Meso-scale engineered peptides and methods of selecting

Abstract

To provide engineered peptides comprising a combination of spatially-associated topological constraints, and methods of selecting the engineered peptides.SOLUTION: A method of selecting an engineered peptide, comprising: identifying one or more topological characteristics of a reference target; designing spatially-associated constraints for each topological characteristic so as to produce a combination of spatially-associated topological constraints derived from the reference target; comparing spatially-associated topological characteristics of candidate peptides with the combination of spatially-associated topological constraints derived from the reference target; and selecting a candidate peptide having spatially-associated topological characteristics that overlap with the combination of spatially-associated topological constraints derived from the reference target, thereby producing the engineered peptide.SELECTED DRAWING: None

Description

[Background technology]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application was filed on May 31, 2019, and is entitled "Mesoscale Engineered Peptides and This application claims priority to and benefit of U.S. patent application Ser. No. 62 / 855,767, entitled "Selection Method and Method for Selecting a Target." No. 6,239,999, which is incorporated herein by reference in its entirety.

[0002] The majority of basic research in the therapeutic field is focused on new peptide therapeutics or new peptide immunotherapeutics. Identifying novel molecules with desirable properties, such as antigens (from which to develop new therapeutic antibodies) However, the standard molecular discovery paradigm is promising. It relies on random sampling using stochastic processes to identify suitable functional molecules. These molecular candidates are then selected to demonