells. Each dot represents the average of 3–4 wells from 4 independent experiments. *p<0.05, Mann-Whitney U test. [Figure 26E] Figure 26E is a graph showing the percent confluency of WT and Pdpn- / - CAFs over time. Each point represents the average of 16 different fields from four wells per condition, and plots are representative of four to six independent experiments. ****p<0.0001, ANOVA. [Figure 27A] Figure 27A is a set of photomicrographs showing representative images from a single well of SW480 tumor organoids expressing red fluorescent protein (RFP) in 3D culture containing no CAFs (tumor only), Pdpn- / - CAFs, or wild-type (WT) CAFs at day 8 of growth. Wells without tumor cells (WT CAFs only and Pdpn- / - CAFs only) are shown as controls. [Figure 27B] Figure 27B is a graph showing the total area (in μm) of spheroids from maximum intensity projections of the wells in Figure 27A at time points between 1 and 8 days of culture. Data are representative of two independent experiments (n=4 wells per condition per experiment). Dunn's multiple comparison test was performed on all conditions versus tumor-only controls. *p<0.05, ****p<0.0001. [Figure 27C] Figure 27C is a set of photomicrographs showing representative images from a single well of RFP-expressing SW480 tumor organoids in 3D culture at day 17 of growth, containing no CAFs (tumor only), WT CAFs, or wild-type WT CAFs treated with control, tetrameric CD177, or tetrameric CLEC-2. A well without tumor cells (WT CAFs only) is shown as a control. [Figure 27D] Figure 27D is a graph showing the total area (in μm) of spheroids from maximum intensity projections of the wells in Figure 27C at time points between 1 and 17 days of culture. Data are representative of two independent experiments (n=4 wells per condition per experiment). Dunn's multiple comparison test showed a significant difference of ****p<0.0001 for the tumor WT CAF group compared to all other conditions except the tumor + WT CAF + control group. [Figure 28A] Figure 28A is a heatmap showing the association of all genes in the IgSF interactome with gene sets based on the CD8+ Teff cell signature, the pan-fibroblast TFGb signature, and the Lund subtyping scheme of immune desert (UroA: urothelial-like A), immune exclusion (Inf: infiltrating), or immune inflammatory tumors (UroB: urothelial-like B, SCCL: basal / SCC-like). [Figure 28B] Figure 28B is a volcano plot showing protein interactions significantly associated with positive (blue) or negative (red) clinical outcomes. Selected interactions are highlighted. [Figure 28C] Figure 28C is a scatter plot showing interactions with high synergy, visualized by comparing the hazard ratio calculated from the co-expression of interacting pairs with the combined hazard ratio of individual genes. Selected interactions that improved the predictive power of clinical outcomes over single genes are highlighted. Red represents interactions predicting lack of response, while blue indicates interactions associated with response. [Figure 28D]Figure 28D is a forest plot showing increased significance for the association with lack of response and patient survival for each of the EFNB1 interactions compared to the individual genes. The EFNB1 / EVC2 interaction (HR: 1.61; 95% CI: 1.24; 2.09, P=3.78e-4) improves prediction of poor overall survival compared to the individual genes EFNB1 (HR: 1.25; 95% CI: 0.97; 1.62, P=0.087) and EVC2 (HR: 1.32; 95% CI: 1.02; 1.71, P=0.035, n=348 patients). [Figure 28E] Figure 28E is a survival plot showing the survival probability of individuals whose EFNB1 expression level is above, below, or equal to the median expression level, individuals whose EVC2 expression level is above, below, or equal to the median expression level, or individuals whose EFNB1 and EVC2 expression levels are above, below, or equal to the median expression level. [Figure 28F] Figure 28F shows boxplots showing the interaction of EFNB1 / ECV2 with each gene individually. Whisker plots represent minimum and maximum values, and filled circles represent outliers. Y-axis: Z-score expression. Responders: complete response (CR) and partial response (PR); non-responders: progressive disease (PD) and stable disease (SD). (p-values: EFBN1: 0.0008901, EVC2: 0.009046, EFBN1 / EVC2: 9.19x10-5, n=298 patients). [Figure 29A] Figure 29A is a heatmap showing differential expression clustered between tumor and adjacent normal samples by TCGA tumor indication (columns) for all IgSF interactome genes (rows). Cell values ​​represent the mean change between the log2 rsem values ​​of tumor and adjacent normal. [Figure 29B] FIG. 29B is a network diagram showing genes that are jointly upregulated (red) and downregulated (blue) among the LILR family of proteins in the TCGA tumor indication HNSC (head and neck squamous cell carcinoma). [Figure 29C]FIG. 29C is a network diagram showing jointly upregulated (red) and downregulated (blue) genes among the LILR family of proteins in the TCGA tumor indication KIRC (kidney renal clear cell carcinoma). [Figure 29D] FIG. 29D is a clustered heatmap showing the normalized protein expression values ​​of network genes in CCLE, partitioned by tissue. [Figure 29E] Figure 29E is a scatter plot comparing the relative protein expression of reported binding partners IGSF3 and PTGFRN in lung CCLE tissue subsets. Expression patterns were significantly correlated (q<0.2) with the overlaid regression model (red). [Figure 29F] Figure 29F is a scatter plot comparing the relative protein expression of reported binding partners IGSF3 and PTGFRN in upper aerodigestive tract CCLE tissue subsets. Expression patterns were significantly correlated (q<0.2) with the overlaid regression model (red). [Figure 29G] Figure 29G is a scatter plot comparing the relative protein expression of reported binding partners IGSF3 and PTGFRN in esophageal CCLE tissue subsets. Expression patterns were significantly correlated (q<0.2) with the overlaid regression model (red). [Figure 29H] Figure 29H is a scatter plot comparing the relative protein expression of reported binding partners CEACAM5 and CEACAM6 in lung CCLE tissue subsets. Expression patterns were significantly correlated (q<0.2) with the overlaid regression model (red). [Figure 29I] Figure 29I is a scatter plot showing a comparison of the relative protein expression of the reported binding partners CEACAM5 and CEACAM6 in lung CCLE tissue subsets. The expression patterns were significantly negatively correlated (q<0.2) with the overlaid regression model (red). [Figure 29J]Figure 29J is a scatter plot comparing the relative protein expression of the reported binding partners ICOSLG and NTM in lung CCLE tissue subsets. Expression patterns were significantly negatively correlated (q<0.2) with the overlaid regression model (red). [Figure 30A] Figure 30A is a volcano plot showing protein interactions significantly correlated with CD8+ Teff cells. Selected interactions are highlighted. [Figure 30B] Figure 30B is a network diagram showing the PD-1 / PD-L1 community of immune-related interactions color-coded by hazard ratio, visualizing the inhibition of the interaction by atezolizumab. [Figure 30C] Figure 30C is a network diagram showing selected binding partners of LRRC4B color-coded by hazard ratio. [Figure 30D] FIG. 30D is a network diagram showing selected interactions within the ephrin family, color-coded by hazard ratio. [Figure 30E] Figure 30E is a forest plot showing increased significance of association with lack of response and patient survival for each of the CD274(PD-L1) interactions compared to individual genes. [Figure 30F] Figure 30F is a forest plot showing increased significance of association with lack of response and patient survival for each of the LRRC4B interactions compared to individual genes. DETAILED DESCRIPTION OF THE INVENTION

[0173] Detailed Description of the Invention I. Definition As used herein, the term "about" refers to a normal range of error for the respective value, which is readily known to one of ordinary skill in the art. Reference to "about" with respect to a value or parameter herein includes (and describes) aspects directed to the value or parameter itself.

[0174] As used herein, the term "single-spanning membrane receptor," "single-spanning membrane receptor," or "STM receptor" refers to a protein having a single transmembrane domain. In some embodiments, the STM receptor is expressed on the cell surface. Exemplary STM receptors are provided in Tables 5, 7, and 8, as well as Martinez-Martin et al., Cell, 174(5): 1158-1171, 2018 and Clark et al., Genome Res, 13: 2265-2270, 2003. In some embodiments, STM proteins have the UniProt annotation "leucine-rich," "cysteine-rich," "ITIM / ITAM" (immunoreceptor tyrosine-based inhibitory motif / immunoreceptor tyrosine-based activation motif), "TNFR" (tumor necrosis factor receptor), "TLR / ILR" (Toll-like receptor / interleukin receptor), "semaphorin," "kinase-like," "Ig-like" (immunoglobulin-like), "fibronectin," "ephrin," "EGF," "cytokine R," or "cadherin." STM receptors can be identified, for example, based on the presence of a signal peptide or predicted transmembrane region in the amino acid sequence. In some embodiments, STM receptors are expressed as extracellular domains.

[0175] As used herein, the term "immunoglobulin superfamily protein" or "IgSF protein" refers to a protein that contains at least one immunoglobulin (Ig) domain or immunoglobulin fold and that has, for example, an annotation in the UniProt database of "immunoglobulin-like superfamily" or is otherwise shown to have structural or functional similarity to such proteins. In some embodiments, an IgSF protein has the annotation "immunoglobulin-like domain superfamily" in the UniProt database. In some embodiments, an IgSF protein is included based on its primary biological activity, e.g., its involvement in leukocyte activation, cell-cell adhesion, cell communication, or signal transduction. In some embodiments, an IgSF protein has the UniProt annotation "TFNR" (transcription factor-like nuclear regulator), "TLR / ILR (Toll-like receptor / interleukin receptor)," "semaphorin," "kinase-like," "IgSF / Ig-like fold," "Ig-like fold," "fibronectin," "ephrin," "EGF," "cytokine R," or "cadherin." In some embodiments, the IgSF protein is expressed on the cell surface. In other embodiments, the IgSF protein is secreted. Exemplary IgSF proteins are provided in Table 4 and in Ozkan et al., Cell, 154(1): 228-239, 2013 and Yap et al., J Mol Biol, 426(4), 945-961.In some embodiments, the IgSF superfamily protein is programmed cell death 1 ligand 1 (PD-L1; CD274), programmed cell death 1 ligand 2 (PD-L2; CD274; PDCD1LG2), receptor tyrosine-protein phosphatase delta (PTPRD), receptor tyrosine-protein phosphatase S (PTPRS), receptor tyrosine-protein phosphatase S (PTPRF), neural cell adhesion molecule L1-like protein ("near homolog of L1"; CHL1), contactin 1 (CNTN1), leukocyte immunoglobulin-like receptor subfamily B member 1 (LILRB1), leukocyte immunoglobulin-like receptor subfamily B member 3 (LILRB3), leukocyte immunoglobulin-like receptor subfamily B member 4 (LILRB4), leukocyte immunoglobulin-like receptor subfamily B member 5 (LILRB5), MAM domain-containing glycosylphosphatidylinositol-anchored protein 1 (MDGA1), or tyrosine protein kinase receptor AXL (UFO). In some embodiments, the IgSF protein is expressed as an extracellular domain. In some embodiments, the IgSF protein is a secreted protein.

[0176] As used herein, the term "immunoglobulin domain" or "Ig domain" refers to a domain of an IgSF protein characterized by about 7-9 antiparallel β-strands comprising a two-layer β-sheet sandwich spanning about 70-125 amino acid residues. In some embodiments, an immunoglobulin domain contains a conserved disulfide bond connecting its B and F strands. Exemplary IgSF proteins are described in Bork et al., JMB, 242(4): 309-320, 1194 and Yap et al., J Mol Biol, 426(4), 945-961.

[0177] As used herein, the term "extracellular domain" or "ECD" refers to a protein domain predicted to be localized on the outer side of the plasma membrane outside the cell. In some examples, the ECD is the ECD of a receptor, e.g., an STM receptor. In some embodiments, the ECD is the ECD of an IgSF protein. In some embodiments, the ECD is the ECD of PDPN. In some embodiments, the boundaries of an extracellular domain can be identified by the prediction of a domain that indicates that the protein crosses the plasma membrane, e.g., a transmembrane domain (e.g., a transmembrane helix). In some embodiments, the presence of an extracellular domain can be predicted by the presence of a domain, sequence, or motif that indicates that the protein is transported to the plasma membrane, e.g., a signal sequence or a glycosylphosphatidylinositol (GPI) attachment site. In some embodiments, the boundaries of an ECD are determined according to UniProt annotation. In some embodiments, the ECD is soluble. In some embodiments, the extracellular domain is expressed in the context of a full-length protein. In other embodiments, the extracellular domain is expressed as an isolated extracellular domain, eg, a sequence of amino acid residues that includes only amino acid residues of the protein that are predicted to be extracellular.

[0178] In some embodiments, the isolated ECD is comprised in a fusion protein. In some embodiments, inclusion in a fusion protein increases solubility, ease of expression, ease of capture (e.g., on a protein A-coated plate), multimerization, or other desirable properties of the ECD. In some embodiments, the ECD or ECD fusion protein is monomeric. In other embodiments, the ECD or ECD fusion protein is multimeric, e.g., a tetramer or pentamer. In some embodiments, the ECD is fused to human IgG. In some embodiments, the ECD is fused to a human Fc tag. In some embodiments, the ECD is fused to an Avidity AVITAG TMIn some embodiments, the ECD is fused to a glycoprotein D (gD) tag and a glycosylphosphatidylinositol (GPI) linker, e.g., a gD-GPI tag. In other embodiments, the ECD is fused to the pentamerization domain of rat cartilage oligomeric matrix protein (COMP) and a β-lactamase protein, e.g., as described in Bushell et al., Genome Res, 18: 622-630, 2008. In some embodiments, the ECD fusion protein further comprises a cleavage sequence, e.g., a TEV cleavage sequence, to allow for removal of one or more domains. In some examples, an ECD fusion protein having an Avi tag and an Fc tag cleavable at a cleavage sequence is further treated to remove the Fc tag, biotinylate the Avi tag, and fuse the biotinylated ECD fusion protein to fluorescent streptavidin (SA), e.g., to form a tetramerized ECD fusion protein. In some examples, the isolated ECD or ECD fusion protein is purified.

[0179] As used herein, a "modulator" is an agent that modulates (e.g., increases, decreases, activates, or inhibits) a given biological activity, e.g., an interaction or a downstream activity resulting from the interaction. A modulator or candidate modulator can be, for example, a small molecule, an antibody, an antigen-binding fragment (e.g., bis-Fab, Fv, Fab, Fab'-SH, F(ab')2, diabody, linear antibody, scFv, ScFab, VH domain, or VHH domain), a peptide, a mimetic, an antisense oligonucleotide, or a small interfering RNA (siRNA).

[0180] By "increase" or "activate" is meant the ability to cause an overall increase, e.g., of 20% or more, 50% or more, or 75%, 85%, 90%, or 95% or more. In certain aspects, increasing or activating can refer to downstream activity of a protein-protein interaction.

[0181] "Decrease" or "inhibit" refers to the ability to cause an overall decrease, e.g., by 20% or more, 50% or more, or 75%, 85%, 90%, 95% or more. In certain aspects, decrease or inhibit can refer to a downstream activity of a protein-protein interaction.

[0182] "Affinity" refers to the strength of the sum total of non-covalent interactions between a single binding site of a molecule (e.g., a receptor) and its binding partner (e.g., a ligand). As used herein, "binding affinity," unless otherwise indicated, refers to the intrinsic binding affinity that reflects a 1:1 interaction between members of a binding pair (e.g., a receptor and a ligand). Typically, the affinity of a molecule X for its partner Y can be expressed by a dissociation constant (kd). Affinity can be measured by common methods known in the art, including those described herein.

[0183] As used herein, "complex" or "complex-type" refers to an association of two or more molecules that interact with each other through bonds and / or forces (e.g., van der Waals, hydrophobic, hydrophilic forces) that are not peptide bonds. In certain aspects, the complex is a heteromultimer. It should be understood that the term "protein complex" or "polypeptide complex" as used herein includes complexes having non-protein entities (e.g., including, but not limited to, chemical molecules such as toxins or detection agents) conjugated to proteins in the protein complex.

[0184] A "disorder" is any condition that would benefit from treatment, including, but not limited to, chronic and acute disorders or diseases, including the pathologies that predispose a mammal to the disorder in question. In one aspect, the disorder is cancer.

[0185] The terms "cancer" and "cancerous" refer to or describe a physiological condition in mammals typically characterized by uncontrolled cell growth / proliferation. Embodiments of cancer include solid tumor cancers and non-solid tumor cancers. Solid cancer tumors include, but are not limited to, colorectal cancer, head and neck cancer (e.g., squamous cell carcinoma of the head and neck), glioma, melanoma, breast cancer, lung cancer, bladder cancer, kidney cancer, ovarian cancer, pancreatic cancer, or prostate cancer, or metastatic forms thereof. In some embodiments, the cancer is colorectal cancer (CRC). In some embodiments, the cancer is head and neck cancer. Further embodiments of head and neck cancer include squamous cell carcinoma of the head and neck (SCCHN). In some embodiments, the cancer is breast cancer. Further embodiments of breast cancer include hormone receptor-positive (HR+) breast cancer, e.g., estrogen receptor-positive (ER+) breast cancer, progesterone receptor-positive (PR+) breast cancer, or ER+ / PR+ breast cancer. Other aspects of breast cancer include HER2-positive (HER2+) breast cancer. Yet other aspects of breast cancer include triple-negative breast cancer (TNBC). In some aspects, the breast cancer is early-stage breast cancer. In some aspects, the cancer is lung cancer. Further aspects of lung cancer include epidermal growth factor receptor-positive (EGFR+) lung cancer. Other aspects of lung cancer include epidermal growth factor receptor-negative (EGFR-) lung cancer. Still other aspects of lung cancer include non-small cell lung cancer, e.g., squamous cell lung cancer or non-squamous cell lung cancer. Other aspects of lung cancer include small cell lung cancer. In some aspects, the cancer is urinary tract cancer. Urinary tract cancer includes urothelial carcinoma (UC), non-urothelial carcinoma of the urinary tract, and urinary tract cancer with mixed histology. Non-urothelial carcinoma of the urinary tract includes all subtypes described in the World Health Organization classification, such as squamous cell carcinoma, verrucous carcinoma, adenocarcinoma, glandular carcinoma, carcinoma of the Bellini collecting duct, neuroendocrine carcinoma, or small cell carcinoma. Adenocarcinoma can be intestinal adenocarcinoma, mucinous adenocarcinoma, signet ring cell adenocarcinoma, or clear cell adenocarcinoma. Urinary tract cancer can occur in the bladder, renal pelvis, ureter, or urethra. In some embodiments, urinary tract cancer (e.g., urothelial carcinoma, non-urothelial carcinoma, or urinary tract cancer with mixed histology) is classified as stage T4b N according to the TNM classification at the time of treatment initiation. anyor T any The cancer is locally advanced, such as N2-3. In some embodiments, the urinary tract cancer is metastatic urothelial carcinoma (mUC), metastatic non-urothelial carcinoma of the urinary tract, or metastatic urinary tract cancer with mixed histology. In some embodiments, the urinary tract cancer is TNM stage M1 according to the TNM classification at the start of treatment. In some embodiments, the cancer is bladder cancer. Further embodiments of bladder cancer include urothelial bladder cancer (UBC), muscle-invasive bladder cancer (MIBC), or non-muscle-invasive bladder cancer (NMIBC). In some embodiments, the cancer is kidney cancer. Further embodiments of kidney cancer include renal cell carcinoma (RCC). In some embodiments, the cancer is liver cancer. Further embodiments of liver cancer include hepatocellular carcinoma. In some embodiments, the cancer is prostate cancer. Further embodiments of prostate cancer include castration-resistant prostate cancer (CRPC). In some embodiments, the cancer is a metastatic form of a solid tumor. In some embodiments, the metastatic form of a solid tumor is a metastatic form of melanoma, breast cancer, colorectal cancer, lung cancer, head and neck cancer, bladder cancer, kidney cancer, ovarian cancer, pancreatic cancer, or prostate cancer. In some embodiments, the cancer is a non-solid tumor cancer. Non-solid tumor cancers include, but are not limited to, B-cell lymphoma. Further embodiments of B-cell lymphoma include, for example, chronic lymphocytic leukemia (CLL), diffuse large B-cell lymphoma (DLBCL), follicular lymphoma, myelodysplastic syndrome (MDS), non-Hodgkin's lymphoma (NHL), acute lymphoblastic leukemia (ALL), multiple myeloma, acute myeloid leukemia (AML), or mycosis fungoides (MF).

[0186] The terms "host cell," "host cell line," and "host cell culture" are used interchangeably and refer to cells into which exogenous nucleic acid has been introduced, including the progeny of such cells. Host cells include "transfected cells," "transformed cells," and "transformants," which include the primary transformed cell and its progeny, regardless of the number of passages. The progeny may not be completely identical in nucleic acid content to the parent cell and may contain mutations. Included herein are mutant progeny that have the same function or biological activity as screened or selected for in the originally transformed cell. In some aspects, the host cell is stably transformed with the exogenous nucleic acid. In other aspects, the host cell is transiently transformed with the exogenous nucleic acid.

[0187] The term "cancer-associated fibroblasts" ("CAFs") or "tumor-associated fibroblasts," as used herein, refers to fibroblasts (e.g., mammalian stromal cells) present in or associated with the tumor microenvironment (TME), e.g., the stroma. In some aspects, CAFs are activated fibroblasts. CAFs can regulate the structure and / or function of the TME, for example, through extracellular matrix (ECM) remodeling and / or secretion of soluble factors, e.g., growth factors and / or inflammatory factors. CAFs can contribute to tumorigenesis, tumor growth, tumor invasion, angiogenesis, or metastasis. CAFs can impair anti-tumor immunity. In some aspects, CAFs express podoplanin (PDPN). In some aspects, CAFs are characterized by actomyosin contractility, a property that influences tissue stiffness. In some aspects, CAFs exhibit an activated fibroblast signature and / or FAP. + Associated with the fibroblast signature, for example, expressing the genes provided in Table 11 and / or Table 12.

[0188] As used herein, the term "podoplanin" or "PDPN" broadly refers to any native PDPN from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. PDPN is also referred to as gp38, Aggrus, and T1α. The term encompasses full-length PDPN and isolated regions or domains of PDPN, such as the PDPN extracellular domain (ECD). The term also encompasses naturally occurring variants of PDPN, such as splice variants or allelic variants. An exemplary amino acid sequence of human PDPN is set forth under UniProt accession number Q86YL7. Minor sequence variations, particularly conservative amino acid substitutions of PDPN that do not affect the function and / or activity of PDPN, are also contemplated by the present invention.

[0189] As used herein, the term "cluster of differentiation 177" or "CD177" broadly refers to any native CD177 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length CD177 and isolated regions or domains of CD177, such as CD177 ECD. The term also encompasses naturally occurring variants of CD177, such as splice variants or allelic variants. The amino acid sequence of an exemplary human CD177 is set forth under UniProt accession number Q8N6Q3. Minor sequence variations, particularly conservative amino acid substitutions of CD177 that do not affect the function and / or activity of CD177, are also contemplated by the present invention.

[0190] The term "agonist of CD177 activity" or "CD177 agonist" refers to a molecule that increases signaling resulting from the interaction of CD177 with one or more of its binding partners, e.g., PDPN. An agonist of CD177 activity can result in increased binding of CD177 to one or more of its binding partners (e.g., PDPN) compared to binding of the two proteins in the absence of the agonist. Agonists of CD177 activity can include antibodies, antigen-binding fragments thereof, immunoadhesins, fusion proteins, peptides (e.g., multimerized peptides, e.g., multimerized CD177 polypeptides), oligopeptides, and other molecules that increase signaling resulting from the interaction of CD177 with one or more of its binding partners, e.g., PDPN.

[0191] As used herein, the term "programmed cell death 1 ligand 1" or "PD-L1" refers broadly to any native PD-L1 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. PD-L1 is also known as CD274. The term encompasses full-length PD-L1 and isolated regions or domains of PD-L1, such as the PD-L1 ECD. The term also encompasses naturally occurring variants of PD-L1, such as splice variants or allelic variants. An exemplary amino acid sequence of human PD-L1 is set forth under UniProt Accession Number Q9NZQ7. Minor sequence variations, particularly conservative amino acid substitutions in PD-L1 that do not affect the function and / or activity of PD-L1, are also contemplated by the present invention.

[0192] As used herein, the term "ephrin type-A receptor 3" or "EPHA3" broadly refers to any native EPHA3 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length EPHA3 and isolated regions or domains of EPHA3, such as the EPHA3 ECD. The term also encompasses naturally occurring variants of EPHA3, such as splice variants or allelic variants. An exemplary amino acid sequence of human EPHA3 is shown under UniProt accession number P29320. Minor sequence variations, particularly conservative amino acid substitutions of EPHA3 that do not affect the function and / or activity of EPHA3, are also contemplated by the present invention.

[0193] As used herein, the term "programmed cell death 1 ligand 2" or "PD-L2" refers broadly to any native PD-L2 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. PD-L2 is also referred to as PDCD1LG2. The term encompasses full-length PD-L2 and isolated regions or domains of PD-L2, such as the PD-L2 ECD. The term also encompasses naturally occurring variants of PD-L2, such as splice variants or allelic variants. An exemplary amino acid sequence of human PD-L2 is shown under UniProt Accession Number Q9BQ51. Minor sequence variations, particularly conservative amino acid substitutions in PD-L2 that do not affect the function and / or activity of PD-L2, are also contemplated by the present invention.

[0194] As used herein, the term "CEACAM4" broadly refers to any native CEACAM4 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length CEACAM4 and isolated regions or domains of CEACAM4, such as the CEACAM4 ECD. The term also encompasses naturally occurring variants of CEACAM4, such as splice variants or allelic variants. An exemplary amino acid sequence of human CEACAM4 is set forth under UniProt accession number O75871. Minor sequence variations, particularly conservative amino acid substitutions of CEACAM4 that do not affect the function and / or activity of CEACAM4, are also contemplated by the present invention.

[0195] As used herein, the term "receptor tyrosine-protein phosphatase delta" or "PTPRD" broadly refers to any naturally occurring PTPRD from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length PTPRD and isolated regions or domains of PTPRD, such as the PTPRD ECD. The term also encompasses naturally occurring variants of PTPRD, such as splice variants or allelic variants. The amino acid sequence of an exemplary human PTPRD is set forth under UniProt accession number P23468. Minor sequence variations, particularly conservative amino acid substitutions of PTPRD that do not affect the function and / or activity of PTPRD, are also contemplated by the present invention.

[0196] As used herein, the term "receptor tyrosine-protein phosphatase F" or "PTPRF" refers broadly to any naturally occurring PTPRF from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length PTPRF and isolated regions or domains of PTPRF, such as the PTPRF ECD. The term also encompasses naturally occurring variants of PTPRF, such as splice variants or allelic variants. The amino acid sequence of an exemplary human PTPRF is set forth under UniProt Accession Number P10586. Minor sequence variations, particularly conservative amino acid substitutions of PTPRF that do not affect the function and / or activity of PTPRF, are also contemplated by the present invention.

[0197] As used herein, the term "receptor tyrosine-protein phosphatase S" or "PTPRS" refers broadly to any naturally occurring PTPRS from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length PTPRSs and isolated regions or domains of PTPRSs, such as PTPRS ECDs. The term also encompasses naturally occurring variants of PTPRSs, such as splice variants or allelic variants. The amino acid sequence of an exemplary human PTPRS is set forth under UniProt Accession Number Q13332. Minor sequence variations, particularly conservative amino acid substitutions of PTPRSs that do not affect the function and / or activity of the PTPRS, are also contemplated by the present invention.

[0198] As used herein, the term "CHL1" broadly refers to any native CHL1 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length CHL1 and isolated regions or domains of CHL1, such as the CHL1 ECD. The term also encompasses naturally occurring variants of CHL1, such as splice variants or allelic variants. An exemplary amino acid sequence of human CHL1 is set forth under UniProt accession number O00533. Minor sequence variations, particularly conservative amino acid substitutions of CHL1 that do not affect the function and / or activity of CHL1, are also contemplated by the present invention.

[0199] As used herein, the term "contactin 1" or "CNTN1" broadly refers to any native CNTN1 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length CNTN1 and isolated regions or domains of CNTN1, such as the CNTN1 ECD. The term also encompasses naturally occurring variants of CNTN1, such as splice variants or allelic variants. An exemplary amino acid sequence of human CNTN1 is set forth under UniProt accession number Q12860. Minor sequence variations, particularly conservative amino acid substitutions of CNTN1 that do not affect the function and / or activity of CNTN1, are also contemplated by the present invention.

[0200] As used herein, the term "leukocyte immunoglobulin-like receptor subfamily B member 1" or "LILRB1" broadly refers to any naturally occurring LILRB1 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length LILRB1 and isolated regions or domains of LILRB1, such as the LILRB1 ECD. The term also encompasses naturally occurring variants of LILRB1, such as splice variants or allelic variants. An exemplary amino acid sequence of human LILRB1 is set forth under UniProt accession number Q8NHL6. Minor sequence variations, particularly conservative amino acid substitutions of LILRB1 that do not affect the function and / or activity of LILRB1, are also contemplated by the present invention.

[0201] As used herein, the term "leukocyte immunoglobulin-like receptor subfamily B member 2" or "LILRB2" broadly refers to any naturally occurring LILRB2 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length LILRB2 and isolated regions or domains of LILRB2, such as the LILRB2 ECD. The term also encompasses naturally occurring variants of LILRB2, such as splice variants or allelic variants. An exemplary amino acid sequence of human LILRB2 is set forth under UniProt accession number Q8N423. Minor sequence variations, particularly conservative amino acid substitutions of LILRB2 that do not affect the function and / or activity of LILRB2, are also contemplated by the present invention.

[0202] As used herein, the term "leukocyte immunoglobulin-like receptor subfamily B member 3" or "LILRB3" broadly refers to any native LILRB3 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length LILRB3 and isolated regions or domains of LILRB3, such as the LILRB3 ECD. The term also encompasses naturally occurring variants of LILRB3, such as splice variants or allelic variants. The amino acid sequence of an exemplary human LILRB3 is set forth under UniProt accession number O75022. Minor sequence variations, particularly conservative amino acid substitutions of LILRB3 that do not affect the function and / or activity of LILRB3, are also contemplated by the present invention.

[0203] As used herein, the term "leukocyte immunoglobulin-like receptor subfamily B member 4" or "LILRB4" broadly refers to any naturally occurring LILRB4 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length LILRB4 and isolated regions or domains of LILRB4, such as the LILRB4 ECD. The term also encompasses naturally occurring variants of LILRB4, such as splice variants or allelic variants. An exemplary amino acid sequence of human LILRB4 is set forth under UniProt accession number Q8NHJ6. Minor sequence variations, particularly conservative amino acid substitutions of LILRB4 that do not affect the function and / or activity of LILRB4, are also contemplated by the present invention.

[0204] As used herein, the term "leukocyte immunoglobulin-like receptor subfamily B member 5" or "LILRB5" broadly refers to any naturally occurring LILRB5 from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length LILRB5 and isolated regions or domains of LILRB5, such as the LILRB5 ECD. The term also encompasses naturally occurring variants of LILRB5, such as splice variants or allelic variants. An exemplary amino acid sequence of human LILRB5 is set forth under UniProt accession number O75023. Minor sequence variations, particularly conservative amino acid substitutions of LILRB5 that do not affect the function and / or activity of LILRB5, are also contemplated by the present invention.

[0205] As used herein, the term "AXL" broadly refers to any naturally occurring AXL from any mammalian source, including primates (e.g., humans) and rodents (e.g., mice and rats), unless otherwise specified. The term encompasses full-length AXL and isolated regions or domains of AXL, such as the AXL ECD. AXL is also known as UFO. The term also encompasses naturally occurring variants of AXL, such as splice variants or allelic variants. An exemplary amino acid sequence of human AXL is set forth under UniProt accession number P30530. Minor sequence variations, particularly conservative amino acid substitutions of AXL that do not affect the function and / or activity of AXL, are also contemplated by the present invention.

[0206] Unless otherwise indicated, the term "protein," as used herein, refers to any naturally occurring protein from any vertebrate source, including mammals such as primates (e.g., humans) and rodents (e.g., mice, rats), unless otherwise specified. The term encompasses "full-length," unprocessed proteins and all forms of proteins that result from processing within a cell. The term also encompasses naturally occurring variants of proteins, such as splice variants or allelic variants, including amino acid substitution or deletion variants. The term also encompasses isolated regions or domains of proteins, such as the extracellular domain (ECD).

[0207] An "isolated" protein or peptide is one that is separated from a component of its natural environment. In some embodiments, the protein or peptide is purified to greater than 95% or 99% purity, as determined, for example, by electrophoresis (e.g., SDS-PAGE, isoelectric focusing (IEF), capillary electrophoresis) or chromatography (e.g., ion exchange or reverse-phase HPLC).

[0208] An "isolated" nucleic acid refers to a nucleic acid molecule that has been separated from a component of its natural environment. Isolated nucleic acid includes a nucleic acid molecule contained in cells that normally contain the nucleic acid molecule, but where the nucleic acid molecule is present extrachromosomally or in a chromosomal location that is different from its natural chromosomal location.

[0209] As used herein, the term "interactome" refers to a set of molecular interactions, e.g., protein-protein interactions, that occur within a set of molecules. In some embodiments, the interactome is represented as a network, e.g., a network in which nodes represent specific molecules and edges connect the nodes where an assay (e.g., a cell surface interaction assay or an extracellular interaction assay) detects an interaction between two nodes.

[0210] As used herein, the term "vector" refers to a nucleic acid molecule capable of propagating another nucleic acid to which it is linked. The term includes vectors as self-replicating nucleic acid structures as well as vectors that integrate into the genome of a host cell into which they are introduced. Certain vectors are capable of directing the expression of nucleic acids to which they are operatively linked. Such vectors are referred to herein as "expression vectors."

[0211] As used herein, the term "immune checkpoint inhibitor" refers to a therapeutic agent that targets at least one immune checkpoint protein to alter the regulation of an immune response, e.g., a therapeutic agent that downmodulates, inhibits, upmodulates, or activates an immune response. The term "immune checkpoint blockade" can be used to refer to a therapy that includes an immune checkpoint inhibitor. Immune checkpoint proteins are known in the art and include, but are not limited to, cytotoxic T-lymphocyte antigen 4 (CTLA-4), programmed cell death 1 (PD-1), programmed cell death-ligand 1 (PD-L1), programmed cell death-ligand 2 (PD-L2), V-domain Ig suppressor of T-cell activation (VISTA), B7-H2, B7-H3, B7-H4, B7-H6, 2B4, ICOS, HVEM, CD160, gp49B, PIR-B, KIR family receptors, TIM-1, TIM-3, TIM-4, LAG-3, BTLA, SIRPalpha (CD47), CD48, 2B4 (CD244), B7.1, B7.2, ILT-2, ILT-4, TIGIT, LAG-3, BTLA, IDO, OX40, and A2aR. In some embodiments, immune checkpoint proteins can be expressed on the surface of activated T cells. Therapeutic agents that can act as immune checkpoint inhibitors useful in the methods of the present invention include, but are not limited to, therapeutic agents that target one or more of CTLA-4, PD-1, PD-L1, PD-L2, VISTA, B7-H2, B7-H3, B7-H4, B7-H6, 2B4, ICOS, HVEM, CD160, gp49B, PIR-B, KIR family receptors, TIM-1, TIM-3, TIM-4, LAG-3, BTLA, SIRPalpha (CD47), CD48, 2B4 (CD244), B7.1, B7.2, ILT-2, ILT-4, TIGIT, LAG-3, BTLA, IDO, OX40, and A2aR. Immune checkpoint inhibitors enhance or suppress the function of one or more target immune checkpoint proteins. In some embodiments, the immune checkpoint inhibitor is a PD-L1 axis binding antagonist, such as atezolizumab.

[0212] The term "PD-1 axis binding antagonist" refers to a molecule that inhibits the interaction of a PD-1 axis binding partner with one or more of its binding partners, thereby restoring or enhancing T cell function (e.g., proliferation, cytokine production, target cell killing) to eliminate T cell dysfunction resulting from signaling on the PD-1 signaling axis. As used herein, PD-L1 axis binding antagonists include PD-L1 binding antagonists, PD-1 binding antagonists, and PD-L2 binding antagonists.

[0213] The term "PD-1 binding antagonist" refers to a molecule that reduces, blocks, inhibits, abrogates, or prevents signaling resulting from the interaction of PD-1 with one or more of its binding partners, such as PD-L1 or PD-L2. In some embodiments, a PD-1 binding antagonist is a molecule that inhibits the binding of PD-1 to one or more of its binding partners. In particular embodiments, a PD-1 binding antagonist inhibits the binding of PD-1 to PD-L1 and / or PD-L2. For example, PD-1 binding antagonists include anti-PD-1 antibodies, antigen-binding fragments thereof, immunoadhesins, fusion proteins, oligopeptides, and other molecules that reduce, block, inhibit, abrogate, or prevent signaling resulting from the interaction of PD-1 with PD-L1 and / or PD-L2. In one embodiment, the PD-1 binding antagonist reduces negative costimulatory signals mediated by or through cell surface proteins expressed on T lymphocytes via signaling through PD-1, such that dysfunctional T cells are rendered non-dysfunctional (e.g., enhance effector responses to antigen recognition). In some embodiments, the PD-1 binding antagonist is an anti-PD-1 antibody. In a particular embodiment, the PD-1 binding antagonist is MDX-1106 (nivolumab). In another particular embodiment, the PD-1 binding antagonist is MK-3475 (pembrolizumab). In another particular embodiment, the PD-1 binding antagonist is AMP-224. In another particular embodiment, the PD-1 binding antagonist is MED1-0680. In another particular embodiment, the PD-1 binding antagonist is PDR001 (spartalizumab). In another specific embodiment, the PD-1 binding antagonist is REGN2810 (cemiplimab). In another specific embodiment, the PD-1 binding antagonist is BGB-108.

[0214] The term "PD-L1 binding antagonist" refers to a molecule that reduces, blocks, inhibits, abrogates, or prevents signaling resulting from the interaction of PD-L1 with any one or more of its binding partners, such as PD-1 and B7-1. In some embodiments, a PD-L1 binding antagonist is a molecule that inhibits the binding of PD-L1 to its binding partners. In particular embodiments, a PD-L1 binding antagonist inhibits the binding of PD-L1 to PD-1 and / or B7-1. In some embodiments, PD-L1 binding antagonists include anti-PD-L1 antibodies, antigen-binding fragments thereof, immunoadhesins, fusion proteins, oligopeptides, and other molecules that reduce, block, inhibit, abrogate, or prevent signaling resulting from the interaction of PD-L1 with one or more of its binding partners, such as PD-1 and B7-1. In one aspect, the PD-L1 binding antagonist reduces negative costimulatory signals mediated by or through cell surface proteins expressed on T lymphocytes via signaling through PD-L1, such that dysfunctional T cells become non-dysfunctional (e.g., enhance effector responses to antigen recognition). In some aspects, the PD-L1 binding antagonist is an anti-PD-L1 antibody. In yet another particular aspect, the anti-PD-L1 antibody is MPDL3280A (atezolizumab, known as TECENTRIQ in WHO non-proprietary name). TM Recommended INN:List 74, Vol. 29, No. 3, 2015 (see page 387), sold under the trademark YW243.55.S70. In another specific embodiment, the anti-PD-L1 antibody is MDX-1105. In another specific embodiment, the anti-PD-L1 antibody is MSB0015718C. In yet another specific embodiment, the anti-PD-L1 antibody is MEDI4736.

[0215] The term "PD-L2 binding antagonist" refers to a molecule that reduces, blocks, inhibits, abrogates, or prevents signaling resulting from the interaction of PD-L2 with any one or more of its binding partners, such as PD-1. In some embodiments, a PD-L2 binding antagonist is a molecule that inhibits the binding of PD-L2 to one or more of its binding partners. In particular embodiments, a PD-L2 binding antagonist inhibits the binding of PD-L2 to PD-1. In some embodiments, PD-L2 antagonists include anti-PD-L2 antibodies, antigen-binding fragments thereof, immunoadhesins, fusion proteins, oligopeptides, and other molecules that reduce, block, inhibit, abrogate, or prevent signaling resulting from the interaction of PD-L2 with any one or more of its binding partners, such as PD-1. In one aspect, the PD-L2 binding antagonist reduces the negative costimulatory signal mediated by or through cell surface proteins expressed on T lymphocytes via signaling through PD-L2, such that dysfunctional T cells become non-dysfunctional (e.g., enhance the effector response to antigen recognition). In some aspects, the PD-L2 binding antagonist is an immunoadhesin.

[0216] The term "antibody" as used herein is used in the broadest sense and encompasses a variety of antibody structures, including, but not limited to, monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), and antibody fragments (e.g., bis-Fab) so long as they exhibit the desired antigen-binding activity.

[0217] An "antigen-binding fragment" or "antibody fragment" refers to a molecule other than an intact antibody that contains a portion of an intact antibody that binds to the antigen to which the intact antibody binds. Examples of antigen-binding fragments include, but are not limited to, bis-Fab; Fv; Fab; Fab, Fab'-SH; F(ab')2; diabodies; linear antibodies; single-chain antibody molecules (e.g., scFv, ScFab); and multispecific antibodies formed from antibody fragments.

[0218] A "single domain antibody" refers to an antibody fragment comprising all or part of the heavy chain variable domain or all or part of the light chain variable domain of an antibody. In certain aspects, a single domain antibody is a human single domain antibody (see, e.g., U.S. Pat. No. 6,248,516). Examples of single domain antibodies include, but are not limited to, VHHs.

[0219] A "Fab" fragment is an antigen-binding fragment produced by papain digestion of an antibody and consists of an entire light chain, the variable region domain (VH) of the heavy chain, and the first constant domain (CH1) of one heavy chain. Papain digestion of an antibody produces two identical Fab fragments. Pepsin treatment of an antibody produces a single large F(ab')2 fragment, which roughly corresponds to two disulfide-linked Fab fragments with bivalent antigen-binding activity and is still capable of cross-linking antigen. Fab' fragments differ from Fab fragments in that they contain additional residues at the carboxy terminus of the CH1 domain, including one or more cysteines from the antibody hinge region. Fab'-SH is the designation used herein for Fab' in which the cysteine ​​residues in the constant domains bear free thiol groups. F(ab')2 antibody fragments are produced as pairs of Fab' fragments with hinge cysteines between them. Other chemical couplings of antibody fragments are also known.

[0220] The term "Fc region" herein is used to define the C-terminal region of an immunoglobulin heavy chain, including native-sequence Fc regions and variant Fc regions. Although the boundaries of the Fc region of an immunoglobulin heavy chain can vary, the human IgG heavy chain Fc region is usually defined to stretch from the amino acid residue at position Cys226, or from Pro230, to the carboxyl-terminus thereof. The C-terminal lysine of the Fc region (residue 447 according to the EU numbering system) can be removed, for example, during antibody production or purification, or by recombinantly engineering the nucleic acid encoding the antibody heavy chain. Thus, an intact antibody composition can include antibody populations in which all Lys447 residues have been removed, antibody populations in which the Lys447 residue has not been removed, and antibody populations having a mixture of antibodies with and without the Lys447 residue.

[0221] An "Fv" consists of a dimer of one heavy-chain and one light-chain variable region domain in tight, non-covalent association. The folding of these two domains generates six hypervariable loops (three loops each from the H chain and L chain) that contribute antigen-binding amino acid residues and confer antigen-binding specificity to the antibody. However, even a single variable domain (or half of an Fv containing only three antigen-specific CDRs) has the ability to recognize and bind antigen, although often with lower affinity than the entire binding site.

[0222] The terms "full length antibody," "intact antibody," and "whole antibody" are used interchangeably herein and refer to an antibody having a structure substantially similar to a native antibody structure or having a heavy chain that includes an Fc region as defined herein.

[0223] "Single-chain Fv," also abbreviated as "sFv" or "scFv," is an antibody fragment comprising the VH and VL antibody domains connected in a single polypeptide chain. Preferably, the scFv polypeptide further comprises a polypeptide linker between the VH and VL domains, which enables the scFv to form the desired structure for antigen binding. For a review of scFvs, see Pluckthun, "The Pharmacology of Monoclonal Antibodies," vol. 113, Rosenburg and Moore eds., Springer-Verlag, New York, pp. 269-315 (1994); Malmborg et al., J. Immunol. Methods 183:7-13, 1995.

[0224] The term "small molecule" refers to any molecule having a molecular weight of about 2000 daltons or less, for example, about 1000 daltons or less. In some embodiments, a small molecule is a small organic molecule.

[0225] As used herein, the term "mimetic" or "molecular mimetic" refers to a polypeptide that has sufficient similarity (e.g., secondary structure, tertiary structure) in conformation and / or binding capacity to a given polypeptide or to a portion of said polypeptide to bind to a binding partner. A mimetic may bind to a binding partner with equal, lesser, or greater affinity than the polypeptide it mimics. A molecular mimetic may or may not have appreciable amino acid sequence similarity to the polypeptide it mimics. Mimetics can be naturally occurring or engineered. In some embodiments, a mimetic is a mimetic of a protein in Table 1. In other embodiments, a mimetic is a mimetic of a protein in Table 2. In yet other embodiments, a mimetic is a mimetic of another protein that binds to a protein in Table 1 or a protein in Table 2. In some embodiments, a mimetic can perform all of the functions of the mimicked polypeptide. In other embodiments, a mimetic does not perform all of the functions of the mimicked polypeptide.

[0226] As used herein, the term "conditions that allow binding" of two or more proteins (e.g., a protein in Table 1 and a protein in Table 2) to one another refers to conditions (e.g., protein concentration, temperature, pH, salt concentration) under which the two or more proteins would interact in the absence of a modulator or candidate modulator. Conditions that allow binding will vary for individual proteins and can differ between protein-protein interaction assays (e.g., surface plasmon resonance assays, biolayer interferometry, enzyme-linked immunosorbent assays (ELISAs), extracellular interaction assays, and cell surface interaction assays).

[0227] The term "survival" refers to patients who are surviving, and includes overall survival and progression-free survival.

[0228] As used herein, "recurrence-free survival" or "RFS" refers to the length of time a patient lives without signs or symptoms of cancer after primary treatment for that cancer. RFS may also be referred to as "disease-free survival" or "DFS." In some embodiments, disease-free survival (RFS) is defined as the time from randomization (e.g., assignment to an adjuvant therapy group) to disease (e.g., cancer) recurrence, new-onset disease (e.g., cancer), or death from any cause.

[0229] As used herein, "progression-free survival" (PFS) refers to the length of time during and after treatment during which the disease being treated (e.g., cancer) does not worsen. Progression-free survival can include the amount of time a patient experiences a complete or partial remission, as well as the amount of time a patient experiences stable disease.

[0230] As used herein, "overall survival" or "OS" refers to the percentage of individuals in a group who are likely to be alive after a particular period of time.

[0231] "Percent (%) amino acid sequence identity" with respect to a reference polypeptide sequence is defined as the percentage of amino acid residues in a candidate sequence that are identical to those in the reference polypeptide, after aligning the sequences and introducing gaps, if necessary, to achieve the maximum percent sequence identity, without considering any conservative substitutions as part of the sequence identity. Alignment for purposes of determining percent amino acid sequence identity can be accomplished by a variety of methods within the skill of the art, for example, using publicly available computer software such as BLAST, BLAST-2, ALIGN, or Megalign (DNASTAR) software. Those skilled in the art can determine appropriate parameters for aligning sequences, including any algorithms necessary to achieve maximum alignment over the full length of the sequences being compared. However, for purposes herein, % amino acid sequence identity values ​​are generated using the sequence comparison computer program ALIGN-2. The ALIGN-2 sequence comparison computer program was written by Genentech, Inc., and its source code, along with user documentation, has been filed with the U.S. Copyright Office (Washington, DC 20559), where it is registered under U.S. Copyright Registration No. TXU510087. The ALIGN-2 program is publicly available from Genentech, Inc. (South San Francisco, CA) or can be compiled from its source code. The ALIGN-2 program must be compiled for use on UNIX operating systems, including Digital UNIX V4.0D. All sequence comparison parameters are set by the ALIGN-2 program and do not vary.

[0232] In situations where ALIGN-2 is used for amino acid sequence comparison, the % amino acid sequence identity of a given amino acid sequence A with (or relative to) a given amino acid sequence B (alternatively, given amino acid sequence A can be said to have or contain a particular % amino acid sequence identity with (or relative to) given amino acid sequence B) is calculated as follows: 100 times the fraction X / Y where X is the number of amino acid residues scored as identical by the sequence alignment program ALIGN-2 in that program's alignment of A and B, and Y is the total number of amino acid residues in B. It will be understood that if the length of amino acid sequence A differs from the length of amino acid sequence B, the % amino acid sequence identity of A to B will differ from the % amino acid sequence identity of B to A. Unless otherwise specified, all % amino acid sequence identity values ​​used herein are obtained as described in the immediately preceding paragraph using the ALIGN-2 computer program.

[0233] As used herein, "treatment" (and grammatical variations thereof, such as "treat" or "treating") refers to a clinical intervention that attempts to alter the natural course of the individual being treated and can be performed prophylactically or during the course of clinical pathology. Desirable effects of treatment include, but are not limited to, preventing the onset or recurrence of disease, alleviating symptoms, reducing the direct or indirect pathological consequences of disease, preventing metastasis, slowing the rate of disease progression, improving or mitigating the disease state, and achieving remission or improved prognosis. In some embodiments, an agent (e.g., a modulator, a PD-L1 axis binding antagonist, or an agonist of CD177 activity) is used to delay the onset of disease or to slow the progression of disease.

[0234] A "subject" or "individual" is a mammal. Mammals include, but are not limited to, domestic animals (e.g., cows, sheep, cats, dogs, horses), primates (e.g., humans and non-human primates such as monkeys), rabbits, and rodents (e.g., mice and rats). In certain embodiments, the subject or individual is human.

[0235] As used herein, "administering" refers to a method of providing a dosage of a compound to a subject. In some embodiments, the compositions utilized in the methods herein are administered intravenously. The compositions utilized in the methods described herein can be administered, for example, intramuscularly, intravenously, intradermally, transdermally, intraarterially, intraperitoneally, intralesionally, intracranially, intraarticularly, intraprostatically, intrapleurally, intratracheally, intranasally, intravitreally, intravaginally, intrarectally, topically, intratumorally, peritoneally, subcutaneously, subconjunctivally, intravesicularly, mucosally, intrapericardially, intraumbilically, intraocularly, orally, topically, locally, by inhalation, by injection, by infusion, by continuous infusion, by a local perfusion bath target cell directly, by catheter, by perfusion, in a cream, or in a lipid composition. The administration method can vary depending on various factors (e.g., the compound or composition to be administered and the severity of the condition, disease, or disorder to be treated).

[0236] As used herein, the term "sample" refers to a composition obtained or derived from a subject and / or individual of interest that contains cellular and / or other molecular entities to be characterized and / or identified, for example, based on physical, biochemical, chemical, and / or physiological properties. For example, the phrase "disease sample" and variations thereof refer to any sample obtained from a subject of interest that is expected to contain or is known to contain the cellular and / or molecular entities to be characterized. Samples include, but are not limited to, tissue samples, primary or cultured cells or cell lines, cell supernatants, cell lysates, platelets, serum, plasma, vitreous fluid, lymphatic fluid, synovial fluid, follicular fluid, semen, amniotic fluid, milk, whole blood, plasma, serum, blood-derived cells, urine, cerebrospinal fluid, saliva, buccal swabs, sputum, tears, sweat, mucus, tumor lysates, and tissue culture media, tissue extracts such as homogenized tissue, tumor tissue, cell extracts, and combinations thereof. Samples may be archival samples, fresh samples, or frozen samples. In some embodiments, the sample is a formalin-fixed, paraffin-embedded (FFPE) tumor tissue sample.

[0237] II. Methods for identifying protein-protein interactions A. Protein-Protein Interaction Assays Interactions between two proteins are commonly tested using yeast two-hybrid (Y2H) assays and / or biochemical purification assays (e.g., affinity purification-mass spectrometry (AP / MS)); however, these methods are not well suited to testing interactions between cell surface proteins. Cell surface proteins are often not soluble in the cytoplasm and / or nucleoplasm (e.g., due to hydrophobic transmembrane regions) and are therefore unsuitable for Y2H assays. Furthermore, interactions between cell surface proteins are often low affinity and / or very transient, e.g., with half-lives of less than 1 second, and therefore are not compatible with assays involving lengthy purification protocols (e.g., AP / MS) (Bushell et al., Genome Res, 18: 622-630, 2008). The protein-protein interaction assays described herein, e.g., extracellular interaction assays and cell surface interaction assays, are compatible with cell surface proteins and can therefore identify novel interactions between these proteins.

[0238] i. Extracellular interaction assay In some embodiments of the invention, the protein-protein interaction assay is an extracellular interaction assay, e.g., avidity-based extracellular interaction screening (AVEXIS) (Bushell et al., Genome Res, 18: 622-630, 2008; Martinez-Martin et al., J Immunol Res, 2197615, 2017). In this type of assay, one or more prey proteins (e.g., one or more STM receptors) and one or more bait proteins (e.g., one or more IgSF proteins) are expressed as soluble extracellular domain (ECD) fusion proteins and assayed for interaction (e.g., using a colorimetric assay) in conditioned medium, as described below.

[0239] ia. Extracellular interaction assay of prey proteins In some embodiments, the one or more prey proteins comprise one or more fusion proteins (e.g., a library of prey fusion proteins) in which the extracellular domain (ECD) of a prey protein of interest (e.g., an STM protein) is conjugated (e.g., fused) to one or more additional moieties (e.g., an IgG or Fc tag, e.g., a human Fc tag) such that the prey fusion protein is soluble. The ECD can be identified as described in Section 2B(i).

[0240] ib. Extracellular interaction assay of bait proteins In some embodiments, the bait protein (query protein) comprises one or more fusion proteins (e.g., a library of bait fusion proteins) in which the ECD of the bait protein of interest is conjugated (e.g., fused) to one or more additional moieties such that the bait fusion protein is soluble. The additional moiety may also increase the binding activity of the bait fusion protein for the prey protein, for example, by multimerizing the bait protein ECD. Increased binding activity may increase the detection of low-affinity interactions. In some embodiments, the additional moiety or moieties cause pentamerization of the bait protein ECD. In some embodiments, the additional moiety is the pentamerization domain of rat cartilage oligomer maintenance protein (COMP). The ECD can be identified as described in Section 2B(i).

[0241] The bait protein can also be conjugated to a moiety that allows for detection of the bait protein, e.g., beta-lactamase (β-lactamase) protein. β-lactamase hydrolyzes the substrate nitrocefin, resulting in a color change from yellow to red, which can be detected colorimetrically (e.g., measuring absorbance at 485 nm).

[0242] In some embodiments, the bait fusion protein comprises both a COMP pentamerization domain and a β-lactamase protein.

[0243] ic. expression The bait and / or prey fusion proteins can be expressed (e.g., transfected, e.g., transiently transfected) in a cell. The cell can be a human cell, e.g., a HEK293 cell (e.g., Expi293F cell). In some embodiments, the bait and / or prey fusion protein is expressed in conditioned medium, e.g., conditioned medium of a transfected cell, e.g., conditioned medium of a transfected human cell. Expression in a human cell can increase the likelihood that post-translational modifications (e.g., disulfide bonds, addition of one or more glycans) occur, thus increasing the likelihood that functionally relevant interactions will be detected.

[0244] The cells can be removed (e.g., by centrifugation) from the conditioned medium after a period of growth (e.g., 7 days); the bait fusion protein and / or prey fusion protein will be present in the conditioned medium from which the cells were removed (e.g., the supernatant of the centrifugation step).

[0245] id. Interaction assay The prey fusion protein can be captured from the conditioned medium, for example, onto a substrate coated with Protein A, based on the affinity between Protein A and the Fc tag of the prey fusion protein. The substrate can be a well, for example, the well of a 384-well plate.

[0246] The bait fusion protein can be assayed directly in the conditioned medium. The concentration of the bait protein can be normalized, for example, by dilution, before performing the interaction assay.

[0247] To conduct a protein-protein interaction assay, a solution containing the bait protein can be added to one or more substrates containing the prey protein (e.g., to one or more wells of a 384-well plate). The assay can then be incubated and washed one or more times to remove unbound bait protein. In embodiments in which the bait fusion protein comprises β-lactamase, the interaction between the bait and prey fusion proteins can be detected by adding the substrate nitrocefin and measuring nitrocefin hydrolysis, e.g., by measuring absorbance at 485 nm. A relatively high absorbance indicates that the bait fusion protein has been retained, i.e., the bait and prey fusion proteins have interacted.

[0248] In some embodiments, the assay is quantitative, and the level of absorbance can be used to measure the relative strength of the interaction (eg, as in Figure 6G).

[0249] i.e. automated extracellular interaction screening The extracellular interaction assay can be a high-throughput assay, e.g., a screen. The screen can include 1 to more than 1500 prey fusion proteins (e.g., more than 1, more than 10, more than 100, more than 250, more than 500, more than 750, more than 1000, more than 1250, or more than 1500 prey fusion proteins) and 1 to more than 1500 bait fusion proteins (e.g., more than 1, more than 10, more than 100, more than 250, more than 500, more than 750, more than 1000, more than 1250, or more than 1500 bait fusion proteins). In some embodiments, the assay uses an integrated robotic system. In some embodiments, the assay is performed in one or more 384-well plates. In some embodiments, the assay is evaluated by a computational pipeline, such as a supervised classification algorithm, to determine whether an interaction has occurred.

[0250] ii. Cell surface interaction assays In some embodiments of the present invention, the protein-protein interaction assay is a cell surface interaction assay. In this type of assay, one or more prey proteins (e.g., one or more STM receptors) are expressed as extracellular domain (ECD) fusion proteins on the cell surface and tested for interaction with one or more bait proteins (e.g., IgSF proteins or PDPN) expressed as soluble extracellular domains, e.g., using a fluorescent assay in which the bait proteins contain fluorescent tags.

[0251] In some aspects, the present invention includes a method for identifying protein-protein interactions, comprising: (a) providing a collection of polypeptides, each polypeptide comprising an extracellular domain, a tag, and an anchor, wherein the collection of polypeptides comprises the extracellular domains of all or a subset of the proteins in Table 7, and optionally the polypeptides are immobilized (e.g., attached or fixed) on one or more solid surfaces, e.g., a plate or the wells of a set of plates; (b) contacting the collection of step (a) with a multimerized query protein under conditions that allow binding of the multimerized query protein to at least one of the extracellular domains of the polypeptides; and (c) detecting an interaction between the multimerized query protein and at least one extracellular domain, thereby identifying a protein-protein interaction. Each of the polypeptides can be localized (e.g., immobilized) to a distinct location (e.g., a separately interrogable location, e.g., a location separately interrogable by the methods described herein) on one or more solid surfaces. For example, each distinct location can contain one or more cells that display the polypeptide. Exemplary collections of polypeptides that can be used in the methods are described in Section IIB.

[0252] In some aspects, the invention includes methods for identifying protein-protein interactions, comprising: (a) providing a solid surface or a set of solid surfaces comprising a plurality of locations, each of the locations comprising a unique polypeptide from a collection of polypeptides, wherein each polypeptide comprises an extracellular domain, a tag, and an anchor, and wherein the collection of polypeptides comprises the extracellular domains of all or a subset of the proteins in Table 7; (b) contacting the solid surface of step (a) with a multimerized query protein under conditions that allow binding of the multimerized query protein to the extracellular domains of the polypeptides; and (c) detecting an interaction between the multimerized query protein and at least one polypeptide from the collection of polypeptides, thereby identifying a protein-protein interaction.

[0253] In some embodiments, the solid surface or set of solid surfaces collectively comprise at least 965 locations, each of the locations comprising a unique polypeptide from a collection of polypeptides, wherein each polypeptide comprises an extracellular domain, a tag, and an anchor, and wherein the collection of polypeptides comprises at least 81% extracellular domains of the proteins in Table 7.

[0254] In some embodiments, each of the locations comprises a cell, e.g., a mammalian cell, that displays a unique polypeptide. In some embodiments, the cells are transfected, e.g., transiently transfected, with a vector encoding the unique polypeptide. In some embodiments, the transient transfection is semi-automated.

[0255] In some embodiments, the multimerized query protein is a query protein described in Section IIA(iib). The multimerized query protein can be, for example, a dimerized, trimerized, tetramerized, or pentamerized query protein. In some embodiments, the multimerized query protein is a tetramerized query protein. The multimerized query protein can comprise an isolated extracellular domain of a query protein. For example, the isolated extracellular domain can be biotinylated and conjugated to fluorescent streptavidin to tetramerize the query protein.

[0256] In some embodiments, protein-protein interactions are identified as described in Section IIA(iid).

[0257] iia. Cell surface interaction assay of prey proteins In some embodiments, the one or more prey proteins comprise one or more fusion proteins (e.g., a library of prey fusion proteins) in which the extracellular domain (ECD) of a prey protein of interest (e.g., an STM protein) is conjugated (e.g., fused) to one or more additional moieties (e.g., a glycosylphosphatidylinositol (GPI)-gD (gDGPI) tag so that the prey fusion protein is expressed on the cell surface) such that the prey fusion protein is soluble. The ECD can be identified as described in Section 2B(i).

[0258] In some embodiments, where the polypeptide comprises an extracellular domain, a tag, and an anchor, the anchor can tether the extracellular domain to the surface of the plasma membrane of a cell. In some embodiments, the anchor is a glycosylphosphatidylinositol (GPI) polypeptide. In some embodiments, the anchor is a moiety used in protein lipidation, such as a moiety used in cysteine ​​palmitoylation, glycine myristoylation, lysine fatty acylation, cholesterol esterification, cysteine ​​prenylation, or serine fatty acylation.

[0259] In some embodiments, the tag can be directly or indirectly visualized or otherwise detected. For example, the tag can include a moiety that can be detected using an antibody or antibody fragment, such as a glycoprotein D (gD) polypeptide. In some embodiments, the tag includes a fluorescent protein.

[0260] iib. Cell surface interaction assay of bait protein The bait protein (query protein) may comprise one or more fusion proteins (e.g., a library of bait fusion proteins) in which the ECD of the bait protein of interest is conjugated to one or more additional moieties such that the bait fusion protein is soluble. The additional moiety or moieties may also increase the binding activity of the bait fusion protein for the prey protein, for example, by multimerizing the bait protein ECD. Increased binding activity may increase the detection of low-affinity interactions. In some embodiments, the additional moiety causes tetramerization of the bait protein ECD.

[0261] In some embodiments, the bait fusion protein comprises an Avi tag, a cleavage sequence (e.g., a TEV cleavage sequence), and an Fc tag, such that the Fc tag can be cleaved from the protein by addition of the enzyme TEV protease. To prepare the protein for a cell surface interaction assay, the Fc tag is cleaved, the Avi tag is biotinylated, and the biotinylated bait fusion protein is conjugated to fluorescent streptavidin (SA), such as streptavidin conjugated to allophycocyanin (APC), to form a tetrameric bait fusion protein that can be detected in a fluorescent assay.

[0262] iic.expression The prey fusion protein can be expressed in cells (e.g., transfected, e.g., transiently transfected). The cells can be human cells, e.g., COS7 cells. The transfected cells can be placed in wells, e.g., in a 384-well plate.

[0263] The bait fusion protein can be expressed (e.g., transfected, e.g., transiently transfected) in a cell, e.g., a mammalian cell. The bait fusion protein can be purified using standard protocols, e.g., as described in Ramani et al., Anal Biochem, 420: 127-138, 2012.

[0264] iid. Interaction assay To conduct a protein-protein interaction assay, a solution containing a bait protein (e.g., a purified bait fusion protein conjugated to a fluorescent SA) can be added to one or more wells (e.g., one or more wells of a 384-well plate) containing cells expressing the prey protein. The assay can then be incubated and washed one or more times to remove unbound bait protein. Cells can then be fixed, for example, with 4% paraformaldehyde, to preserve the protein-protein interaction.

[0265] In some embodiments, detecting the interaction comprises detecting a signal, e.g., a fluorescent signal, above a threshold level at a location on the solid surface (e.g., a signal indicative of the presence of the query protein at that location, e.g., a signal from a moiety comprised by the bait fusion protein (e.g., a multimerized query protein)). The signal may be directly or indirectly visualized or otherwise detectable. In some embodiments, the detection is semi-automated or automated. The interaction may be a transient interaction and / or a low affinity interaction, e.g., a micromolar affinity interaction.

[0266] In embodiments where the bait fusion protein (e.g., a multimerized query protein) contains a fluorescent SA, the interaction between the bait and prey fusion proteins can be detected by fluorescence microscopy. Relatively high fluorescence indicates the presence of the bait fusion protein, i.e., the bait and prey fusion proteins interact.

[0267] i.e. Automated cell surface interaction screening The extracellular interaction assay can be a high-throughput assay, e.g., a screen. The screen can include 1 to more than 1500 prey fusion proteins (e.g., more than 1, more than 10, more than 100, more than 250, more than 500, more than 750, more than 1000, more than 1250, or more than 1500 prey fusion proteins) and 1 to more than 1500 bait fusion proteins (e.g., more than 1, more than 10, more than 100, more than 250, more than 500, more than 750, more than 1000, more than 1250, or more than 1500 bait fusion proteins). In some embodiments, the assay uses an integrated robotic system. In some embodiments, the assay is performed in one or more 384-well plates. In some embodiments, the assay is evaluated by a computational pipeline, such as a custom analysis module, to determine whether an interaction has occurred.

[0268] iii. Surface plasmon resonance (SPR) method Protein-protein interactions can also be assayed using surface plasmon resonance (SPR) assays. In some embodiments, SPR assays are used to confirm or validate extracellular or cell surface interaction assays, such as assays detected in high-throughput extracellular or cell surface interaction screens.

[0269] In some embodiments, the prey protein is expressed as a fusion protein containing an additional moiety, e.g., the ECD of the protein linked to an Fc tag. The prey fusion protein can be purified. The prey protein can be immobilized on a sensor chip, e.g., a GLC or CM5 sensor chip, by amine coupling.

[0270] The bait protein can be provided in a soluble form, for example, an ECD fused to a soluble tag. The bait fusion protein can be purified.

[0271] B. Protein, Vector, and Cell Libraries Proteins assayed in the present invention include cell surface proteins, e.g., STM receptors and IgSF proteins. Proteins can be full-length proteins (e.g., secreted proteins), one or more domains or regions of a full-length protein (e.g., the extracellular domain), or fusion proteins containing one or more domains of a protein of interest and one or more additional polypeptide sequences. In some embodiments, the protein is a fusion protein having the extracellular domain of a protein of interest, e.g., an STM receptor or IgSF protein, and one or more additional polypeptide sequences, e.g., additional polypeptide sequences that enable the protein to be used in an assay. Proteins can be organized into "libraries," i.e., collections of proteins of a particular class (i.e., STM receptors or IgSF proteins) with a shared format, construction, or modification (e.g., fusion proteins containing the ECD of a protein of interest and the same or similar additional polypeptide sequences). Specific examples of libraries are described below.

[0272] i. extracellular domain In some embodiments, the protein is expressed as one or more domains of the full-length protein, e.g., the extracellular domain (ECD). The ECD is a domain of a protein that is predicted to be localized on the outside of the plasma membrane of a cell. This domain of the protein is therefore available to interact with the extracellular environment, e.g., interact with soluble proteins and the ECDs of other proteins on the cell or adjacent cells. In some embodiments, the ECD is soluble.

[0273] One or more ECDs of a protein can be identified by bioinformatics analysis, e.g., analysis of UniProt annotations. For example, the boundaries of an ECD can be identified relative to the boundaries of adjacent predicted transmembrane regions, e.g., transmembrane helices. In some embodiments, the presence of an extracellular domain can be predicted by the presence of a domain, sequence, or motif that indicates that the protein is transported to the plasma membrane, e.g., a signal sequence or a glycosylphosphatidylinositol (GPI) attachment site.

[0274] In some embodiments, the extracellular domain is expressed in the context of a full-length protein. In other embodiments, the extracellular domain is expressed as an isolated extracellular domain, e.g., a sequence of amino acid residues that includes only amino acid residues of the protein that are predicted to be extracellular. In some embodiments, the isolated extracellular domain is expressed as a fusion protein.

[0275] ia.ECD fusion protein In some embodiments, the isolated ECD is included in a fusion protein, e.g., expressed as part of a polypeptide chain that includes one or more other polypeptide sequences. The isolated ECD can be fused directly or indirectly to one or more moieties that impart or enhance one or more desirable properties, such as solubility, ease of expression, ease of capture, multimerization, etc. In some embodiments, the ECD fusion protein can be for use in an assay, e.g., an extracellular interaction assay, a cell surface interaction assay, or a surface plasmon resonance assay.

[0276] In some embodiments, the ECD fusion protein is monomeric. In other embodiments, the ECD fusion protein is multimeric, e.g., tetrameric or pentameric. In some embodiments, the ECD is fused to human IgG. In some embodiments, the ECD is fused to a human Fc tag. In some embodiments, the ECD is fused to an Avi tag. In some embodiments, the ECD is fused to a polyhistidine (His) tag. In some embodiments, the ECD is fused to a (GPI)-gD (gDGPI) tag. In other embodiments, the ECD is fused to the pentamerization domain of rat cartilage oligomeric matrix protein (COMP) and β-lactamase protein, as described, for example, in Bushell et al., Genome Res, 18: 622-630, 2008.

[0277] In some embodiments, the ECD fusion protein further comprises a cleavage sequence, e.g., a TEV cleavage sequence, to allow for selective removal of one or more domains. In some examples, an ECD fusion protein having an Avi tag and an Fc tag cleavable with the cleavage sequence is further processed to remove the Fc tag, biotinylate the Avi tag, and fuse the biotinylated ECD fusion protein to fluorescent streptavidin (SA), e.g., to form a tetramerized ECD fusion protein. In some examples, the isolated ECD or ECD fusion protein is purified.

[0278] In some aspects, a fusion protein comprises the complete sequence of a protein (e.g., the complete sequence of a secreted protein, e.g., a secreted IgSF protein) fused to one or more of the other polypeptide sequences described herein.

[0279] ii.STM receptor Single-spanning membrane (STM) receptor proteins are a large category of membrane-bound receptors with a single domain that spans the plasma membrane. Many STM receptors are expressed on the cell surface and may therefore be involved in the extracellular interactome. Exemplary STM receptors are provided in Tables 5, 7, and 8, as well as Martinez-Martin et al., Cell, 174(5): 1158-1171, 2018 and Clark et al., Genome Res, 13: 2265-2270, 2003.

[0280] The STM receptors included in the STM library were identified by bioinformatics analysis, e.g., by using algorithms to predict protein features, such as protein domains and subcellular localization. Exemplary algorithms for predicting protein domains and / or subcellular localization include the TMHMM and SignalP servers (University of Denmark) and Phobius (Stockholm Bioinformatics Center).

[0281] In some embodiments, the STM receptor has the UniProt annotation "leucine-rich", "cysteine-rich", "ITIM / ITAM" (immunoreceptor tyrosine-based inhibitory motif / immunoreceptor tyrosine-based activation motif), "TNFR" (tumor necrosis factor receptor), "TLR / ILR" (Toll-like receptor / interleukin receptor), "semaphorin", "kinase-like", "Ig-like" (immunoglobulin-like), "fibronectin", "ephrin", "EGF", "cytokine R", or "cadherin".

[0282] iia. STM receptor library for extracellular interaction assays In some embodiments, the invention features an STM receptor library for use in extracellular interaction assays. The proteins included in this library are listed in Table 5 and are "prey" fusion proteins constructed as described in Section IIA(ia) herein.

[0283] iib. STM receptor library for cell surface interaction assays In some embodiments, the invention features an STM receptor library for use in cell surface interaction assays. The proteins included in this library are listed in Table 7 and are "prey" fusion proteins constructed as described in Section IIA(iia) herein.

[0284] Polypeptide Library In some aspects, the invention features a collection of polypeptides (e.g., a polypeptide library), where each polypeptide comprises an extracellular domain, a tag, and an anchor, and the collection of polypeptides comprises the extracellular domains of all or a subset of the proteins in Table 7. In some aspects, the anchor is at the N-terminus of the polypeptide and the extracellular domain is at the C-terminus of the polypeptide. In other aspects, the anchor is at the C-terminus of the polypeptide and the extracellular domain is at the N-terminus of the polypeptide. In some aspects, the collection of polypeptides comprises the extracellular domains of at least 81% of the proteins in Table 7.

[0285] In some embodiments, the collection of polypeptides comprises at least 8%, at least 9%, at least 10%, at least 11%, at least 12%, at least 13%, at least 14%, at least 15%, at least 16%, at least 17%, at least 18%, at least 19%, at least 20%, at least 21%, at least 22%, at least 23%, at least 24%, at least 25%, at least 26%, at least 27%, at least 28%, at least 29%, at least 30%, at least 31%, at least 32%, at least 33%, at least 34%, at least 35%, at least 36%, at least 37%, at least 38%, at least 39%, at least 40%, at least 41%, at least 42%, at least 43%, at least 44%, at least 45%, at least 46%, at least 47%, at least 48%, at least 49%, at least 50%, at least 51%, at least 52%, at least 53%, at least 54%, at least 55%, at least 56%, at least 57%, at least 58%, at least 59%, at least 60%, at least 61%, at least 62%, at least 63%, at least 64%, at least 65%, at least 66%, at least 67%, at least 68%, at least 69%, at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least At least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or 100% of the extracellular domain.

[0286] In some embodiments, the collection of polypeptides comprises at least 81% to 100% of the extracellular domains of the proteins in Table 7, e.g., at least 85%, 90%, 95%, or 100% (e.g., including all) of the proteins in Table 7, e.g., 81%-85%, 83%-87%, 85%-89%, 87%-91%, 89%-93%, 91%-95%, 93%-97%, 95%-99%, or 100% of the extracellular domains of the proteins in Table 7.

[0287] In some embodiments, the collection of polypeptides comprises at least 80% to 81% extracellular domains of the proteins in Table 7, e.g., at least 80.1%, 80.2%, 80.3%, 80.4%, 80.5%, 80.6%, 80.7%, 80.75%, 80.8%, or 80.9% of the proteins in Table 7.

[0288] In some embodiments, the collection of polypeptides comprises at least 100, at least 150, at least 200, at least 250, at least 300, at least 350, at least 400, at least 450, at least 500, at least 550, at least 600, at least 650, at least 700, at least 750, at least 800, at least 850, at least 900, at least 950, at least 1000, at least 1050, at least 1100, at least 1200, at least 1300, at least 1400, at least 1500, at least 1600, at least 1700, at least 1800, at least 1900, at least 2000, at least 2100, at least 2200, at least 2300, at least 2400, at least 2500, at least 2600, at least 2700, at least 2800, at least 2900, at least 3000, at least 3100, at least 3200, at least 3300, at least 3400, at least 3500, at least 3600, at least 3700, at least 3800, at least 3900, at least 4000, at least 4100, at least 4200, at least 4300, at least 4400, at least 4500, at least 5000, at least 5500, at least 6000, at least 6500, at least 7000, at least 7500, at least 8000, at least 8500, at least 9000, at least 9500 The polypeptides may contain all 1,150 or 1,195 extracellular domains, for example, all 100 to 150, 150 to 200, 200 to 250, 250 to 300, 300 to 350, 350 to 400, 400 to 450, 450 to 500, 500 to 550, 550 to 600, 600 to 650, 650 to 700, 750 to 800, 800 to 850, 850 to 900, 900 to 950, 950 to 1,000, 1,000 to 1,050, 1,050 to 1,100, 1,100 to 1,150, or all 1,195 extracellular domains of the polypeptides in Table 7.

[0289] In some embodiments, the collection of polypeptides comprises the extracellular domains of at least 965 to at least 970 of the proteins in Table 7, e.g., at least 965, 966, 967, 968, 969, or 970 of the proteins in Table 7.

[0290] In some embodiments, the collection of polypeptides comprises at least one extracellular domain of a protein of Table 17, e.g., at least 2, at least 5, at least 10, at least 20, at least 30, at least 40, at least 50, at least 60, at least 70, at least 80, at least 90, at least 100, at least 110, at least 120, at least 130, at least 140, at least 150, at least 160, at least 170, at least 180, at least 190, at least 200, at least 210, at least 220, at least 230, or all 231 extracellular domains of the proteins of Table 17, e.g., 1-5, 5-10, 10-15, 11-20, 12-23, 13-24, 14-25, 15-26, 16-27, 17-28, 18-29, 19-30, 19-31, 20-32, 21-33, 22-34, 23-35, 24-36, 25-37, 26-38, 27-39, 28-40, 29-41, 29-42, 29-43, 29-44, 29-45, 30-31, 30-32, 30-33, 30-34, 30-35, 30-36, 30-37, 30-38, 30-41, 30-42, 30-43, 30-44, 30-45, 30-46, 30-47, 30- ~15, 15~20, 20~25, 25~30, 30~35, 35~40, 40~45, 45~50, 50~55, 55~60, 60~65, 65~70, 70~75, 75~80, 80~85, 85~90, 90~95, 95~100, 101~105, 105~110, 110~115, 115~120, 120~125, 125~130, 130~135, 135~ 140, 140-145, 145-150, 150-155, 155-160, 160-165, 165-170, 170-175, 175-180, 180-185, 185-190, 190-195, 195-200, 200-205, 205-210, 210-215, 214-220, 220-225, 225-230, or all 231 extracellular domains.

[0291] In some embodiments, the extracellular domain of a prey protein (e.g., an STM protein) has a native conformation, e.g., a conformation observed in the wild-type protein. In some embodiments, the extracellular domain of a prey protein (e.g., an STM protein) includes a native post-translational modification.

[0292] In some embodiments, the cells are COS7 cells.

[0293] In some embodiments, the cells are transiently transfected with a plasmid encoding the polypeptide.

[0294] Plasmid library In some aspects, the invention includes a collection of vectors (e.g., plasmids) encoding polypeptides each comprising an extracellular domain, a tag, and an anchor, wherein the collection of polypeptides encoded by the vectors comprises the extracellular domains of all or a subset of the proteins in Table 7, e.g., at least 8%, at least 9%, at least 10%, at least 11%, at least 12%, at least 13%, at least 14%, at least 15%, at least 16%, at least 17%, at least a few of the proteins in Table 7. at least 18%, at least 19%, at least 20%, at least 21%, at least 22%, at least 23%, at least 24%, at least 25%, at least 26%, at least 27%, at least 28%, at least 29%, at least 30%, at least 31%, at least 32%, at least 33%, at least 34%, at least 35%, at least 36%, at least 37%, at least 38%, at least 39%, at least 40%, at least 41%, at least 42%, at least 43%, at least 44%, at least Also 45%, at least 46%, at least 47%, at least 48%, at least 49%, at least 50%, at least 51%, at least 52%, at least 53%, at least 54%, at least 55%, at least 56%, at least 57%, at least 58%, at least 59%, at least 60%, at least 61%, at least 62%, at least 63%, at least 64%, at least 65%, at least 66%, at least 67%, at least 68%, at least 69%, at least 70%, at least 71%, at least 72%, 2%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%,or 100%. Exemplary collections of polypeptides that can be encoded by the plasmids are described in Section IIB.

[0295] In some embodiments, the collection of polypeptides encoded by the vectors comprises at least 81% of the proteins in Table 7, e.g., comprises the extracellular domains of at least 81% to 100% of the proteins in Table 7, e.g., comprises at least 85%, 90%, 95%, or 100% (e.g., including all) of the proteins in Table 7, e.g., comprises between 81% and 85%, 83% and 87%, 85% and 89%, 87% and 91%, 89% and 93%, 91% and 95%, 93% and 97%, 95% and 99%, or 100% of the extracellular domains of the proteins in Table 7.

[0296] In some embodiments, the collection of polypeptides encoded by the vectors comprises the extracellular domains of at least 100, at least 150, at least 200, at least 250, at least 300, at least 350, at least 400, at least 450, at least 500, at least 550, at least 600, at least 650, at least 700, at least 750, at least 800, at least 850, at least 900, at least 950, at least 1000, at least 1050, at least 1100, at least 1150, or all 1195 of the proteins in Table 7.

[0297] In some embodiments, the collection of polypeptides encoded by the vector comprises at least 965 of the proteins in Table 7. In some embodiments, the collection of polypeptides encoded by the vector comprises the extracellular domains of at least 965 to at least 970 of the proteins in Table 7, e.g., at least 965, 966, 967, 968, 969, or 970 of the proteins in Table 7.

[0298] In some embodiments, the collection of polypeptides encoded by the vector comprises the extracellular domain of at least one protein of Table 17, e.g., the extracellular domains of at least 2, at least 5, at least 10, at least 20, at least 30, at least 40, at least 50, at least 60, at least 70, at least 80, at least 90, at least 100, at least 110, at least 120, at least 130, at least 140, at least 150, at least 160, at least 170, at least 180, at least 190, at least 200, at least 210, at least 220, at least 230, or all 231 of the proteins of Table 17, e.g., 1-5, 5-10, 10-15, 15-20, 20-25, 25-30, 30-35, 35-40, 40-45, 45-50, 50-55, 55-60, 60-65, 65-70, 70-75, 75-80, 80-85, 85-90, 90-95, 95-100, 101-105, 105-110, 110-115, 115-120, 120-125, 125-130, 130-135, 1 It includes 35 to 140, 140 to 145, 145 to 150, 150 to 155, 155 to 160, 160 to 165, 165 to 170, 170 to 175, 175 to 180, 180 to 185, 185 to 190, 190 to 195, 195 to 200, 200 to 205, 205 to 210, 210 to 215, 214 to 220, 220 to 225, 225 to 230, or all 231 extracellular domains.

[0299] Cell Library In some aspects, the invention includes a collection of cells transformed with the above-described vectors (e.g., plasmids), i.e., a collection of cells transformed with vectors each encoding a polypeptide comprising an extracellular domain, a tag, and an anchor, wherein the collection of polypeptides encoded by the vectors comprises the extracellular domains of all or a subset of the proteins in Table 7, e.g., at least 8%, at least 9%, at least 10%, at least 11%, at least 12%, at least 13%, at least 14%, at least 15%, at least 16%, at least 17%, at least 18%, at least 19%, at least 20%, at least 21%, at least 22%, at least 23%, at least 24%, at least 25%, at least 26%, at least 27%, at least 28%, at least 29%, at least 30%, at least 31%, at least 32%, at least 33%, at least 34%, at least 35%, at least 36%, at least 37%, at least 38%, at least 39%, at least 40%, at least 41%, at least 42%, at least 43%, at least 44%, at least 45%, at least 46%, at least 47%, at least 48%, at least 49%, at least 50%, at least 51%, at least 52%, at least 53%, at least 54%, at least 55%, at least 56%, at least 57%, at least 58%, at least 59%, at least 60%, at least 61%, at least 62%, at least 63%, at least 64%, at least 65%, at least 66%, at least 67%, at least 68%, at least 69%, at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, 1%, at least 42%, at least 43%, at least 44%, at least 45%, at least 46%, at least 47%, at least 48%, at least 49%, at least 50%, at least 51%, at least 52%, at least 53%, at least 54%, at least 55%, at least 56%, at least 57%, at least 58%, at least 59%, at least 60%, at least 61%, at least 62%, at least 63%, at least 64%, at least 65%, at least 66%, at least 67%, at least 68%, at least 69%, at least 70%, at least 71%, at least 72%, at least 73%, at least 74%, at least 75%, at least 76%, at least 77%, at least 78%, at least 79%, at least 80%, at least 81%, at least 82%, at least 83%, at least 84%, at least 85%, at least 86%, at least 87%, at least 88%, at least 89%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%,At least 97%, at least 98%, at least 99%, or 100% of the extracellular domain. Exemplary collections of polypeptides that may be comprised by the cell are described in Section IIB.

[0300] In some embodiments, the collection of vectors contained in the cells encodes the extracellular domains of at least 81% of the proteins in Table 7, e.g., includes at least 81% to 100% of the extracellular domains of the proteins in Table 7, e.g., includes at least 85%, 90%, 95%, or 100% (e.g., including all) of the proteins in Table 7, e.g., includes between 81% and 85%, 83% and 87%, 85% and 89%, 87% and 91%, 89% and 93%, 91% and 95%, 93% and 97%, 95% and 99%, or 100% of the extracellular domains of the proteins in Table 7.

[0301] In some embodiments, the collection of vectors contained in the cells encodes the extracellular domains of at least 100, at least 150, at least 200, at least 250, at least 300, at least 350, at least 400, at least 450, at least 500, at least 550, at least 600, at least 650, at least 700, at least 750, at least 800, at least 850, at least 900, at least 950, at least 1000, at least 1050, at least 1100, at least 1150, or all 1195 of the proteins in Table 7.

[0302] In some embodiments, the collection of vectors contained in the cells encodes at least 965 of the proteins in Table 7.

[0303] In some embodiments, the collection of vectors contained in the cells comprises at least 965 to at least 970 extracellular domains of the proteins in Table 7, e.g., at least 965, 966, 967, 968, 969, or 970 extracellular domains of the proteins in Table 7.

[0304] In some embodiments, the collection of vectors contained in the cells comprises the extracellular domain of at least one protein of Table 17, e.g., at least 2, at least 5, at least 10, at least 20, at least 30, at least 40, at least 50, at least 60, at least 70, at least 80, at least 90, at least 100, at least 110, at least 120, at least 130, at least 140, at least 150, at least 160, at least 170, at least 180, at least 190, at least 200, at least 210, at least 220, at least 230, or all 231 of the proteins of Table 17, e.g., 1-5, 5-10 of the polypeptides of Table 17. , 10-15, 15-20, 20-25, 25-30, 30-35, 35-40, 40-45, 45-50, 50-55, 55-60, 60-65, 65-70, 70-75, 75-80, 80-85, 85-90, 90-95, 95-100, 101-105, 105-110, 110-115, 115-120, 120-125, 125-130, 130-135, 135 The extracellular domains may include up to 140, 140-145, 145-150, 150-155, 155-160, 160-165, 165-170, 170-175, 175-180, 180-185, 185-190, 190-195, 195-200, 200-205, 205-210, 210-215, 214-220, 220-225, 225-230, or all 231.

[0305] In some embodiments, each cell of the collection of cells is capable of expressing a polypeptide encoded by a vector with which it has been transformed. In some embodiments, each of a plurality of cells of the collection of cells is capable of expressing a polypeptide encoded by a vector with which it has been transformed.

[0306] iic.PDPN Podoplanin (PDPN) is an STM receptor. PDPN can be highly expressed on the surface of fibroblasts (such as cancer-associated fibroblasts), lymphatic endothelial cells, and type I pneumocytes. PDPN is overexpressed in many tumor tissues, and its expression in tumor tissues is associated with poor prognosis. PDPN is known to function as a master regulator of actomyosin contractility in mouse fibroblasts through interaction with the C-type lectin receptor CLEC-2 (CLEC1B) (Astarita et al., Nat Immunol, 16: 75-84, 2015; Acton et al., Nature, 514: 498-502, 2014). In some embodiments, PDPN functions as a regulator of actomyosin contractility in human CAFs.

[0307] PDPN for cell surface interaction assays In some aspects, the invention features a PDPN fusion protein for use in a cell surface interaction assay, which is a "bait" fusion protein constructed as described in Section IIA(iib) herein.

[0308] iii.IgSF protein The immunoglobulin superfamily (IgSF) is the largest family of secreted and cell-surface-expressed proteins encoded by the human genome and is the most highly represented extracellular protein domain in humans. IgSF proteins are known to function through the formation of homophilic and heterophilic complexes that mediate a wide range of functionalities, such as regulating axon guidance, synaptic plasticity, cell migration, cell adhesion, and self- and non-self-recognition. As such, these proteins have become a major focus of drug development efforts. Some IgSF proteins are expressed on the cell surface (e.g., transmembrane proteins); others are secreted. Exemplary IgSF proteins are provided in Table 4 and in Ozkan et al., Cell, 154(1): 228-239, 2013.

[0309] Immunoglobulin superfamily (IgSF) libraries include proteins that contain at least one immunoglobulin (Ig) domain or immunoglobulin fold, for example, proteins that are annotated in the SwissProt database as "immunoglobulin-like superfamily" or are otherwise shown to have structural or functional similarity to such proteins. In some embodiments, IgSF proteins are identified by the protein annotation "immunoglobulin-like domain superfamily" in the SwissProt database. In some embodiments, IgSF proteins are identified based on the protein's involvement in a primary biological activity.

[0310] In some embodiments, the IgSF protein has the UniProt annotation "TFNR" (transcription factor-like nuclear regulator), "TLR / ILR (Toll-like receptor / interleukin receptor)", "semaphorin", "kinase-like", "IgSF / Ig-like fold", "Ig-like fold", "fibronectin", "ephrin", "EGF", "cytokine R", or "cadherin".

[0311] In some embodiments, the IgSF superfamily protein is selected from the group consisting of programmed death-ligand 1 (PD-L1; CD274), programmed cell death 1 ligand 2 (PD-L2; CD274; PDCD1LG2), receptor tyrosine-protein phosphatase delta (PTPRD), receptor tyrosine-protein phosphatase S (PTPRS), receptor tyrosine-protein phosphatase F (PTPRF), neural cell adhesion molecule L1-like protein ("near homolog of L1"; CHL1), contactin 11 (CNTN1), leukocyte immunoglobulin G (LEGG), and leukocyte immunoglobulin G (LEGG). These include leukocyte immunoglobulin-like receptor subfamily B member 1 (LILRB1), leukocyte immunoglobulin-like receptor subfamily B member 2 (LILRB2), leukocyte immunoglobulin-like receptor subfamily B member 3 (LILRB3), leukocyte immunoglobulin-like receptor subfamily B member 4 (LILRB4), leukocyte immunoglobulin-like receptor subfamily B member 5 (LILRB5), MAM domain-containing glycosylphosphatidylinositol-anchored protein 1 (MDGA1), and tyrosine protein kinase receptor AXL.

[0312] iiia. IgSF library for extracellular interaction assays In some embodiments, the invention features an IgSF receptor library for use in extracellular interaction assays. The proteins included in this library are listed in Table 4 and are "bait" fusion proteins constructed as described in Section IIA(ib) herein.

[0313] iiib. IgSF proteins for cell surface interaction assays In some embodiments, the invention features IgSF receptors for use in cell surface interaction assays. The proteins included in this library are listed in Table 4 and are "bait" fusion proteins constructed as described in Section IIA(iib) herein.

[0314] iv. Bait and Prey Protein Libraries In some aspects, the invention features a bait protein library that includes a protein provided in Table 1. The proteins can be constructed as "bait" fusion proteins, as described in Section IIA(ib) or Section IIA(iib) herein.

[0315] In some aspects, the invention features a prey protein library that includes the proteins provided in Table 2. The proteins can be constructed as "prey" fusion proteins, as described in Section IIA(ia) or Section IIA(iia) herein.

[0316] C. Protein-protein interaction network In some embodiments, interactions between bait proteins (Table 1) and prey proteins (Table 2) can be tested using extracellular interaction assays (including high-throughput extracellular interaction screens), cell surface assays (including high-throughput cell surface interaction screens), and surface plasmon resonance assays as described in Section IIA. In some embodiments, an assay can include from 1 to more than 1500 prey proteins (e.g., more than 1, more than 10, more than 100, more than 250, more than 500, more than 750, more than 1000, more than 1250, or more than 1500 prey fusion proteins) and from 1 to more than 1500 bait proteins (e.g., more than 1, more than 10, more than 100, more than 250, more than 500, more than 750, more than 1000, more than 1250, or more than 1500 bait proteins). In some embodiments, the assays identify between 1 and 900 protein-protein interactions (e.g., 1, more than 1, more than 10, more than 100, more than 250, more than 500, more than 750, more than 850, or 900 protein-protein interactions). In some embodiments, these analyses identify previously unrecognized communities of functionally related proteins, reveal binding partners of orphan proteins, and reveal receptor-ligand interactions significantly deregulated in cancer.

[0317] i.STM receptor-IgSF interactome In some embodiments, the IgSF proteins provided in Table 4 and the STM proteins provided in Table 5 are tested for interactions in a high-throughput extracellular interaction screen, as described in Section IIA(ie) herein. In some embodiments, the IgSF proteins are constructed as a bait library (as in Section IIB(iiia)) and the STM proteins are constructed as a prey library (as in Section IIB(iia)). In some embodiments, the detected interactions are assembled into a network. In some embodiments, selected interactions are further tested, for example, in a cell surface interaction screen (as described in Section IIa(ii)) or an SPR assay (as described in Section IIa(ii)).

[0318] ia. Proteins that interact with PD-L1 and / or PD-L2 The IgSF proteins PD-L1 (CD274) and PD-L2 (PDCD1LG2) are immune checkpoint proteins that play important roles in immunosuppressive functions that can lead to tumor immune escape (Chen and Mellman, Immunity, 39: 1-10, 2013). Therefore, PD-L1-targeted therapies have become a major focus of research. Antibody blockade of PD-L1 is a preferred immunotherapy strategy for the treatment of solid tumors; however, many patients do not respond or do not show a durable response to antibody blockade of PD-L1, indicating the need for therapies targeting other immunosuppressive pathways.

[0319] In some embodiments, the assays described herein can identify the interaction between PD-L1 and ephrin type A receptor 3 (EPHA3). EPHA3 is a receptor tyrosine kinase activated by the binding of ephrin proteins and plays a role in signal transduction and the regulation of multiple cellular processes, such as cell proliferation, migration, and adhesion (Lisabeth et al., Cold Spring Harb Perspect Biol, 5, 2013). EPHA3 has also been identified as one of the most frequently mutated genes in certain tumors (Andretta et al., Sci Rep, 7: 41576, 2017). Thus, downstream effects of the interaction between PD-L1 and EPHA3 can include modulation of immune checkpoint function, e.g., immunosuppression, and / or modulation of EPHA3 kinase function.

[0320] In some embodiments, the assays described herein can identify the interaction between PD-L2 and CEACAM4. The CEACAM family has been shown to play a role in regulating the immune system: CEACAM1 has been identified as a ligand for the inhibitory receptor TIM-3 (Huang et al., Nature, 517: 386-390, 2002). The interaction between CEACAM4 and PD-L2 can contribute to PD-1-independent functions of PD-L2, such as those described in Liu et al., J Exp Med, 197: 1721-1730, 2003, e.g., phagocytosis. Thus, downstream effects of the interaction between PD-L2 and CEACAM4 may include modulation of immune checkpoint function, such as immunosuppression (Delgado Tascon et al., J Leukoc Biol, 97: 521-531, 2015; Xiao et al., J Exp Med, 211: 943-959, 2014).

[0321] In some aspects, one or more of the assays described herein may identify interactions between PD-L2 and CEACAM4; PD-L2 and ICAM5; PD-L2 and NECTIN3; PD-L2 and PSG9; and PD-L2 and TNFRSF11A.

[0322] ib. Proteins that interact with PTPR proteins The PTPR proteins PTPRD, PTPRS, and PTPRF are receptor tyrosine protein phosphatases. Tyrosine phosphorylation and dephosphorylation regulate numerous cellular processes, and abnormalities in tyrosine phosphorylation / dephosphorylation are associated with tumorigenesis. In particular, PTPRD and PTPRS have been described as important modulators of nervous system processes, such as synaptogenesis and axon growth, and have also been identified as tumor suppressors with high mutation rates in neuroblastoma, glioma, colon cancer, and breast cancer (Veeriah et al., Proc Natl Acad Sci USA, 106: 9435-9440, 2009; Wang et al., Hepatology, 62: 1201-1214, 2015).

[0323] In some embodiments, the assays described herein can identify interactions between PTPRS, PTPRD, and / or PTPRF and members of the SLIT and NTRK-like (SLITRK) family (such as SLITRK1, SLITRK2, SLITRK3, SLITRK4, and SLITRK6), and can identify interactions between PTPRS, PTPRF, and / or PTPRD and members of the leucine-rich repeat and fibronectin type III domain-containing protein (LRFN) family, interleukin-1 receptor accessory protein (IL1RAP), and the related proteins IL1RAPL1 and IL1RAPL2. SLITRK, LRFN, and IL1RAP-like proteins have been implicated in nervous system disorders, including tumors. For example, IL1RAP is required for IL-1 signaling, as reported to be a biomarker for chronic myeloid leukemia stem cells (Zhao et al., Int J Clin Exp Med, 7: 4787-4798, 2014). Thus, downstream effects of the identified interactions may include alterations in phosphatase activity, e.g., alterations in tyrosine phosphorylation / dephosphorylation, e.g., tyrosine phosphorylation / dephosphorylation of proteins implicated in diseases, including cancer.

[0324] In some embodiments, the one or more assays may identify interactions between PTPRD and BMP5, CEACAM3, IL1RAP, IL1RAPL2, LECT1, LRFN5, SIRPG, SLITRK3, SLITRK4, SLITRK6, and TGFA. PTPRD is associated with suppression of cell proliferation and STAT3 phosphorylation (Veeriah et al., PNAS, 106(23): 9435-9440, 2009; Peyser et al., PLoS ONE, 10.1371 / journal.pone.0135750, 2015).

[0325] In some embodiments, one or more assays may identify interactions between PTPRS and C6orf25, IL1RAP, IL1RAPL1, IL1RAPL2, LRFN1, LRFN5, LRRC4B, NCAM1, SLITRK1, SLITRK2, SLITRK3, SLITRK4, and SLITRK6. PTPRS have been associated with inhibition of cell migration (Wang et al., Hepatology, 62(4): 1201-1214, 2015) and activation of the PI3K signaling pathway (Suarez Pestana et al., Oncogene, 18: 4069-4079, 1999; Morris et al., PNAS, 108(47): 19024-19029, 2011).

[0326] In some embodiments, one or more assays may identify interactions between PTPRF and CD177, IL1RAP, and LRFN5. PTPRF is associated with inhibition of cell migration and EGFR phosphorylation (Du et al., J Cell Sci, 126: 1440-1453, 2013).

[0327] Proteins interacting with PTPRD mutants In some embodiments, the one or more assays may identify interactions between forms of PTPRD with one or more disease-associated (e.g., cancer-associated) mutations (PTPRD mutants), such as amino acid substitution or deletion mutations, and IL1RAP, IL1RAPL1, IL1RAPL2, LRFN4, LRFN5, LRRC4B, NTRK3, SLITRK1, SLITRK3, or SLITRK6. Disease-associated variations in PTPRD (Table 9) occur in the immunoglobulin (IG) domains IG1, IG2, IG3, and fibronectin (FN) domains FN1-FN8 of the PTPRD ECD (Figure 6F). Specific amino acid substitution or deletion mutations include ΔG61ΔE106 (IG1), G203EK204E (IG2), R232CR233C (IG2), P249L (IG3), G285E (IG3), E406K (FN1), S431L (FN2), R561Q (FN3), P666S (FN4), E755K (FN5), V892I (FN6), S912F (FN7), R995C (FN7), and R1088C (FN8). In some aspects, PTPRD variants are expressed as bait proteins as described in Section IIA(ib) herein and assayed for interaction with a prey protein in an extracellular interaction assay as described in Section IIA(i) herein. In some embodiments, the strength of the interaction between the PTPRD variant and the binding partner may be decreased compared to the strength of the interaction between the wild-type PTPRD protein and the binding partner, while in other embodiments, the strength of the interaction may be similar for the variant and wild-type PTPRD proteins.

[0328] ic. Proteins that interact with CHL1 and / or CNTN1 Neural cell adhesion molecule L1-like protein (CHL1) and contactin 1 (CNTN1) are IgSF proteins that function as cell adhesion molecules (CAMs). Overexpression of CAMs is associated with poor prognosis in cancer patients (Yan et al., Cancer Res., 76(6), 1603-1614, 2016).

[0329] CHL1 is involved in neuronal cell adhesion and plays a role in central nervous system (CNC) development. CHL1 is associated with suppression of cell proliferation (Tang et al., Oncogene, doi: 10.1038 / s41388-018-0648-7), tumorigenesis (Tang et al., Oncogene, doi: 10.1038 / s41388-018-0648-7), and cell invasion (He et al., Biochem Biophys Res Commun, 438: 433-438, 2013).

[0330] In some embodiments, the assays described herein may identify interactions between CHL1 and contactin 5 (CNTN5), L1 cell adhesion molecule (L1CAM), two proteins involved in cell adhesion, and B and T lymphocyte attenuator (BTLA).

[0331] Upregulation of CNTN1 is strongly associated with the invasion of prostate cancer cells. CNTN1 is associated with the suppression of cell proliferation and invasion (Yan et al., Cancer Res, 76(6): 1603-1614, 2016), and activation of the RhoA and AKT signaling pathways (Yan et al., PLoS ONE, 8(5): e65463, 2013; Chen et al., Front Mol Neurosci, 11, Article 422 (2018); Su et al., Cancer Res, 66(5): 2553-2561, 2006).

[0332] In some embodiments, the one or more assays may identify interactions between CHL1 and SIRPA, CNTN1, CNTN5, L1CAM, and TMEM132A.

[0333] In some embodiments, the one or more assays may identify interactions between CNTN1 and CDH6, CHL1, FCGRT, PCDHB7, and SGCG. Downstream effects of the identified interactions may include modulation of cell adhesion, e.g., neural cell adhesion.

[0334] Proteins that interact with id.LILRB protein Leukocyte immunoglobulin-like receptor B proteins (LILRBs) are IgSF proteins characterized by the presence of cytosolic immunoreceptor tyrosine-based activation motifs (ITAMs) or immunoreceptor tyrosine-based inhibitory motifs (ITIMs) (Brown et al., Tissue Antigens, 64: 215-225, 2004). LILRB proteins can be activating or inhibitory receptors, are primarily expressed in myeloid cells and lymphocytes, and have been associated with roles in cancer, autoimmune disease, infectious diseases, and phagocytosis by macrophages.

[0335] In some embodiments, the assays described herein may identify interactions between LILLRB5 and the low-density lipoprotein receptor (LDLR), which mediates the endocytosis of low-density lipoproteins (LDL) and is involved in diseases including familial hypercholesterolemia and coronary artery disease.

[0336] The assays described herein can also identify interactions between LILRB1 and CLEC6A, CXADR, EDAR, FLT4, IL6R, ILDR1, and LILRA5. LILRB1 is a component of the MHC class 1-LILRB1 signaling axis and has been shown to protect cells (such as tumor cells) from phagocytosis (Barkal et al., Nature Immunol, 19: 76-84, 2017). Thus, downstream effects of the identified interactions may include modulation of phagocytosis (Section IIIB(vii)). Downregulation of interactions involving LILRB1 may lead to increased phagocytosis, for example, increased phagocytosis of tumor cells. LILRB1 has also been associated with osteoclast differentiation (Mori et al., J Immunol, 181(7): 4742-4751, 2008).

[0337] The assay described herein can also identify interactions between LILLRB2 and IGSF8 and MOG. LILLRB2 is associated with M2 macrophage polarization (Chen et al., J Clin Invest, 128(12), 5647-5662).

[0338] The assays described herein can also identify interactions between LILLRB3 and LRRC15 and LY6G6F. LILLRB3 is associated with osteoclast differentiation (Mori et al., J Immunol, 181(7): 4742-4751, 2008).

[0339] The assay described herein can also identify the interaction between LILLRB4 and CNTFR. LILLRB4 is associated with osteoclast differentiation (Mori et al., J Immunol, 181(7): 4742-4751, 2008).

[0340] In some embodiments, the one or more assays may identify interactions between LILLRB5 and APLP2, CD177, CLEC10A, LDLR, PILRA, and UNC5C.

[0341] i.e., proteins that interact with MDGA1 The MAM domain containing glycosylphosphatidylinositol anchor 1 (MDGA1) is present in IgSF proteins expressed in the nervous system. In some embodiments, one or more of the assays described herein can identify interactions between MDGA1 and NLGN3 and NLGN4X.

[0342] Proteins that interact with if.AXL The tyrosine-protein kinase receptor AXL is an IgSF protein that inhibits innate immune responses. Overexpression of AXL is associated with numerous cancers, including colon cancer, esophageal cancer, thyroid cancer, breast cancer, lung cancer, liver cancer, and astrocytoma-glioblastoma (Verma et al., Mol Cancer Ther, 10(10): 1763-1773, 2011).

[0343] AXL has been shown to regulate cell invasion (Verma et al., Mol Cancer Ther, 10(10): 1763-1773, 2011); JAK / STAT pathway, RAS / RAF / MAPK / ERK1 / 2 pathway, and PI3K signaling pathway (Verma et al., Mol Cancer Ther, 10(10): 1763-1773, 2011); cell motility and morphological changes, such as filopodia formation (Verma et al., Mol Cancer Ther, 10(10): 1763-1773, 2011); and epithelial-mesenchymal transition (EMT) (Gjerdrum et al., PNAS, 107(3): 1124-1129, 2010; Divine et al., Oncotarget, 7: 77291-77305, 2016; Rankin et al., al., Cancer Res, 70(19), 7570-7579, 2010; Chichon et al., Oncogene, 33: 4185-4192, 2013).

[0344] In some aspects, one or more of the assays described herein may identify interactions between AXL and IL1RL1 and VSIG10L.

[0345] ig. LRRC4B-interacting proteins In some embodiments, one or more of the assays described herein may identify an interaction between LRRC4B and BTN3A1 or BTN3A3.

[0346] ii. Interaction between STM receptors and PDPN In some embodiments, the STM proteins provided in Table 7 can be tested for interaction with PDPN in a high-throughput cell surface interaction screen, as described in Section IIA(iie) herein. PDPN can be constructed as a bait protein (as in Section IIB(iic)), and the STM proteins can be constructed as a prey library (as in Section IIB(iib)). In some embodiments, selected interactions can be further tested in an SPR assay (as described in Section IIa(iii)).

[0347] Podoplanin (PDPN) is an STM receptor that functions as a regulator of actomyosin contractility in CAFs. In some embodiments, the assays described herein may identify interactions between PDPN and cluster of differentiation 177 (CD177), a neutrophil receptor recently identified as a marker for tumor-resident Tregs (Plitas et al., Immunity, 45: 1122-1134, 2016). In some embodiments, one or more assays may identify interactions between PDPN and CD177, CLEC-2 (CLEC1B), and SIGLEC7.

[0348] In some aspects, downstream effects of interaction with PDPN (e.g., interaction of CLEC-2 and / or CD177 with PDPN) can include increased fibroblast elongation (e.g., increased CAF elongation) and decreased fibroblast contractility (e.g., decreased CAF contractility) (Section IIIB(i)).

[0349] III. Methods for identifying modulators of protein-protein interactions A. Assays for modulation of interactions In some aspects, the invention features a method for identifying a modulator of the interaction between a protein of Table 1 and a protein of Table 2, the method comprising: (a) providing a candidate modulator (e.g., a candidate modulator described in Section IV herein); (b) contacting the protein of Table 1 with a protein of Table 2 in the presence or absence of the candidate modulator under conditions that allow binding of the protein of Table 1 to the protein of Table 2, wherein the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (c) measuring binding of the protein of Table 1 to the protein of Table 2, wherein an increase or decrease in binding in the presence of the candidate modulator compared to binding in the absence of the candidate modulator identifies the candidate modulator as a modulator of the interaction between the protein of Table 1 and the protein of Table 2.

[0350] In some embodiments, candidate modulators are provided to cells (e.g., mammalian cells), to cell culture medium, to conditioned medium, and / or to purified forms of the proteins of Table 1 and / or the proteins of Table 2. In some embodiments, candidate modulators are provided at a concentration of at least 0.1 nM, 0.5 nM, 1 nM, 10 nM, 50 nM, 100 nM, 250 nM, 500 nM, 750 nM, 1 μM, 2 μM, 3 μM, 5 μM, or 10 μM. In some embodiments, candidate modulators are provided at a concentration of 0.1 nM to 10 μM. In some embodiments, candidate modulators are provided in solution, e.g., in a soluble form.

[0351] In some embodiments, a candidate modulator is identified as a modulator if the increase in binding is at least 70%. In some embodiments, the increase in binding is at least 5%, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 100%, or greater than 100%. In some embodiments, the increase in binding is at least 70%.

[0352] In some embodiments, a candidate modulator is identified as a modulator if there is a decrease in binding of at least 70%. In some embodiments, the decrease in binding is at least 5%, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 100%. In some embodiments, the decrease in binding is at least 70%.

[0353] i. Assays for modulation of protein-protein interactions In some embodiments, binding of a protein of Table 1 and a protein of Table 2 in the presence or absence of a candidate modulator is assessed in a protein-protein interaction assay. Modulation of the interaction between a protein of Table 1 and a protein of Table 2 may be identified as an increase in protein-protein interaction in the presence of the modulator compared to the protein-protein interaction in the absence of the modulator, for example, a 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 80%, 90%, 95%, 100%, or greater than 100% increase in protein-protein interaction. Alternatively, modulation can be identified as a decrease in protein-protein interaction in the presence of a modulator compared to the protein-protein interaction in the absence of the modulator, e.g., a 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 80%, 90%, 95%, or 100% decrease in protein-protein interaction. The protein-protein interaction assay can be, for example, an SPR assay, a biolayer interferometry (BLI) assay, an enzyme-linked immunosorbent assay (ELISA), an extracellular interaction assay described in Section IIAi, or a cell surface interaction assay described in Section IIAii.

[0354] ia. SPR assays for modulation of protein-protein interactions In some embodiments, the assay for protein-protein interaction is a surface plasmon resonance (SPR) assay, as described in Section IIAii herein. In some embodiments, modulation of binding of a protein of Table 1 to a protein of Table 2 is measured as the difference in SPR signal response units (RU) in the presence of a modulator compared to its absence.

[0355] ib. BLI assays for modulation of protein-protein interactions In some embodiments, the protein-protein interaction assay is a BLI assay. In some embodiments, the BLI assay is performed using an isolated ECD, e.g., an isolated ECD, as described in Section IIB(i) herein. In some embodiments, modulation of binding of a protein of Table 1 to a protein of Table 2 is measured as a difference in wavelength shift (Δλ) measured on a biosensor chip in the presence of a modulator compared to its absence.

[0356] ic. ELISA for modulation of protein-protein interactions In some embodiments, the assay for protein-protein interaction is an ELISA. In some embodiments, the first protein is bound to a plate (e.g., directly bound to the plate or bound to the plate via an affinity tag recognized by an antibody bound to the plate), and the second protein is provided in soluble form, e.g., provided as an isolated ECD as described in section IIB(i) herein. The interaction between the first protein and the second protein can be detected by providing an antibody that binds to the second protein or its affinity tag, where the antibody can be detected, e.g., visualized, in an assay for the presence of the antibody.

[0357] Other assays for modulation of protein-protein interactions In some embodiments, the assay is an extracellular interaction assay, as described in Section IIAi herein. In some embodiments, the assay is a cell surface interaction assay, as described in Section IIAii herein. In some embodiments, the assay is an isothermal titration calorimetry (ITC) assay, an assay comprising immunoprecipitation, or an ALPHASCREEN assay. TM The assay involves techniques.

[0358] B. Assays for changes in downstream activities In another aspect, the invention features a method of identifying a modulator of a downstream activity of a protein of Table 1, the method including: (a) providing a candidate modulator (e.g., a candidate modulator described in Section IV herein); (b) contacting the protein of Table 1 with a protein of Table 2 in the presence or absence of the candidate modulator under conditions that allow binding of the protein of Table 1 to the protein of Table 2, where the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (c) measuring a downstream activity of the protein of Table 1 (e.g., CAF contractility, immune checkpoint inhibition, inhibition of cell proliferation, modulation of target phosphorylation, inhibition of cell migration, inhibition of tumorigenesis, inhibition of cell invasion, macrophage polarization, modulation of phagocytosis, osteoclast differentiation, activation of a signaling pathway, or filopodia formation), wherein a change in downstream activity in the presence of the candidate modulator compared to the downstream activity in the absence of the candidate modulator identifies the candidate modulator as a modulator of a downstream activity of the protein of Table 1.

[0359] In another aspect, the invention features a method of identifying a modulator of a downstream activity of a protein of Table 2, the method including: (a) providing a candidate modulator (e.g., a candidate modulator described in Section IV herein); (b) contacting the protein of Table 2 with a protein of Table 1 in the presence or absence of the candidate modulator under conditions that allow binding of the protein of Table 2 to the protein of Table 1, wherein the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (c) measuring a downstream activity of the protein of Table 2 (e.g., CAF contractility, immune checkpoint inhibition, inhibition of cell proliferation, modulation of target phosphorylation, inhibition of cell migration, inhibition of tumorigenesis, inhibition of cell invasion, macrophage polarization, modulation of phagocytosis, osteoclast differentiation, activation of a signaling pathway, or filopodia formation), wherein a change in downstream activity in the presence of the candidate modulator compared to the downstream activity in the absence of the candidate modulator identifies the candidate modulator as a modulator of the downstream activity of the protein of Table 2.

[0360] In some embodiments, the candidate modulator is provided at a concentration of at least 0.1 nM, 0.5 nM, 1 nM, 10 nM, 50 nM, 100 nM, 250 nM, 500 nM, 750 nM, 1 μM, 2 μM, 3 μM, 5 μM, or 10 μM. In some embodiments, the candidate modulator is provided at a concentration of 0.1 nM to 10 μM. In some embodiments, the candidate modulator is provided at a range of concentrations, e.g., as in Figure 4F. In some embodiments, the candidate modulator is provided in solution, e.g., in a soluble form. In some embodiments, the candidate modulator is provided to an organism comprising the proteins of Table 1 and the proteins of Table 2, to tissue comprising the proteins of Table 1 and the proteins of Table 2, to cells (e.g., mammalian cells), to cell culture medium, to conditioned medium, and / or to purified forms of the proteins of Table 1 and / or the proteins of Table 2.

[0361] In some embodiments, a candidate modulator is identified as a modulator if there is an increase in downstream activity of a protein of Table 1 or a protein of Table 2 by at least 30%. In some embodiments, the increase in downstream activity is at least 5%, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 100%, or greater than 100%. In some embodiments, the increase in downstream activity is at least 30%.

[0362] In some embodiments, a candidate modulator is identified as a modulator if there is at least a 30% decrease in downstream activity of a protein of Table 1 or a protein of Table 2. In some embodiments, the decrease in downstream activity is at least 5%, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 100%. In some embodiments, the decrease in downstream activity is at least 30%.

[0363] In some embodiments, the downstream activity of a protein of Table 1 or a protein of Table 2 is assessed in one or more assays, as described below.

[0364] In some embodiments, the downstream activity is an activity associated with the development or progression of a disease, such as cancer.

[0365] i.CAF contractility In some embodiments, the downstream activity is cancer-associated fibroblast (CAF) actomyosin contractility (CAF contractility). In some embodiments, the protein in Table 1 is PDPN and the downstream activity is CAF contractility. In some embodiments, the protein in Table 1 is PDPN and the protein in Table 2 is CD177 and the downstream activity is CAF contractility.

[0366] In some embodiments, the CAF contractility assay is a gel contraction assay. In this assay, a population (e.g., 100,000 cells) of fibroblasts (e.g., cancer-associated fibroblasts) containing at least one of the proteins in Table 1 and Table 2 is mixed in a Matrigel-collagen mixture and placed in a well, e.g., the well of a 96-well plate. In embodiments where an individual cell does not contain both the proteins in Table 1 and Table 2, the protein not contained in the cell can be provided to another cell (e.g., a mammalian cell, e.g., a neutrophil or T cell), or can be provided or added to the Matrigel-collagen medium or cell culture medium. The cells can be further treated with a modulator, e.g., by addition to the Matrigel-collagen medium or cell culture medium. After solidifying the gel for 20 minutes, the gel is detached from the side of the well and cell culture medium is added. The gel is incubated, e.g., for 72 hours, before imaging. The contractility of the cell population is measured by comparing the well diameter with the final gel diameter. The increase or decrease in contractility is calculated by comparing the gel contraction of cells in which the modulator is provided to the gel contraction of cells in which the modulator is not provided. In some instances, a population of cells treated with a modulator exhibits decreased gel contraction compared to a population of cells not treated with the modulator, e.g., CAF contractility is decreased in the presence of the modulator. In some embodiments, CAF contractility is decreased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured by a gel contraction assay. In some embodiments, CAF contractility is decreased by at least 30%, as measured by a gel contraction assay. In some embodiments, CAF contractility is decreased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, or at least 80%, as measured by a gel contraction assay.

[0367] In other embodiments, CAF contractility is measured using a 3D gel elongation assay, e.g., an assay such as that described in Example 8B. In this assay, fibroblasts (e.g., CAFs) contain at least one of the proteins in Table 1 and Table 2. In embodiments in which cells do not contain both the proteins in Table 1 and Table 2, the protein not contained in the cells can be provided to another cell (e.g., a mammalian cell, e.g., a neutrophil or T cell) or can be provided or added in the cell culture medium. The cells can be further treated with a modulator, e.g., by addition to the cell culture medium. Decreased fibroblast contractility is indicated by increased elongation of the cells in the 3D gel compared to an isotype control (Figures 17A-17D). In some examples, cells treated with a modulator are elongated compared to control cells, e.g., CAF contractility is reduced in the presence of the modulator. In some embodiments, CAF contractility is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured in a 3D gel elongation assay. In some embodiments, CAF contractility is reduced by at least 30% in the presence of the modulator, as measured in a 3D gel elongation assay. Morphological index can be calculated as described in Astarita et al., Nat Immunol, 16: 75-84, 2015. In some embodiments, morphological index is calculated by the perimeter of the CAFs. 2 / 4π x area, where "perimeter" is the perimeter of the cell and "area" is the area of ​​the cell. In some embodiments, the morphology index of a cell is between 10 and 30, e.g., 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30.

[0368] ii. Immune checkpoint inhibition In some embodiments, the downstream activity is immune checkpoint inhibition. In some embodiments, the protein in Table 1 is PD-L1 and the downstream activity is immune checkpoint inhibition. In some embodiments, the protein in Table 1 is PD-L1 and the protein in Table 2 is EPHA3 and the downstream activity is immune checkpoint inhibition.

[0369] In other embodiments, the protein in Table 1 is PD-L2 and the downstream activity is immune checkpoint inhibition. In some embodiments, the protein in Table 1 is PD-L2 and the protein in Table 2 is CEACAM4, ICAM5, NECTIN3, PSG9, or TNFRSF11A, and the downstream activity is immune checkpoint inhibition.

[0370] In some embodiments, the immune checkpoint inhibition assay is a cell-based assay, such as the cell-based assay described in Skalniak et al., Oncotarget, 8: 72167-72181, 2017. In some embodiments, the immune checkpoint inhibition assay is the assay described in Mariathasan et al., Nature, 554: 544-548. In some embodiments, the assayed cells are further treated with a modulator, e.g., by adding it to the cell culture medium. In some embodiments, immune checkpoint inhibition is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, immune checkpoint inhibition is increased by at least 30% in the presence of the modulator.

[0371] iii. Inhibition of cell proliferation In some embodiments, the downstream activity is inhibition of cell proliferation. In some embodiments, the protein in Table 1 is PTPRD and the downstream activity is inhibition of cell proliferation. In some embodiments, the protein in Table 1 is PTPRD and the protein in Table 2 is BMP5, CEACAM3, IL1RAP, IL1RAPL2, LECT1, LRFN5, SIRPG, SLITRK3, SLITRK4, SLITRK6, or TGFA and the downstream activity is inhibition of cell proliferation. In some embodiments, the PTPRD protein comprises G203E and K204E; R232C and R233C; P249L; G285E; E406K; S431L; R561Q; P666S; E755K; V892I; S912F; R995C; or R1088C amino acid substitution mutations or a ΔG61ΔE106 amino acid deletion mutation and the downstream activity is inhibition of cell proliferation.

[0372] In other embodiments, the protein in Table 1 is CNTN1 and the downstream activity is inhibition of cell proliferation. In some embodiments, the protein in Table 1 is CNTN1 and the protein in Table 2 is CDH6, CHL1, FCGRT, PCDHB7, or SGCG and the downstream activity is inhibition of cell proliferation.

[0373] In yet other embodiments, the protein in Table 1 is CHL1 and the downstream activity is inhibition of cell proliferation. In some embodiments, the protein in Table 1 is CHL1 and the protein in Table 2 is CNTN1, CNTN5, SIRPA, L1CAM, or TMEM132A and the downstream activity is inhibition of cell proliferation.

[0374] In some embodiments, the assay for inhibition of cell proliferation is a colony formation assay, e.g., a colony formation assay such as that described in Yan et al., Cancer Res, 76(6): 1603-1614, 2016 or Ognibene et al., Oncotarget, 9(40): 25903-25921, 2018. In some embodiments, the cells assayed in the colony formation assay are further treated with a modulator, e.g., by adding it to the cell culture medium. In some embodiments, cell proliferation is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured in the colony formation assay. In some embodiments, cell proliferation is reduced by at least 30% in the presence of the modulator, as measured in the colony formation assay.

[0375] In some embodiments, the assay for inhibition of cell proliferation is a cell proliferation assay, e.g., a cell proliferation assay such as that described in Yan et al., Cancer Res, 76(6): 1603-1614, 2016. In some embodiments, the cells assayed in the cell proliferation assay are further treated with a modulator, e.g., by adding it to the cell culture medium. In some embodiments, cell proliferation is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured in the cell proliferation assay. In some embodiments, cell proliferation is reduced by at least 30% in the presence of the modulator, as measured in the cell proliferation assay.

[0376] iv. Regulation of target phosphorylation In some embodiments, the downstream activity is phosphorylation or inhibition of phosphorylation of a target protein, e.g., phosphorylation of EGFR or inhibition of STAT3 phosphorylation. In some embodiments, the protein in Table 1 is PTPRD and the downstream activity is inhibition of STAT3 phosphorylation. In some embodiments, the protein in Table 1 is PTPRD and the protein in Table 2 is BMP5, CEACAM3, IL1RAP, IL1RAPL2, LECT1, LRFN5, SIRPG, SLITRK3, SLITRK4, SLITRK6, or TGFA and the downstream activity is inhibition of STAT3 phosphorylation. In some embodiments, the PTPRD protein comprises G203E and K204E; R232C and R233C; P249L; G285E; E406K; S431L; R561Q; P666S; E755K; V892I; S912F; R995C; or R1088C amino acid substitution mutations or a ΔG61ΔE106 amino acid deletion mutation, and the downstream activity is inhibition of STAT3 phosphorylation.

[0377] In some embodiments, the assay for inhibition of STAT3 phosphorylation is a Western blot of phosphorylated STAT3, e.g., a Western blot as described in Veeriah et al., PNAS, 106(23): 9435-9440, 2009 or Peyser et al., PLoS ONE, 10.1371 / journal.pone.0135750, 2015. In some embodiments, the cells assayed in the Western blot are further treated with a modulator, e.g., by adding it to the cell culture medium. In some embodiments, inhibition of STAT3 phosphorylation, as measured by Western blot of phosphorylated STAT3, is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, inhibition of STAT3 phosphorylation is increased by at least 30% in the presence of the modulator, as measured by Western blot of phosphorylated STAT3.

[0378] In some embodiments, inhibition of STAT3 phosphorylation is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured by Western blot of phosphorylated STAT3, hi some embodiments, inhibition of STAT3 phosphorylation is reduced by at least 30% in the presence of the modulator, as measured by Western blot of phosphorylated STAT3.

[0379] In other embodiments, the protein in Table 1 is PTPRF and the downstream activity is phosphorylation of EGFR. In some embodiments, the protein in Table 1 is PTPRF and the protein in Table 2 is CD177, IL1RAP, or LRFN5, and the downstream activity is phosphorylation of EGFR.

[0380] In some embodiments, the assay for EGFR phosphorylation is a phospho-EGFR Western blot, e.g., a Western blot as described in Du et al., J Cell Sci, 126: 1440-1453, 2013. In some embodiments, the cells assayed in a Western blot are further treated with a modulator, e.g., by adding it to the cell culture medium. In some embodiments, EGFR phosphorylation is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured by phospho-EGFR Western blot. In some embodiments, EGFR phosphorylation is reduced by at least 30% in the presence of the modulator, as measured by phospho-EGFR Western blot.

[0381] In some embodiments, phosphorylation of EGFR is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured by Western blot of phospho-EGFR, hi some embodiments, phosphorylation of EGFR is increased by at least 30% in the presence of the modulator, as measured by Western blot of phospho-EGFR.

[0382] In other embodiments, the protein in Table 1 is AXL and the downstream activity is phosphorylation of AXL. In some embodiments, the protein in Table 1 is AXL and the protein in Table 2 is IL1RL1 or VSIG10L and the downstream activity is phosphorylation of AXL.

[0383] In some embodiments, the assay for AXL phosphorylation is a Western blot of phosphorylated AXL. In some embodiments, the cells assayed in a Western blot are further treated with a modulator, for example, by adding it to the cell culture medium. In some embodiments, AXL phosphorylation, as measured by Western blot of phosphorylated AXL, increases by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, AXL phosphorylation, as measured by Western blot of phosphorylated AXL, decreases by at least 30% in the presence of the modulator.

[0384] In some embodiments, phosphorylation of AXL is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured by Western blot of phosphorylated AXL, hi some embodiments, phosphorylation of AXL is increased by at least 30% in the presence of the modulator, as measured by Western blot of phosphorylated AXL.

[0385] v. Inhibition of cell migration In some embodiments, the downstream activity is inhibition of cell migration. In some embodiments, the protein in Table 1 is PTPRF and the downstream activity is inhibition of cell migration. In some embodiments, the protein in Table 1 is PTPRF and the protein in Table 2 is CD177, IL1RAP, or LRFN5 and the downstream activity is inhibition of cell migration.

[0386] In some embodiments, the downstream activity is inhibition of cell migration. In some embodiments, the protein of Table 1 is a PTPRS and the downstream activity is inhibition of cell migration. In some embodiments, the protein of Table 1 is a PTPRS and the protein of Table 2 is C6orf25, IL1RAP, IL1RAPL1, IL1RAPL2, LRFN1, LRFN5, LRRC4B, NCAM1, SLITRK1, SLITRK2, SLITRK3, SLITRK4, or SLITRK6 and the downstream activity is inhibition of cell migration.

[0387] In some embodiments, the assay for inhibition of cell migration is a cell migration assay, e.g., a cell migration assay such as that described in Du et al., J Cell Sci, 126: 1440-1453, 2013. In some embodiments, the cells assayed in the cell migration assay are further treated with a modulator, e.g., by adding it to the cell culture medium. In some embodiments, the inhibition of cell migration, as measured in the cell migration assay, is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, the inhibition of cell migration, as measured in the cell migration assay, is decreased by at least 30% in the presence of the modulator.

[0388] vi. Inhibition of tumor formation In some embodiments, the downstream activity is suppression of tumorigenesis. In some embodiments, the protein in Table 1 is CHL1 and the downstream activity is suppression of tumorigenesis. In some embodiments, the protein in Table 1 is CHL1 and the protein in Table 2 is CNTN1, CNTN5, SIRPA, L1CAM, or TMEM132A and the downstream activity is suppression of tumorigenesis.

[0389] In some embodiments, the assay for inhibition of tumor formation is an in vitro tumorigenicity assay, e.g., a tumorigenicity assay such as that described in Tang et al., Oncogene, doi: 10.1038 / s41388-018-0648-7, 2019, an XTT cell proliferation assay, a foci formation assay, a colony formation assay in soft agar, or a tumorigenicity assay in nude mice. In some embodiments, the cells assayed in the tumorigenicity assay are further treated with a modulator, e.g., by adding it to the cell culture medium. In some embodiments, inhibition of tumor formation, as measured in the tumorigenicity assay, is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, inhibition of tumor formation, as measured in the tumorigenicity assay, is increased by at least 30% in the presence of the modulator.

[0390] vii. Inhibition of cell invasion In some embodiments, the downstream activity is cell invasion. In some embodiments, the protein in Table 1 is CNTN1 and the downstream activity is cell invasion. In some embodiments, the protein in Table 1 is CNTN1 and the protein in Table 2 is CDH6, CHL1, FCGRT, PCDHB7, or SGCG and the downstream activity is cell invasion.

[0391] In other embodiments, the protein in Table 1 is AXL and the downstream activity is cell invasion. In some embodiments, the protein in Table 1 is AXL and the protein in Table 2 is IL1RL1 or VSIG10L and the downstream activity is cell invasion.

[0392] In still other embodiments, the protein in Table 1 is CHL1 and the downstream activity is cell invasion. In some embodiments, the protein in Table 1 is CHL1 and the protein in Table 2 is CNTN1, CNTN5, SIRPA, L1CAM, or TMEM132A and the downstream activity is cell invasion.

[0393] In some embodiments, the assay for cell invasion is a gel invasion assay, e.g., a gel invasion assay such as that described in Yan et al., Cancer Res, 76(6): 1603-1614, 2016 or He et al., Biochem Biophys Res Commun, 438: 433-438, 2013. In some embodiments, cells assayed in a gel invasion assay are further treated with a modulator, e.g., by adding it to cell culture medium. In some embodiments, cell invasion is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured in a gel invasion assay. In some embodiments, cell invasion is reduced by at least 30% in the presence of the modulator, as measured in a gel invasion assay.

[0394] viii. Macrophage polarization and phagocytic function In some embodiments, the downstream activity is inhibition of phagocytic function of a phagocyte (e.g., a macrophage), e.g., antibody-dependent cellular phagocytosis (inhibition of phagocytosis). In some embodiments, the protein in Table 1 is LILRB1 and the downstream activity is inhibition of phagocytosis. In some embodiments, the protein in Table 1 is LILRB1 and the protein in Table 2 is CLEC6A, CXADR, EDAR, FLT4, IL6R, ILDR1, or LILRA5 and the downstream activity is inhibition of phagocytosis.

[0395] Phagocytic function can be measured, for example, as the percentage of macrophages containing a fluorescent signal, where the presence of fluorescence indicates phagocytosis of fluorescently labeled target cells. A representative assay for phagocytosis is described in Barkal et al., Nature Immunol, 19: 76-84, 2017. In some embodiments, inhibition of phagocytosis is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured in a phagocytosis assay. In some embodiments, inhibition of phagocytosis is reduced by at least 30% in the presence of the modulator, as measured in a phagocytosis assay.

[0396] In some embodiments, inhibition of phagocytosis is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured in a phagocytosis assay, hi some embodiments, inhibition of phagocytosis is increased by at least 30% in the presence of the modulator, as measured in a phagocytosis assay.

[0397] In some embodiments, the downstream activity is promotion of M2 macrophage polarization. In some embodiments, the protein in Table 1 is LILRB2, and the downstream activity is promotion of M2 macrophage polarization. In some embodiments, the protein in Table 1 is LILRB2 and the protein in Table 2 is IGSF8 or MOG, and the downstream activity is promotion of M2 macrophage polarization. M2 macrophage polarization can be assessed as described in Chen et al., J Clin Invest, 128(12), 5647-5662. In some embodiments, M2 macrophage polarization is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, M2 macrophage polarization is reduced by at least 30% in the presence of the modulator.

[0398] In some embodiments, M2 macrophage polarization is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, hi some embodiments, M2 macrophage polarization is increased by at least 30% in the presence of the modulator.

[0399] ix. Osteoclast differentiation In some aspects, the downstream activity is osteoclast differentiation.

[0400] In some embodiments, the protein in Table 1 is LILRB1 and the downstream activity is osteoclast differentiation. In some embodiments, the protein in Table 1 is LILRB1 and the protein in Table 2 is CLEC6A, CXADR, EDAR, FLT4, IL6R, ILDR1, or LILRA5 and the downstream activity is osteoclast differentiation.

[0401] In some embodiments, the protein in Table 1 is LILRB3 and the downstream activity is osteoclast differentiation. In some embodiments, the protein in Table 1 is LILRB3 and the protein in Table 2 is LRRC15 or LY6G6F and the downstream activity is osteoclast differentiation.

[0402] In some embodiments, the protein in Table 1 is LILRB4 and the downstream activity is osteoclast differentiation. In some embodiments, the protein in Table 1 is LILRB4 and the protein in Table 2 is CNTFR and the downstream activity is osteoclast differentiation.

[0403] In some embodiments, the assay for osteoclast differentiation is an assay for multinucleated cells positive for tartrate-resistant acid phosphatase (TRAP) staining (TRAP+ multinucleated cells). The TRAP+ status and multiple nuclei (e.g., three or more nuclei) are indicative of a cell being an osteoclast. A representative assay for TRAP+ multinucleated cells is described in Mori et al., J Immunol, 181(7): 4742-4751, 2008. In some embodiments, cells assayed in a TRAP+ multinucleated cell assay are further treated with a modulator, for example, by adding it to the cell culture medium. In some embodiments, osteoclast differentiation, as measured in a TRAP+ multinucleated cell assay, is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, osteoclast differentiation is reduced by at least 30% in the presence of the modulator as measured in a TRAP+ multinucleated cell assay.

[0404] In other embodiments, osteoclast differentiation is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, as measured in a TRAP+ multinucleated cell assay. In some embodiments, osteoclast differentiation is increased by at least 30% in the presence of the modulator, as measured in a TRAP+ multinucleated cell assay.

[0405] x. Activation of signaling pathways In some embodiments, the downstream activity is activation of a signal transduction pathway. In some embodiments, the protein in Table 1 is AXL and the downstream activity is activation of the RAS / RAF / MAPK / ERK1 / 2 pathway, activation of the JAK / STAT pathway, or activation of the PI3K signaling pathway. In some embodiments, the protein in Table 1 is AXL and the protein in Table 2 is IL1RL1 or VSIG10L and the downstream activity is activation of the RAS / RAF / MAPK / ERK1 / 2 pathway, activation of the JAK / STAT pathway, or activation of the PI3K signaling pathway.

[0406] In some embodiments, the protein in Table 1 is a PTPRS and the downstream activity is activation of the PI3K signaling pathway. In some embodiments, the protein in Table 1 is a PTPRS and the protein in Table 2 is C6orf25, IL1RAP, IL1RAPL1, IL1RAPL2, LRFN1, LRFN5, LRRC4B, NCAM1, SLITRK1, SLITRK2, SLITRK3, SLITRK4, or SLITRK6 and the downstream activity is activation of the PI3K signaling pathway.

[0407] In other embodiments, the protein in Table 1 is CNTN1 and the downstream activity is activation of the RhoA pathway or the Akt pathway. In some embodiments, the protein in Table 1 is CNTN1 and the protein in Table 2 is CDH6, CHL1, FCGRT, PCDHB7, or SGCG and the downstream activity is activation of the RhoA pathway or the Akt pathway.

[0408] In some embodiments, activation of the RAS / RAF / MAPK / ERK1 / 2 pathway, activation of the JAK / STAT pathway, activation of the RhoA pathway, activation of the Akt pathway, or activation of the PI3K signaling pathway is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, hi some embodiments, activation of the RAS / RAF / MAPK / ERK1 / 2 pathway, activation of the JAK / STAT pathway, activation of the RhoA pathway, activation of the Akt pathway, or activation of the PI3K signaling pathway is reduced by at least 30% in the presence of the modulator.

[0409] In some embodiments, activation of the RAS / RAF / MAPK / ERK1 / 2 pathway, activation of the JAK / STAT pathway, activation of the RhoA pathway, activation of the Akt pathway, or activation of the PI3K signaling pathway is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, hi some embodiments, activation of the RAS / RAF / MAPK / ERK1 / 2 pathway, activation of the JAK / STAT pathway, activation of the RhoA pathway, activation of the Akt pathway, or activation of the PI3K signaling pathway is increased by at least 30% in the presence of the modulator.

[0410] xi. Formation of filopodia In some embodiments, the downstream activity is filopodia formation. In some embodiments, the protein in Table 1 is AXL and the downstream activity is filopodia formation. In some embodiments, the protein in Table 1 is AXL and the protein in Table 2 is IL1RL1 or VSIG10L and the downstream activity is filopodia formation. In some embodiments, filopodia formation is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, filopodia formation is reduced by at least 30% in the presence of the modulator.

[0411] xii. Regulation of EMT In some embodiments, the downstream activity is inhibition of epithelial-mesenchymal transition (EMT). EMT enhances the migratory and survival properties of cancer cells and promotes malignant progression (Gjerdrum et al., PNAS, 107(3): 1124-1129, 2010). In some embodiments, the protein in Table 1 is AXL, and the downstream activity is inhibition of EMT. In some embodiments, the protein in Table 1 is AXL and the protein in Table 2 is IL1RL1 or VSIG10L, and the downstream activity is inhibition of EMT. EMT can be quantified as described in Gjerdrum et al., PNAS, 107(3): 1124-1129, 2010. In some embodiments, inhibition of EMT is increased by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator, hi some embodiments, inhibition of EMT is increased by at least 30% in the presence of the modulator.

[0412] xii. Tumor growth In some embodiments, the downstream activity is tumor growth. In some embodiments, the protein in Table 1 is PDPN and the protein in Table 2 is CD177, and the downstream activity is tumor growth. In some embodiments, tumor growth is reduced by at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or 100% in the presence of the modulator. In some embodiments, tumor growth is reduced by at least 30% in the presence of the modulator.

[0413] IV. Modulators of Protein-Protein Interactions In some aspects, the disclosure features an isolated modulator of an interaction between a protein of Table 1 and a protein of Table 2, where (a) the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (b) the modulator causes an increase or decrease in binding of the protein of Table 1 to the protein of Table 2 compared to binding in the absence of the modulator.

[0414] In some aspects, the disclosure features an isolated modulator of a downstream activity of a protein of Table 1 or a protein of Table 2, where (a) the protein of Table 1 and the protein of Table 2 are reported to interact in Table 3; and (b) the modulator causes a change in the downstream activity of the protein of Table 1 or the protein of Table 2 compared to the downstream activity in the absence of the modulator.

[0415] In some embodiments, the modulator is an inhibitor or an activator of a downstream activity of a protein of Table 1 or Table 2.

[0416] In some embodiments, the alteration in a downstream activity is an increase or decrease in the amount, intensity, or duration of a downstream activity, e.g., a downstream activity described in Section IIIB herein, e.g., CAF contractility, immune checkpoint inhibition, inhibition of cell proliferation, modulation of target phosphorylation, inhibition of cell migration, inhibition of tumorigenesis, inhibition of cell invasion, macrophage polarization, modulation of phagocytosis, osteoclast differentiation, activation of a signaling pathway, or filopodia formation.

[0417] A. Small molecules In some embodiments, the modulator or candidate modulator is a small molecule. A small molecule is a molecule other than a binding polypeptide or antibody as defined herein that can bind, preferably specifically, to a protein of Table 1 and / or a protein of Table 2. Binding small molecules can be identified and chemically synthesized using known methodologies (see, e.g., PCT Publication Nos. WO00 / 00823 and WO00 / 39585). The size of a binding small molecule is typically less than about 2000 daltons (e.g., less than about 2000, 1500, 750, 500, 250, or 200 daltons in size), where such small organic molecules capable of preferably specifically binding to a polypeptide described herein can be identified without undue experimentation using known techniques. In this regard, it should be noted that techniques for screening small molecule libraries for molecules capable of binding to polypeptide targets are well known in the art (see, e.g., PCT Publication Nos. WO00 / 00823 and WO00 / 39585). The conjugated small molecule can be, for example, an aldehyde, ketone, oxime, hydrazone, semicarbazone, carbazide, primary amine, secondary amine, tertiary amine, N-substituted hydrazine, hydrazide, alcohol, ether, thiol, thioether, disulfide, carboxylic acid, ester, amide, urea, carbamate, carbonate, ketal, thioketal, acetal, thioacetal, aryl halide, aryl sulfonate, alkyl halide, alkyl sulfonate, aromatic compound, heterocyclic compound, aniline, alkene, alkyne, diol, amino alcohol, oxazolidine, oxazoline, thiazolidine, thiazoline, enamine, sulfonamide, epoxide, aziridine, isocyanate, sulfonyl chloride, diazo compound, acid chloride, and the like.

[0418] In some embodiments, binding between a protein of Table 1 and a protein of Table 2 is decreased in the presence of a small molecule (e.g., decreased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%). In some embodiments, binding between a protein of Table 1 and a protein of Table 2 is increased in the presence of a small molecule (e.g., increased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, or more than 100%). In some aspects, a downstream activity of a protein of Table 1 and / or a protein of Table 2 (e.g., a downstream activity described in Section IIIB herein, e.g., CAF contractility, immune checkpoint inhibition, inhibition of cell proliferation, modulation of target phosphorylation, inhibition of cell migration, inhibition of tumorigenesis, inhibition of cell invasion, macrophage polarization, modulation of phagocytosis, osteoclast differentiation, activation of a signaling pathway, or filopodia formation) is decreased (e.g., decreased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%) in the presence of the small molecule. In some embodiments, a downstream activity of a protein of Table 1 and / or a protein of Table 2 (e.g., a downstream activity described in Section IIIB herein) is increased (e.g., 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, or greater than 100%) in the presence of the small molecule.

[0419] B. Antibodies and Antigen-Binding Fragments In some embodiments, the modulator or candidate modulator is an antibody or antigen-binding fragment thereof that binds to a protein of Table 1 and / or a protein of Table 2. In some embodiments, the antigen-binding fragment is a bis-Fab, Fv, Fab, Fab'-SH, F(ab')2, diabody, linear antibody, scFv, scFab, VH domain, or VHH domain.

[0420] In some embodiments, binding between a protein of Table 1 and a protein of Table 2 is decreased in the presence of the antibody or antigen-binding fragment (e.g., decreased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%). In some embodiments, binding between a protein of Table 1 and a protein of Table 2 is increased in the presence of the antibody or antigen-binding fragment (e.g., increased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, or more than 100%). In some aspects, a downstream activity of a protein of Table 1 and / or a protein of Table 2 (e.g., a downstream activity described in Section IIIB herein, e.g., CAF contractility, immune checkpoint inhibition, inhibition of cell proliferation, modulation of target phosphorylation, inhibition of cell migration, inhibition of tumorigenesis, inhibition of cell invasion, macrophage polarization, modulation of phagocytosis, osteoclast differentiation, activation of a signaling pathway, or filopodia formation) is decreased (e.g., decreased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%) in the presence of the antibody or antigen-binding fragment. In some embodiments, a downstream activity of a protein of Table 1 and / or a protein of Table 2 (e.g., a downstream activity described in Section IIIB herein) is increased (e.g., increased by 5%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, or more than 100%) in the presence of the antibody or antigen-binding fragment.

[0421] C. Peptides In some embodiments, the modulator or candidate modulator is a peptide that binds to a protein of Table 1 and / or a protein of Table 2. The peptide can be a naturally occurring peptide or can be an engineered peptide. In some embodiments, the peptide is a fragment of a protein of Table 1, a protein of Table 2, or another protein that binds to a protein of Table 1 or a protein of Table 2. The peptide can bind to a binding partner with equal, less, or greater affinity than the full-length protein. In some embodiments, the peptide performs all of the functions of the full-length protein. In other embodiments, the peptide does not perform all of the functions of the full-length protein.

[0422] In some embodiments, binding between a protein of Table 1 and a protein of Table 2 is decreased in the presence of the peptide (e.g., decreased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%). In some embodiments, binding between a protein of Table 1 and a protein of Table 2 is increased in the presence of the peptide (e.g., increased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, or more than 100%). In some embodiments, a downstream activity of a protein of Table 1 and / or a protein of Table 2 (e.g., a downstream activity described in Section IIIB herein, e.g., CAF contractility, immune checkpoint inhibition, inhibition of cell proliferation, modulation of target phosphorylation, inhibition of cell migration, inhibition of tumorigenesis, inhibition of cell invasion, macrophage polarization, modulation of phagocytosis, osteoclast differentiation, activation of a signaling pathway, or filopodia formation) is decreased in the presence of the peptide (e.g., decreased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%). In some embodiments, a downstream activity of a protein of Table 1 and / or a protein of Table 2 (e.g., a downstream activity described in Section IIIB herein) is increased (e.g., by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, or more than 100%) in the presence of the peptide.

[0423] D. Imitation In some embodiments, the modulator or candidate modulator is a mimetic, e.g., a molecular mimetic, that binds to a protein of Table 1 and / or a protein of Table 2. The mimetic can be a molecular mimetic of a protein of Table 1, a protein of Table 2, or another protein that binds to a protein of Table 1 or a protein of Table 2. In some embodiments, the mimetic can perform all of the functions of the mimicked polypeptide. In other embodiments, the mimetic does not perform all of the functions of the mimicked polypeptide.

[0424] In some embodiments, binding between a protein of Table 1 and a protein of Table 2 is decreased in the presence of the mimetic (e.g., decreased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%). In some embodiments, binding between a protein of Table 1 and a protein of Table 2 is increased in the presence of the mimetic (e.g., increased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, or more than 100%). In some embodiments, a downstream activity of a protein of Table 1 and / or a protein of Table 2 (e.g., a downstream activity described in Section IIIB herein, e.g., CAF contractility, immune checkpoint inhibition, inhibition of cell proliferation, modulation of target phosphorylation, inhibition of cell migration, inhibition of tumorigenesis, inhibition of cell invasion, macrophage polarization, modulation of phagocytosis, osteoclast differentiation, activation of a signaling pathway, or filopodia formation) is decreased in the presence of the mimetic (e.g., decreased by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, or 100%). In some embodiments, a downstream activity of a protein of Table 1 and / or a protein of Table 2 (e.g., a downstream activity described in Section IIIB herein) is increased (e.g., by 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 100%, or more than 100%) in the presence of the mimetic.

[0425] V. Therapeutic Methods Involving Identified Modulators of Protein-Protein Interactions In some embodiments, modulators of protein-protein interactions described in Section IIC herein are used to treat or delay the progression of a pathological state, disease, disorder, or condition, e.g., cancer.

[0426] In some embodiments, the modulator increases or decreases the amount, intensity, or duration of a downstream activity described in Section IIIB herein, e.g., CAF contractility, immune checkpoint inhibition, suppression of cell proliferation, modulation of target phosphorylation, inhibition of cell migration, suppression of tumorigenesis, suppression of cell invasion, macrophage polarization, modulation of phagocytosis, osteoclast differentiation, activation of signaling pathways, or filopodia formation, in an individual to whom the modulator is administered.

[0427] A. Cancer In some embodiments, a modulator of a protein-protein interaction (e.g., a small molecule, antibody, antigen-binding fragment, peptide, mimetic, antisense oligonucleotide, or siRNA) described in Section IIC herein is used to treat or delay the progression of cancer in a subject in need thereof. In some embodiments, the subject is a human. The cancer can be a solid tumor cancer or a non-solid tumor cancer. Solid cancer tumors include, but are not limited to, bladder cancer, melanoma, breast cancer, colorectal cancer, lung cancer, head and neck cancer, kidney cancer, ovarian cancer, pancreatic cancer, or prostate cancer, or metastatic forms thereof. In some embodiments, the cancer is bladder cancer. Further embodiments of bladder cancer include urothelial carcinoma, muscle-invasive bladder cancer (MIBC), or non-muscle-invasive bladder cancer (NMIBC). In some embodiments, the bladder cancer is metastatic urothelial carcinoma (mUC). In some embodiments, the cancer is breast cancer. Further embodiments of breast cancer include hormone receptor positive (HR+) breast cancer, e.g., estrogen receptor positive (ER+) breast cancer, progesterone receptor positive (PR+) breast cancer, or ER+ / PR+ breast cancer. Other embodiments of breast cancer include HER2 positive (HER2+) breast cancer. Still other embodiments of breast cancer include triple negative breast cancer (TNBC). In some embodiments, the breast cancer is early stage breast cancer. In some embodiments, the cancer is lung cancer. Further embodiments of lung cancer include epidermal growth factor receptor positive (EGFR+) lung cancer. Other embodiments of lung cancer include epidermal growth factor receptor negative (EGFR-) lung cancer. Still other embodiments of lung cancer include non-small cell lung cancer, e.g., squamous cell lung cancer or non-squamous cell lung cancer. Other embodiments of lung cancer include small cell lung cancer. In some embodiments, the cancer is head and neck cancer. Further embodiments of head and neck cancer include squamous cell carcinoma of the head and neck (SCCHN). In some embodiments, the cancer is kidney cancer. Further kidney cancer embodiments include renal cell carcinoma (RCC). In some embodiments, the cancer is liver cancer. Further liver cancer embodiments include hepatocellular carcinoma. In some embodiments, the cancer is prostate cancer. Further prostate cancer embodiments include castration-resistant prostate cancer (CRPC).In some embodiments, the cancer is a metastatic form of a solid tumor. In some embodiments, the metastatic form of a solid tumor is a metastatic form of melanoma, breast cancer, colorectal cancer, lung cancer, head and neck cancer, bladder cancer, kidney cancer, ovarian cancer, pancreatic cancer, or prostate cancer. In some embodiments, the cancer is metastatic urothelial carcinoma (mUC). In some embodiments, the cancer is a non-solid tumor cancer. Non-solid tumor cancers include, but are not limited to, B-cell lymphoma. Further embodiments of B-cell lymphoma include, for example, chronic lymphocytic leukemia (CLL), diffuse large B-cell lymphoma (DLBCL), follicular lymphoma, myelodysplastic syndrome (MDS), non-Hodgkin's lymphoma (NHL), acute lymphoblastic leukemia (ALL), multiple myeloma, acute myeloid leukemia (AML), or mycosis fungoides (MF). In some embodiments, the cancer is colorectal cancer.

[0428] B. Method of Delivery The compositions utilized in the methods described herein (e.g., modulators of protein-protein interactions described in Section IIC, e.g., small molecules, antibodies, antigen-binding fragments, peptides, mimetics, antisense oligonucleotides, or siRNA) can be administered in any suitable manner, including, but not limited to, intravenously, intramuscularly, subcutaneously, intradermally, transdermally, intra-arterially, intraperitoneally, intralesionally, intracranially, intra-articularly, intraprostatically, intrapleurally, intratracheally, intrathecally, intranasally, intravaginally, intrarectally, topically, intratumorally, peritoneally, subconjunctivally, intravesicularly, mucosally, intrapericardially, intraumbilically, intraocularly, intraorbitally, orally, transdermally, intravitreally (e.g., by intravitreal injection), by eye drop, by inhalation, by injection, by implantation, by infusion, by continuous infusion, directly by a local perfusion bath targeting cells, by catheter, by perfusion, in a cream, or in a lipid composition. The compositions utilized in the methods described herein can also be administered systemically or locally. The method of administration can vary depending on various factors (e.g., the compound or composition being administered and the severity of the condition, disease, or disorder being treated). In some aspects, modulators of protein-protein interactions are administered intravenously, intramuscularly, subcutaneously, topically, orally, transdermally, intraperitoneally, intraorbitally, by implantation, by inhalation, intrathecally, intracerebroventricularly, or intranasally. Dosing can be by any suitable route, for example, injection, such as intravenous or subcutaneous injection, depending in part on whether the administration is brief or chronic. Various dosing schedules are contemplated herein, including, but not limited to, a single dose or multiple doses over various time periods, a bolus dose, or a pulse infusion.

[0429] The modulators of protein-protein interactions described herein (and any additional therapeutic agents) may be formulated, dosed, and administered in a manner consistent with good medical practice. Factors to consider in this regard include the particular disorder being treated, the particular mammal being treated, the clinical condition of the individual patient, the cause of the disorder, the site of drug delivery, the method of administration, the schedule of administration, and other factors known to medical professionals. The modulator need not be, but is optionally, formulated with and / or co-administered with one or more agents currently used to prevent or treat the disorder in question. The effective amount of such other agents will depend on the amount of modulator present in the formulation, the type of disorder or treatment, and other factors discussed above. These will generally be used in the same dosages and by any route of administration as described herein, or at about 1% to 99% of the dosages described herein, or at any dosage and by any route determined empirically / clinically to be appropriate.

[0430] VI. Methods of Treatment Involving PD-L1 Axis Binding Antagonists A. Markers of response to atezolizumab In some aspects, the invention includes methods of identifying an individual having a cancer that is treatable with a PD-L1 axis-binding antagonist (e.g., a cancer that is likely to be treatable with a PD-L1 axis-binding antagonist), comprising determining the expression levels of the first member and the second member of at least one gene pair of Table 15 in a sample from the individual, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual as having a cancer that is treatable with a PD-L1 axis-binding antagonist.

[0431] In some aspects, the invention includes methods of identifying an individual having cancer who may benefit from a treatment comprising a PD-L1 axis-binding antagonist, comprising determining the expression levels of the first member and the second member of at least one gene pair of Table 15 in a sample from the individual, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual as an individual who may benefit from a treatment comprising a PD-L1 axis-binding antagonist.

[0432] In some aspects, the invention includes methods of selecting a therapy for an individual with cancer, comprising determining the expression levels of a first member and a second member of at least one gene pair of Table 15 in a sample from the individual, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual as one that is likely to benefit from therapy comprising a PD-L1 axis-binding antagonist.

[0433] In some aspects, the individual has an expression level of a first member of the gene pair that is greater than a first reference expression level and an expression level of a second member of the gene pair that is greater than a second reference expression level, and the method further comprises administering to the individual an effective amount of a PD-L1 axis-binding antagonist.

[0434] In some aspects, the invention comprises methods of treating an individual with cancer, the method comprising: (a) determining the expression levels of a first member and a second member of at least one gene pair of Table 15 in a sample from the individual, wherein the expression level of the first member of the gene pair is greater than a first reference expression level, and the expression level of the second member of the gene pair is greater than the second reference expression level; and (b) administering to the individual an effective amount of a PD-L1 axis-binding antagonist.

[0435] In some aspects, the invention comprises methods of treating an individual having cancer, the method comprising administering a PD-L1 axis-binding antagonist to the individual determined to have an expression level of a first member of a gene pair in Table 15 that is greater than a first reference expression level and an expression level of a second member of the gene pair that is greater than a second reference expression level.

[0436] In some embodiments, the benefit comprises an increase in the overall survival (OS) of the individual compared to treatment without the PD-L1 axis binding antagonist, hi other embodiments, the benefit may comprise, for example, an increase in the time to recurrence or a decrease in the duration of treatment compared to treatment without the PD-L1 axis binding antagonist.

[0437] In some embodiments, the first member of the gene pair is SIGLEC6 and the second member of the gene pair is NCR1.

[0438] In some embodiments, the first member of the gene pair is BTN3A1 and the second member of the gene pair is LRRC4B.

[0439] In some embodiments, the first member of the gene pair is CD80 and the second member of the gene pair is CTLA4.

[0440] In some embodiments, the first member of the gene pair is BTN3A3 and the second member of the gene pair is LRRC4B.

[0441] In some embodiments, the first member of the gene pair is NCR1 and the second member of the gene pair is SIGLEC8.

[0442] In some embodiments, the first member of the gene pair is CNTFR and the second member of the gene pair is LILRB4.

[0443] In some embodiments, the first member of the gene pair is FGFR3 and the second member of the gene pair is LRRTM2.

[0444] In some embodiments, the first member of the gene pair is FGFR4 and the second member of the gene pair is SIGLEC15.

[0445] In some embodiments, the first member of the gene pair is FGFR1 and the second member of the gene pair is KL.

[0446] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is TRHDE.

[0447] In some embodiments, the first member of the gene pair is CTLA4 and the second member of the gene pair is PCDHGB4.

[0448] In some embodiments, the first member of the gene pair is CTLA4 and the second member of the gene pair is FAM200A.

[0449] In some embodiments, the first member of the gene pair is CA12 and the second member of the gene pair is SIGLEC6.

[0450] In some embodiments, the first member of the gene pair is ILDR2 and the second member of the gene pair is CLEC12B.

[0451] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is ITLN1.

[0452] In some embodiments, the first member of the gene pair is CADM1 and the second member of the gene pair is CRTAM.

[0453] In some embodiments, the first member of the gene pair is CD79B and the second member of the gene pair is CD244.

[0454] In some embodiments, the first member of the gene pair is DAG1 and the second member of the gene pair is EFNB1.

[0455] B. Markers of lack of response to atezolizumab In some aspects, the invention includes methods of identifying an individual having cancer who may benefit from a therapy other than, or in addition to, a PD-L1 axis-binding antagonist, comprising determining the expression levels of the first member and the second member of at least one gene pair of Table 16 in a sample from the individual, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual as one who may benefit from a therapy other than, or in addition to, a PD-L1 axis-binding antagonist.

[0456] In some aspects, the invention includes methods of selecting a therapy for an individual with cancer, comprising determining the expression levels of the first member and the second member of at least one gene pair of Table 16 in a sample from the individual, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual as one that may benefit from a treatment other than, or in addition to, a PD-L1 axis-binding antagonist.

[0457] In some aspects, the invention includes methods of identifying an individual having a cancer that may benefit from a treatment other than, or in addition to, a PD-L1 axis-binding antagonist, comprising determining the expression levels of the first member and the second member of at least one gene pair of Table 16 in a sample from the individual, wherein an expression level of the first member of the gene pair that is greater than the first reference expression level and an expression level of the second member of the gene pair that is greater than the second reference expression level identifies the individual having a cancer that is resistant to treatment with a PD-L1 axis-binding antagonist.

[0458] In some aspects, the invention includes methods of selecting a therapy for an individual with cancer, comprising determining the expression levels of a first member and a second member of at least one gene pair of Table 16 in a sample from the individual, wherein an expression level of the first member of the gene pair greater than the first reference expression level and an expression level of the second member of the gene pair greater than the second reference expression level identifies the individual with a cancer that is resistant to treatment with a PD-L1 axis-binding antagonist.

[0459] In some aspects, the individual has an expression level of a first member of the gene pair that is greater than a first reference expression level and an expression level of a second member of the gene pair that is greater than a second reference expression level, and the method comprises administering to the individual an effective amount of a therapy other than, or in addition to, a PD-L1 axis-binding antagonist.

[0460] In some embodiments, the benefit comprises an increase in overall survival (OS) of the individual compared to treatment with no treatment other than or in addition to a PD-L1 axis-binding antagonist.

[0461] In some embodiments, the sample from the individual is obtained from the individual prior to administration of the anti-cancer therapy, hi some embodiments, the sample from the individual is obtained from the individual after administration of the anti-cancer therapy.

[0462] In some embodiments, the sample from the individual is a tumor tissue sample or a tumor fluid sample, such as a formalin-fixed, paraffin-embedded (FFPE) sample, an archival sample, a fresh sample, or a frozen sample.

[0463] In some embodiments, (a) the expression levels of the first and second members of the gene pair in the sample are protein expression levels; or (b) the expression levels of the first and second members of the gene pair in the sample are mRNA expression levels.

[0464] In some embodiments, the expression levels of the first and second members of the gene pair in the sample are mRNA expression levels of the first and second members of the gene pair, respectively. In some embodiments, the mRNA expression levels of the first and second members of the gene pair are determined by in situ hybridization (ISH), RNA-seq, RT-qPCR, qPCR, multiplex qPCR or RT-qPCR, microarray analysis, SAGE, MassARRAY technology, FISH, or a combination thereof. In some embodiments, the RNA-seq is TruSeq RNA Access technology (Illumina®).

[0465] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is EVC2.

[0466] In some embodiments, the first member of the gene pair is GPC4 and the second member of the gene pair is FGFRL1.

[0467] In some embodiments, the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB4.

[0468] In some embodiments, the first member of the gene pair is PTPRD and the second member of the gene pair is LRFN4.

[0469] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is AQPEP.

[0470] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is DSG4.

[0471] In some embodiments, the first member of the gene pair is LDLR and the second member of the gene pair is LILRB5.

[0472] In some embodiments, the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB3.

[0473] In some embodiments, the first member of the gene pair is PLXNB3 and the second member of the gene pair is SEMA4G.

[0474] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is EPHB6.

[0475] In some embodiments, the first member of the gene pair is FLT4 and the second member of the gene pair is FLRT2.

[0476] In some embodiments, the first member of the gene pair is FLT1 and the second member of the gene pair is ELFN1.

[0477] In some embodiments, the first member of the gene pair is GPC4 and the second member of the gene pair is FGFR4.

[0478] In some embodiments, the first member of the gene pair is GPC3 and the second member of the gene pair is TNFRSF11B.

[0479] In some embodiments, the first member of the gene pair is FGFR4 and the second member of the gene pair is GPC6.

[0480] In some embodiments, the first member of the gene pair is PLXNB1 and the second member of the gene pair is SEMA4B.

[0481] In some embodiments, the first member of the gene pair is EDA and the second member of the gene pair is EDAR.

[0482] In some embodiments, the first member of the gene pair is FGFR4 and the second member of the gene pair is NRXN2.

[0483] In some embodiments, the first member of the gene pair is SEMA4D and the second member of the gene pair is PLXNB2.

[0484] In some embodiments, the first member of the gene pair is FLT4 and the second member of the gene pair is NRP2.

[0485] In some embodiments, the first member of the gene pair is FGFR4 and the second member of the gene pair is GPC3.

[0486] In some embodiments, the first member of the gene pair is FGFR2 and the second member of the gene pair is RAMP1.

[0487] In some embodiments, the first member of the gene pair is AXL1 and the second member of the gene pair is IL1RL1.

[0488] In some embodiments, the first member of the gene pair is CD320 and the second member of the gene pair is IGSF5.

[0489] In some embodiments, the first member of the gene pair is CD59 and the second member of the gene pair is STAB1.

[0490] In some embodiments, the first member of the gene pair is CNTN3 and the second member of the gene pair is PTPRG.

[0491] In some embodiments, the first member of the gene pair is EFNB1 and the second member of the gene pair is EPHA3.

[0492] In some embodiments, the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB2.

[0493] In some embodiments, the first member of the gene pair is EGF and the second member of the gene pair is TNFRSF11B.

[0494] In some embodiments, the first member of the gene pair is ENPEP and the second member of the gene pair is SLITRK1.

[0495] In some embodiments, the first member of the gene pair is FCGR3B and the second member of the gene pair is EDA2R.

[0496] In some embodiments, the first member of the gene pair is IL20RA and the second member of the gene pair is CLEC14A.

[0497] In some embodiments, the first member of the gene pair is IL6R and the second member of the gene pair is BTNL9.

[0498] In some embodiments, the first member of the gene pair is IZUMO1 and the second member of the gene pair is LILRA5.

[0499] In some embodiments, the first member of the gene pair is NGFR and the second member of the gene pair is LRRTM3.

[0500] In some embodiments, the first member of the gene pair is NTM and the second member of the gene pair is AMIGO2.

[0501] In some embodiments, the first member of the gene pair is PCDHB3 and the second member of the gene pair is IGSF11.

[0502] In some embodiments, the first member of the gene pair is PTGFRN and the second member of the gene pair is TMEM59L.

[0503] In some embodiments, the first member of the gene pair is TREM1 and the second member of the gene pair is VSIG8.

[0504] C. Reference expression level In some embodiments, the reference expression level for the first member of the protein pair (i.e., the first reference expression level) is a pre-assigned expression level, and the reference expression level for the first member of the protein pair (i.e., the second reference expression level) is a pre-a...

Claims

1. 15. A method of identifying an individual having cancer that may benefit from a treatment comprising a PD-L1 axis binding antagonist, comprising determining in a sample from the individual an expression level of a first member and a second member of at least one gene pair of Table 15, wherein an expression level of the first member of the gene pair greater than a first reference expression level and an expression level of the second member of the gene pair greater than a second reference expression level identifies the individual as an individual that may benefit from a treatment comprising a PD-L1 axis binding antagonist.

2. 15. A method of selecting a therapy for an individual having cancer, comprising determining in a sample from the individual an expression level of a first member and a second member of at least one gene pair of Table 15, wherein an expression level of the first member of the gene pair greater than a first reference expression level and an expression level of the second member of the gene pair greater than a second reference expression level identifies the individual as one who is likely to benefit from a therapy comprising a PD-L1 axis binding antagonist.

3. 3. The method of claim 1 or 2, wherein the individual has an expression level of a first member of the gene pair that is greater than a first reference expression level and an expression level of a second member of the gene pair that is greater than a second reference expression level, and the method further comprises administering to the individual an effective amount of a PD-L1 axis binding antagonist.

4. 1. A method of treating an individual having cancer, comprising: (a) determining in a sample from the individual an expression level of a first member and a second member of at least one gene pair of Table 15, wherein the expression level of the first member of the gene pair is greater than a first reference expression level, and the expression level of the second member of the gene pair is greater than a second reference expression level; and (b) administering to the individual an effective amount of a PD-L1 axis binding antagonist. The method includes:

5. 15. A method of treating an individual having cancer, comprising administering to the individual determined to have an expression level of a first member of a gene pair of Table 15 that is greater than a first reference expression level and an expression level of a second member of said gene pair that is greater than a second reference expression level, a PD-L1 axis binding antagonist.

6. 13. A method of identifying an individual having cancer that may benefit from a treatment other than, or in addition to, a PD-L1 axis binding antagonist, comprising determining in a sample from the individual the expression levels of a first member and a second member of at least one gene pair of Table 16, wherein an expression level of the first member of the gene pair greater than a first reference expression level and an expression level of the second member of the gene pair greater than a second reference expression level identifies the individual as an individual that may benefit from a treatment other than, or in addition to, a PD-L1 axis binding antagonist.

7. 13. A method of selecting a therapy for an individual having cancer, comprising determining in a sample from the individual an expression level of a first member and a second member of at least one gene pair of Table 16, wherein an expression level of the first member of the gene pair greater than a first reference expression level and an expression level of the second member of the gene pair greater than a second reference expression level identifies the individual as one who may benefit from a treatment other than, or in addition to, a PD-L1 axis binding antagonist.

8. 8. The method of claim 6 or 7, wherein the individual has an expression level of a first member of the gene pair that is greater than a first reference expression level and an expression level of a second member of the gene pair that is greater than a second reference expression level, and the method comprises administering to the individual an effective amount of a therapy other than or in addition to a PD-L1 axis binding antagonist.

9. 9. The method of any one of claims 1 to 8, wherein the first reference expression level is a pre-assigned expression level and the second reference expression level is a pre-assigned reference expression level.

10. The method of any one of claims 1 to 9, wherein the sample from the individual is obtained from the individual prior to administration of an anti-cancer therapy.

11. The method of any one of claims 1 to 9, wherein the sample from the individual is obtained from the individual after administration of an anti-cancer therapy.

12. The method according to any one of claims 1 to 11, wherein the sample from the individual is a tumour tissue sample or a tumour fluid sample.

13. 13. The method of claim 12, wherein the sample is a formalin-fixed paraffin-embedded (FFPE) sample, an archived sample, a fresh sample, or a frozen sample.

14. 14. The method of claim 13, wherein the tumor tissue sample is a FFPE sample.

15. (a) the expression levels of the first and second members of the gene pair in the sample are protein expression levels; or (b) the expression levels of the first member and the second member of the gene pair in the sample are mRNA expression levels. The method according to any one of claims 12 to 14.

16. 16. The method of claim 15, wherein the expression levels of the first and second members of the gene pair in the sample are the mRNA expression levels of the first and second members of the gene pair, respectively.

17. 17. The method of claim 16, wherein the mRNA expression levels of the first and second members of the gene pair are determined by in situ hybridization (ISH), RNA-seq, RT-qPCR, qPCR, multiplex qPCR or RT-qPCR, microarray analysis, SAGE, MassARRAY technology, FISH, or a combination thereof.

18. 18. The method of claim 16 or 17, wherein the first reference expression level is between about 0.25 and about 0.5 counts per million (CPM) and the second reference expression level is between about 0.25 and about 0.5 CPM.

19. 20. The method of claim 18, wherein the first reference expression level is 0.25 CPM and the second reference expression level is 0.25 CPM.

20. 20. The method of any one of claims 1 to 19, wherein the first reference expression level and the second reference expression level are the expression levels of the first member and the second member of the gene pair, respectively, in a reference population of individuals with cancer.

21. The method of any one of claims 1 to 20, wherein the cancer is a urinary tract cancer.

22. 22. The method of claim 21, wherein the urinary tract cancer is a urinary tract cancer.

23. 23. The method of claim 22, wherein the urinary tract cancer is locally advanced urothelial carcinoma.

24. 23. The method of claim 22, wherein the urinary tract cancer is metastatic urothelial carcinoma (mUC).

25. 25. The method of any one of claims 1-3 and 6-24, wherein the benefit comprises an extension of the individual's overall survival (OS) compared to treatment without the PD-L1 axis binding antagonist.

26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is SIGLEC6 and the second member of the gene pair is NCR1.

27. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is BTN3A1 and the second member of the gene pair is LRRC4B.

28. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is CD80 and the second member of the gene pair is CTLA4.

29. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is BTN3A3 and the second member of the gene pair is LRRC4B.

30. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is EFNB1 and the second member of the gene pair is TRHDE.

31. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is CTLA4 and the second member of the gene pair is PCDHGB4.

32. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is CTLA4 and the second member of the gene pair is FAM200A.

33. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is CA12 and the second member of the gene pair is SIGLEC6.

34. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is ILDR2 and the second member of the gene pair is CLEC12B.

35. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is EFNB1 and the second member of the gene pair is ITLN1.

36. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is CADM1 and the second member of the gene pair is CRTAM.

37. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is CD79B and the second member of the gene pair is CD244.

38. 26. The method of any one of claims 1 to 5 and 9 to 25, wherein the first member of the gene pair is DAG1 and the second member of the gene pair is EFNB1.

39. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EFNB1 and the second member of the gene pair is EVC2.

40. 26. The method of any one of claims 6 to 25, wherein the first member of the gene pair is GPC4 and the second member of the gene pair is FGFRL1.

41. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB4.

42. 26. The method of any one of claims 6 to 25, wherein the first member of the gene pair is PTPRD and the second member of the gene pair is LRFN4.

43. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EFNB1 and the second member of the gene pair is AQPEP.

44. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EFNB1 and the second member of the gene pair is DSG4.

45. The method of any one of claims 6 to 25, wherein the first member of the gene pair is LDLR and the second member of the gene pair is LILRB5.

46. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB3.

47. 26. The method of any one of claims 6 to 25, wherein the first member of the gene pair is PLXNB3 and the second member of the gene pair is SEMA4G.

48. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EFNB1 and the second member of the gene pair is EPHB6.

49. The method of any one of claims 6 to 25, wherein the first member of the gene pair is FLT4 and the second member of the gene pair is FLRT2.

50. The method of any one of claims 6 to 25, wherein the first member of the gene pair is AXL1 and the second member of the gene pair is IL1RL1.

51. The method of any one of claims 6 to 25, wherein the first member of the gene pair is CD320 and the second member of the gene pair is IGSF5.

52. The method of any one of claims 6 to 25, wherein the first member of the gene pair is CD59 and the second member of the gene pair is STAB1.

53. 26. The method of any one of claims 6 to 25, wherein the first member of the gene pair is CNTN3 and the second member of the gene pair is PTPRG.

54. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EFNB1 and the second member of the gene pair is EPHA3.

55. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EFNB3 and the second member of the gene pair is EPHB2.

56. 26. The method of any one of claims 6 to 25, wherein the first member of the gene pair is EGF and the second member of the gene pair is TNFRSF11B.

57. The method of any one of claims 6 to 25, wherein the first member of the gene pair is ENPEP and the second member of the gene pair is SLITRK1.

58. 26. The method of any one of claims 6 to 25, wherein the first member of the gene pair is FCGR3B and the second member of the gene pair is EDA2R.

59. The method of any one of claims 6 to 25, wherein the first member of the gene pair is IL20RA and the second member of the gene pair is CLEC14A.

60. The method of any one of claims 6 to 25, wherein the first member of the gene pair is IL6R and the second member of the gene pair is BTNL9.

61. The method of any one of claims 6 to 25, wherein the first member of the gene pair is IZUMO1 and the second member of the gene pair is LILRA5.

62. The method of any one of claims 6 to 25, wherein the first member of the gene pair is NGFR and the second member of the gene pair is LRRTM3.

63. The method of any one of claims 6 to 25, wherein the first member of the gene pair is NTM and the second member of the gene pair is AMIGO2.

64. 26. The method of any one of claims 6 to 25, wherein the first member of the gene pair is PCDHB3 and the second member of the gene pair is IGSF11.

65. 26. The method of any one of claims 6 to 25, wherein the first member of the gene pair is PTGFRN and the second member of the gene pair is TMEM59L.

66. The method of any one of claims 6 to 25, wherein the first member of the gene pair is TREM1 and the second member of the gene pair is VSIG8.

67. 67. The method of any one of claims 1 to 66, wherein the PD-L1 axis binding antagonist is selected from the group consisting of a PD-L1 binding antagonist, a PD-1 binding antagonist, and a PD-L2 binding antagonist.

68. 68. The method of claim 67, wherein the PD-L1 axis binding antagonist is a PD-L1 binding antagonist.

69. 69. The method of claim 68, wherein the PD-L1 binding antagonist inhibits binding of PD-L1 to one or more of its ligand binding partners.

70. 69. The method of claim 68, wherein the PD-L1 binding antagonist inhibits binding of PD-L1 to PD-1.

71. 69. The method of claim 68, wherein the PD-L1 binding antagonist inhibits binding of PD-L1 to B7-1.

72. 69. The method of claim 68, wherein the PD-L1 binding antagonist inhibits binding of PD-L1 to both PD-1 and B7-1.

73. 73. The method of any one of claims 68 to 72, wherein the PD-L1 binding antagonist is an antibody or an antigen-binding fragment thereof.

74. 74. The method of claim 73, wherein the antibody is selected from the group consisting of atezolizumab, MDX-1105, MEDI4736 (durvalumab), and MSB0010718C (avelumab).

75. The anti-PD-L1 antibody comprises the following hypervariable regions: (a) the HVR-H1 sequence of GFTFSDSWIH (SEQ ID NO: 19); (b) the HVR-H2 sequence of AWISPYGGSTYYADSVKG (SEQ ID NO:20); (c) the HVR-H3 sequence of RHWPGGFDY (SEQ ID NO:21); (d) the HVR-L1 sequence of RASQDVSTAVA (SEQ ID NO:22); (e) the HVR-L2 sequence of SASFLYS (SEQ ID NO:23); and (f) the HVR-L3 sequence of QQYLYHPAT (SEQ ID NO: 24):

75. The method of claim 74, comprising:

76. The anti-PD-L1 antibody (a) a heavy chain variable (VH) domain comprising an amino acid sequence having at least 90% sequence identity to the amino acid sequence of SEQ ID NO:3; (b) a light chain variable (VL) domain comprising an amino acid sequence having at least 90% sequence identity to the amino acid sequence of SEQ ID NO:4; or (c) a VH domain according to (a) and a VL domain according to (b) 76. The method of claim 74 or 75, comprising:

77. The anti-PD-L1 antibody (a) a heavy chain variable (VH) domain comprising an amino acid sequence having at least 95% sequence identity to the amino acid sequence of SEQ ID NO:3; (b) a light chain variable (VL) domain comprising an amino acid sequence having at least 95% sequence identity to the amino acid sequence of SEQ ID NO:4; or (c) a VH domain according to (a) and a VL domain according to (b) 77. The method of claim 76, comprising:

78. The anti-PD-L1 antibody (a) a heavy chain variable (VH) domain comprising an amino acid sequence having at least 96% sequence identity to the amino acid sequence of SEQ ID NO:3; (b) a light chain variable (VL) domain comprising an amino acid sequence having at least 96% sequence identity to the amino acid sequence of SEQ ID NO:4; or (c) a VH domain according to (a) and a VL domain according to (b) 78. The method of claim 77, comprising:

79. The anti-PD-L1 antibody (a) a heavy chain variable (VH) domain comprising an amino acid sequence having at least 97% sequence identity to the amino acid sequence of SEQ ID NO:3; (b) a light chain variable (VL) domain comprising an amino acid sequence having at least 97% sequence identity to the amino acid sequence of SEQ ID NO:4; or (c) a VH domain according to (a) and a VL domain according to (b) 79. The method of claim 78, comprising:

80. The anti-PD-L1 antibody (a) a heavy chain variable (VH) domain comprising an amino acid sequence having at least 98% sequence identity to the amino acid sequence of SEQ ID NO:3; (b) a light chain variable (VL) domain comprising an amino acid sequence having at least 98% sequence identity to the amino acid sequence of SEQ ID NO:4; or (c) a VH domain according to (a) and a VL domain according to (b) 80. The method of claim 79, comprising:

81. The anti-PD-L1 antibody (a) a heavy chain variable (VH) domain comprising an amino acid sequence having at least 99% sequence identity to the amino acid sequence of SEQ ID NO:3; (b) a light chain variable (VL) domain comprising an amino acid sequence having at least 99% sequence identity to the amino acid sequence of SEQ ID NO:4; or (c) a VH domain according to (a) and a VL domain according to (b) 81. The method of claim 80, comprising:

82. The anti-PD-L1 antibody (a) a VH domain comprising the amino acid sequence of SEQ ID NO:3; and (b) a VL domain comprising the amino acid sequence of SEQ ID NO:4 82. The method of claim 81 , comprising:

83. The method of claim 82, wherein the anti-PD-L1 antibody is atezolizumab (MPDL3280A).

84. 68. The method of claim 67, wherein the PD-L1 axis binding antagonist is a PD-1 binding antagonist.

85. 85. The method of claim 84, wherein the PD-1 binding antagonist inhibits binding of PD-1 to one or more of its ligand binding partners.

86. 86. The method of claim 85, wherein the PD-1 binding antagonist inhibits binding of PD-1 to PD-L1.

87. 86. The method of claim 85, wherein the PD-1 binding antagonist inhibits binding of PD-1 to PD-L2.

88. 86. The method of claim 85, wherein the PD-1 binding antagonist inhibits binding of PD-1 to both PD-L1 and PD-L2.

89. 89. The method of any one of claims 85-88, wherein the PD-1 binding antagonist is an antibody or an antigen-binding fragment thereof.

90. 90. The method of claim 89, wherein the antibody is selected from the group consisting of MDX-1106 (nivolumab), MK-3475 (pembrolizumab), MEDI-0680 (AMP-514), PDR001, REGN2810, and BGB-108.

91. 89. The method of any one of claims 85-88, wherein the PD-1 binding antagonist is an Fc fusion protein.

92. A method of treating an individual having cancer comprising administering to the individual an effective amount of an agonist of CD177 activity.

93. A method for identifying an individual having cancer who may benefit from a treatment comprising an agonist of CD177 activity, the method comprising determining an expression level of podoplanin (PDPN) in a sample from the individual, wherein an expression level of PDPN in the sample that exceeds a reference PDPN expression level identifies the individual as an individual who may benefit from a treatment comprising an agonist of CD177 activity.

94. A method for selecting a therapy for an individual having cancer, comprising determining an expression level of PDPN in a sample from the individual, wherein an expression level of PDPN in the sample that exceeds a reference PDPN expression level identifies the individual as one who may benefit from treatment comprising an agonist of CD177 activity.

95. The method of claim 93 or 94, wherein the individual has an expression level of PDPN in the sample that is greater than a reference PDPN expression level, and the method further comprises administering to the individual an effective amount of an agonist of CD177 activity.

96. 1. A method of treating an individual having cancer, comprising: (a) determining an expression level of PDPN in a sample from the individual, wherein the expression level of PDPN in the sample is greater than a reference PDPN expression level; and (b) administering to the individual an effective amount of an agonist of CD177 activity. A method comprising:

97. A method for treating an individual having cancer, comprising administering to the individual an effective amount of an agonist of CD177 activity, wherein the expression level of PDPN in a sample from the individual has been determined to be greater than a reference PDPN expression level.

98. The method of any one of claims 92 to 97, wherein the CD177 activity is inhibition of PDPN.

99. The method of any one of claims 93 to 98, wherein the sample from the individual is a tumour tissue sample or a tumour fluid sample.

100. 100. The method of claim 99, wherein the tumor tissue sample is a formalin-fixed paraffin-embedded (FFPE) sample, an archived sample, a fresh sample, or a frozen sample.

101. The method according to any one of claims 93 to 100, wherein the expression level of PDPN in the sample is the protein expression level of PDPN or the RNA expression level of PDPN.

102. The method of claim 101, wherein the expression level of PDPN in the sample is an RNA expression level of PDPN.

103. 103. The method of claim 102, wherein the RNA expression level of PDPN is determined by in situ hybridization (ISH), RNA-seq, RT-qPCR, qPCR, multiplex qPCR or RT-qPCR, microarray analysis, SAGE, MassARRAY technology, FISH, or a combination thereof.

104. The method of any one of claims 93 to 103, wherein the reference PDPN expression level is the expression level of PDPN in a population of individuals with cancer.

105. 105. The method of claim 104, wherein the cancer is colorectal cancer (CRC), squamous cell carcinoma of the head and neck, or glioma.

106. The method of claim 104 or 105, wherein the reference PDPN expression level is the 50th percentile of the expression level in a population.

107. The method of claim 104 or 105, wherein the reference PDPN expression level is the 66th percentile of expression levels in a population.

108. The method of any one of claims 93 to 107, wherein the reference PDPN expression level is a pre-assigned PDPN expression level.

109. The method of any one of claims 92 to 108, wherein the cancer is CRC, squamous cell carcinoma of the head and neck, or glioma.

110. 110. The method of claim 109, wherein the cancer is CRC.

111. 111. The method of claim 110, wherein the CRC is stage II CRC.

112. The method of claim 110, wherein the CRC is stage IV CRC.

113. 13. The method of any one of claims 93-95 and 98-112, wherein the benefit comprises an extension of the individual's recurrence-free survival (RFS) compared to treatment without an agonist of CD177 activity.

114. The method of any one of claims 92 to 113, wherein the agonist of CD177 activity results in increased binding of PDPN to CD177 compared to binding of the two proteins in the absence of the agonist.

115. The method of any one of claims 92 to 114, wherein the agonist of CD177 activity results in a change in downstream activity of PDPN compared to the downstream activity in the absence of the agonist of CD177 activity.

116. The method of claim 115, wherein the change in downstream activity is a decrease in tumor growth.

117. The method of claim 115, wherein the alteration in downstream activity is a decrease in cancer-associated fibroblast (CAF) contractility.

118. The method of any one of claims 92 to 117, wherein the agonist of CD177 activity is a small molecule, an antibody or antigen-binding fragment thereof, a peptide, or a mimetic.

119. The method of claim 118, wherein the agonist of CD177 activity is a peptide.

120. The method of claim 119, wherein the peptide is a CD177 peptide.

121. The method of claim 120, wherein the CD177 peptide is the extracellular domain of CD177.

122. The method of claim 121, wherein the peptide is multimerized.

123. 123. The method of claim 122, wherein the peptide is tetramerized.

124. 124. The method of claim 123, wherein the peptide is tetramerized using streptavidin.

125. The method of claim 118, wherein the agonist of CD177 activity is an antibody or an antigen-binding fragment thereof.

126. The method of claim 125, wherein the antibody or antigen-binding fragment thereof binds to PDPN.

127. The method of claim 126, wherein the antibody or antigen-binding fragment thereof is an antagonist antibody or antigen-binding fragment thereof.

128. The method of claim 125, wherein the antibody or antigen-binding fragment thereof binds to CD177.

129. The method of claim 128, wherein the antibody or antigen-binding fragment thereof is an agonist antibody or antigen-binding fragment thereof.

130. The antigen-binding fragment is selected from the group consisting of bis-Fab, Fv, Fab, Fab'-SH, and F(ab'). ２ , a diabody, a linear antibody, an scFv, an scFab, a VH domain, or a VHH domain.

131. The method of any one of claims 1 to 130, wherein the individual is a human.

132. A collection of polypeptides, each polypeptide comprising an extracellular domain, a tag, and an anchor, wherein the collection of polypeptides comprises at least 81% of the extracellular domains of the proteins of Table 7.

133. 133. The method of claim 132, wherein the collection of polypeptides comprises at least 85% of the extracellular domains of the proteins in Table 7.

134. 134. The method of claim 133, wherein the collection of polypeptides comprises at least 90% of the extracellular domains of the proteins in Table 7.

135. 135. The method of claim 134, wherein the collection of polypeptides comprises at least 95% of the extracellular domains of the proteins in Table 7.

136. 136. The method of claim 135, wherein the collection of polypeptides comprises the extracellular domains of all of the proteins in Table 7.

137. 137. The collection of polypeptides according to any one of claims 132 to 136, wherein the anchor is capable of anchoring the extracellular domain to the surface of the plasma membrane of a cell.

138. The collection of polypeptides according to any one of claims 132 to 137, wherein the anchor is a glycosylphosphatidylinositol (GPI) polypeptide.

139. A collection of polypeptides according to any one of claims 132 to 138, wherein the tag can be directly or indirectly visualized.

140. 140. The collection of polypeptides of claim 139, wherein the tag comprises a moiety that can be detected using an antibody or an antibody fragment.

141. 141. The collection of polypeptides described in claim 139 or 140, wherein the tag is a glycoprotein D (gD) polypeptide.

142. 140. The collection of polypeptides of claim 139, wherein the tag comprises a fluorescent protein.

143. 143. The collection of polypeptides according to any one of claims 132 to 142, wherein the extracellular domain has a native conformation.

144. 144. The collection of polypeptides according to any one of claims 132 to 143, wherein the extracellular domain comprises a naturally occurring post-translational modification.

145. The collection of polypeptides according to any one of claims 137 to 144, wherein the cells are mammalian cells.

146. The collection of polypeptides described in claim 145, wherein the cells are COS7 cells.

147. A collection of polypeptides according to any one of claims 137 to 146, wherein cells are transiently transfected with plasmids encoding the polypeptides.

148. A collection of vectors encoding the collection of polypeptides of any one of claims 132-147.

149. A collection of cells comprising the collection of vectors of claim 148.

150. 150. The collection of cells according to claim 149, wherein a plurality of cells are capable of expressing at least one polypeptide according to any one of claims 132 to 147, and optionally different cells express different polypeptides.

151. 148. The collection of polypeptides of any one of claims 132-147, wherein one or more of said polypeptides are each immobilized at a distinct location on one or more solid surfaces.

152. 1. A method for identifying protein-protein interactions, comprising: (a) providing a collection of polypeptides according to any one of claims 132 to 147, optionally wherein said polypeptides are immobilised on one or more solid surfaces; (b) contacting the collection of step (a) with a multimerized query protein under conditions that allow binding of the multimerized query protein to at least one of the extracellular domains of the polypeptide; and (c) detecting an interaction between the multimerized query protein and the at least one extracellular domain, thereby identifying a protein-protein interaction. The method includes:

153. 153. The method of claim 152, wherein one or more of said polypeptides are each immobilized at a distinct location on said one or more solid surfaces.

154. 154. The method of claim 152 or 153, wherein the distinct location comprises a cell that displays the polypeptide.

155. 155. The method of claim 154, wherein the cell is a mammalian cell.

156. 156. The method of any one of claims 152 to 155, wherein the contacting step is semi-automated.

157. 157. The method of any one of claims 152 to 156, wherein detecting the interaction comprises detecting a signal, optionally a fluorescent signal, above a threshold level at a location on the solid surface.

158. 158. The method of claim 157, wherein the detection is automated.

159. 159. The method of any one of claims 152 to 158, wherein the interaction is a transient interaction.

160. 160. The method of any one of claims 152 to 159, wherein the interaction is a low affinity interaction.

161. 161. The method of claim 160, wherein the low affinity interaction is a micromolar affinity interaction.

162. The method of any one of claims 152 to 161, wherein the multimerized query protein is a dimerized, trimerized, tetramerized, or pentamerized query protein.

163. The method of claim 162, wherein the multimerized query protein is a tetramerized query protein.

164. The method of any one of claims 152 to 163, wherein the multimerized query protein comprises an isolated extracellular domain of the query protein.

165. The method of claim 164, wherein the isolated extracellular domain of the query protein is biotinylated and conjugated to fluorescent streptavidin to tetramerize the query protein.

166. 1. A method for identifying a modulator of an interaction between a protein of Table 1 and a protein of Table 2, comprising: (a) providing candidate modulators; (b) contacting a protein of Table 1 with a protein of Table 2 in the presence or absence of a candidate modulator under conditions that allow binding of the protein of Table 1 to the protein of Table 2, where the proteins of Table 1 and the proteins of Table 2 are reported to interact in Table 3; and (c) measuring binding of the protein of Table 1 to the protein of Table 2, wherein an increase or decrease in binding in the presence of the candidate modulator compared to binding in the absence of the candidate modulator identifies the candidate modulator as a modulator of the interaction between the protein of Table 1 and the protein of Table 2. The method includes:

167. 1. A method for identifying a modulator of a downstream activity of a protein of Table 1, comprising: (a) providing candidate modulators; (b) contacting a protein of Table 1 with a protein of Table 2 in the presence or absence of a candidate modulator under conditions that allow binding of the protein of Table 1 to the protein of Table 2, where the proteins of Table 1 and the proteins of Table 2 are reported to interact in Table 3; and (c) measuring a downstream activity of the protein of Table 1, wherein a change in the downstream activity in the presence of the candidate modulator compared to the downstream activity in the absence of the candidate modulator identifies the candidate modulator as a modulator of the downstream activity of the protein of Table 1. The method includes:

168. 1. A method for identifying a modulator of a downstream activity of a protein of Table 2, comprising: (a) providing candidate modulators; (b) contacting a protein of Table 2 with a protein of Table 1 in the presence or absence of a candidate modulator under conditions that allow binding of the protein of Table 2 to the protein of Table 1, where the proteins of Table 1 and the proteins of Table 2 are reported to interact in Table 3; and (c) measuring a downstream activity of a protein of Table 2, wherein a change in the downstream activity in the presence of the candidate modulator compared to the downstream activity in the absence of the candidate modulator identifies the candidate modulator as a modulator of the downstream activity of a protein of Table 2. The method includes:

169. 167. The method of claim 166, wherein the increase or decrease in binding is at least 70% as measured by surface plasmon resonance, biolayer interferometry, or enzyme-linked immunosorbent assay (ELISA).

170. 169. The method of claim 167 or 168, wherein the modulator is an inhibitor of a downstream activity of a protein of Table 1 or Table 2.

171. 169. The method of claim 167 or 168, wherein the modulator is an activator of a downstream activity of a protein of Table 1 or Table 2.

172. The method of claim 167 or 168, wherein the change in downstream activity is a decrease in the amount, intensity, or duration of the downstream activity.

173. The method of claim 167 or 168, wherein the change in downstream activity is an increase in the amount, intensity, or duration of the downstream activity.

174. 169. The method of any one of claims 166 to 168, wherein the modulator is a small molecule, an antibody or antigen-binding fragment thereof, a peptide, a mimetic, an antisense oligonucleotide, or a small interfering RNA (siRNA).

175. The antigen-binding fragment is selected from the group consisting of bis-Fab, Fv, Fab, Fab'-SH, and F(ab'). ２ , a diabody, a linear antibody, an scFv, an scFab, a VH domain, or a VHH domain.

176. 175. The method of claim 174, wherein the antibody or antigen-binding fragment thereof binds to a protein of Table 1.

177. 175. The method of claim 174, wherein the antibody or antigen-binding fragment thereof binds to a protein of Table 2.

178. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is podoplanin (PDPN).

179. 179. The method of claim 178, wherein the protein of Table 2 is CD177.

180. The method of claim 178 or 179, wherein the downstream activity is cancer-associated fibroblast (CAF) contractility.

181. The method of claim 180, wherein CAF contractility is decreased in the presence of a modulator.

182. The method of claim 181, wherein CAF contractility is reduced by at least 20% as measured by a gel contraction assay.

183. The method of claim 182, wherein CAF contractility is reduced by at least 20% as measured in a 3D gel stretch assay.

184. 180. The method of claim 178 or 179, wherein the downstream activity is tumor proliferation.

185. The method of claim 184, wherein tumor growth is reduced in the presence of the modulator.

186. 186. The method of claim 185, wherein tumor growth is reduced by at least 20% as measured in a tumor growth assay.

187. 187. The method of any one of claims 178-186, wherein the modulator is an antibody or antigen-binding fragment thereof that targets PDPN.

188. 187. The method of any one of claims 178 to 186, wherein the modulator is an antibody or antigen-binding fragment thereof that targets CD177.

189. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is PD-L1 (CD274).

190. 190. The method of claim 189, wherein the protein of Table 2 is EPHA3.

191. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is PD-L2 (PDCD1LG2).

192. 192. The method of claim 191, wherein the protein of Table 2 is CEACAM4, ICAM5, NECTIN3, PSG9, or TNFRSF11A.

193. 193. The method of claim 192, wherein the protein of Table 2 is CEACAM4.

194. The method of any one of claims 189 to 193, wherein the downstream activity is immune checkpoint inhibition.

195. The method of claim 194, wherein immune checkpoint inhibition is decreased in the presence of a modulator.

196. The method of claim 195, wherein immune checkpoint inhibition is reduced by at least 30% as measured in a cell-based assay.

197. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is PTPRD.

198. The method of claim 197, wherein PTPRD comprises G203E and K204E; R232C and R233C; P249L; G285E; E406K; S431L; R561Q; P666S; E755K; V892I; S912F; R995C; or R1088C amino acid substitution mutation or a ΔG61ΔE106 amino acid deletion mutation.

199. 200. The method of claim 197 or 198, wherein the protein of Table 2 is BMP5, CEACAM3, IL1RAP, IL1RAPL2, LECT1, LRFN5, SIRPG, SLITRK3, SLITRK4, SLITRK6, or TGFA.

200. 200. The method of any one of claims 197 to 199, wherein the downstream activity is inhibition of cell proliferation.

201. The method of claim 200, wherein inhibition of cell proliferation is increased in the presence of a modulator.

202. The method of claim 201, wherein the inhibition of cell proliferation is increased by at least 30% as measured by a colony formation assay.

203. 200. The method of any one of claims 197 to 199, wherein the downstream activity is inhibition of STAT3 phosphorylation.

204. The method of claim 203, wherein inhibition of STAT3 phosphorylation is increased in the presence of a modulator.

205. The method of claim 204, wherein the inhibition of STAT3 phosphorylation is increased by at least 30% as measured by Western blot of phosphorylated STAT3.

206. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is a PTPRS.

207. 207. The method of claim 206, wherein the protein of Table 2 is C6orf25, IL1RAP, IL1RAPL1, IL1RAPL2, LRFN1, LRFN5, LRRC4B, NCAM1, SLITRK1, SLITRK2, SLITRK3, SLITRK4, or SLITRK6.

208. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is PTPRF.

209. 209. The method of claim 208, wherein the protein in Table 2 is CD177, IL1RAP, or LRFN5.

210. The method of any one of claims 206 to 209, wherein the downstream activity is inhibition of cell migration.

211. The method of claim 210, wherein inhibition of cell migration is increased in the presence of a modulator.

212. The method of claim 211, wherein inhibition of cell migration is increased by at least 20%.

213. The method of any one of claims 206 to 209, wherein the downstream activity is phosphorylation of EGFR.

214. The method of claim 213, wherein phosphorylation of EGFR is decreased in the presence of the modulator.

215. 215. The method of claim 214, wherein EGFR phosphorylation is reduced by at least 30% as measured by an EGFR phosphorylation assay.

216. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is CHL1.

217. 217. The method of claim 216, wherein the protein of Table 2 is CNTN1, CNTN5, SIRPA, L1CAM, or TMEM132A.

218. The method of claim 216 or 217, wherein the downstream activity is inhibition of tumorigenesis.

219. The method of claim 218, wherein inhibition of tumor formation is increased in the presence of a modulator.

220. The method of claim 219, wherein inhibition of tumor formation is increased by at least 20%.

221. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is CNTN1.

222. 222. The method of claim 221, wherein the protein of Table 2 is CDH6, CHL1, FCGRT, PCDHB7, or SGCG.

223. The method of claim 221 or 222, wherein the downstream activity is cell proliferation or cell invasion.

224. 224. The method of claim 223, wherein cell proliferation or cell invasion is reduced in the presence of the modulator.

225. 225. The method of claim 224, wherein cell proliferation is reduced by at least 20% as measured by a colony formation assay.

226. 225. The method of claim 224, wherein cell invasion is reduced by at least 20% as measured by a gel invasion assay.

227. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is LILRB1.

228. 228. The method of claim 227, wherein the protein of Table 2 is CLEC6A, CXADR, EDAR, FLT4, IL6R, ILDR1, or LILRA5.

229. The method of claim 227 or 228, wherein the downstream activity is inhibition of phagocytosis.

230. 230. The method of claim 229, wherein the inhibition of phagocytosis is decreased in the presence of a modulator.

231. The method of claim 230, wherein inhibition of phagocytosis is reduced by at least 20%.

232. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is LILRB2.

233. 233. The method of claim 232, wherein the protein of Table 2 is IGSF8 or MOG.

234. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is LILRB3.

235. 235. The method of claim 234, wherein the protein of Table 2 is LRRC15 or LY6G6F.

236. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is LILRB4.

237. 237. The method of claim 236, wherein the protein of Table 2 is CNTFR.

238. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is LILRB5.

239. 239. The method of claim 238, wherein the protein of Table 2 is APLP2, CD177, CLEC10A, CLECSF13, LDLR, PILRA, or UNC5C.

240. 240. The method of claim 239, wherein the protein of Table 2 is LDLR.

241. The method of any one of claims 227, 228, and 234-240, wherein the downstream activity is osteoclast differentiation.

242. The method of claim 241, wherein osteoclast differentiation is reduced by at least 20% in the presence of the modulator.

243. The method of claim 242, wherein osteoclast differentiation is measured by a TRAP+ multinucleated cell assay.

244. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is AXL.

245. 245. The method of claim 244, wherein the protein of Table 2 is IL1RL1 or VSIG10L.

246. The method of claim 244 or 245, wherein the downstream activity is activation of the RAS / RAF / MAPK / ERK1 / 2 pathway, activation of the JAK / STAT pathway, activation of the PI3K signaling pathway, cell migration, formation of filopodia, or phosphorylation of AXL.

247. 247. The method of claim 246, wherein cell migration is reduced by at least 20% as measured by a gel invasion assay.

248. 169. The method of any one of claims 166-168, wherein the protein of Table 1 is LRRC4B.

249. 249. The method of claim 248, wherein the protein of Table 2 is BTN3A1 or BTN3A3.

250. 250. The method of any one of claims 166-249, wherein expression of a protein of Table 1 or a protein of Table 2 is upregulated or downregulated in tumor tissue compared to healthy tissue.

251. 1. An isolated modulator of the interaction between a protein of Table 1 and a protein of Table 2, comprising: i. a protein in Table 1 and a protein in Table 2 are reported to interact in Table 3; and ii. the modulator causes an increase or decrease in binding of a protein of Table 1 to a protein of Table 2 compared to binding in the absence of the modulator. Isolated modulators.

252. 1. An isolated modulator of a downstream activity of a protein of Table 1 or a protein of Table 2, comprising: i. a protein in Table 1 and a protein in Table 2 are reported to interact in Table 3; and ii. the modulator causes a change in the downstream activity of a protein of Table 1 or a protein of Table 2 compared to the downstream activity in the absence of the modulator. Isolated modulators.

253. The modulator of claim 251, wherein the increase or decrease in binding is at least 70% as measured by surface plasmon resonance, biolayer interferometry, or enzyme-linked immunosorbent assay (ELISA).

254. 253. The modulator of claim 251 or 252, wherein the modulator is an inhibitor of a downstream activity of a protein of Table 1 or Table 2.

255. 253. The modulator of claim 251 or 252, wherein the modulator is an activator of a downstream activity of a protein of Table 1 or Table 2.

256. The modulator of claim 252, wherein the change in downstream activity is a decrease in the amount, intensity, or duration of the downstream activity.

257. The modulator of claim 252, wherein the change in downstream activity is an increase in the amount, intensity, or duration of the downstream activity.

258. The modulator of any one of claims 251 to 257, wherein the modulator is a small molecule, an antibody or antigen-binding fragment thereof, a peptide, or a mimetic.

259. The antigen-binding fragment is selected from the group consisting of bis-Fab, Fv, Fab, Fab'-SH, and F(ab'). ２ , a diabody, a linear antibody, an scFv, an scFab, a VH domain, or a VHH domain.

260. The modulator of claim 259, wherein the antibody or antigen binding fragment thereof binds to a protein of Table 1.

261. The modulator of claim 259, wherein the antibody or antigen binding fragment thereof binds to a protein of Table 2.

Images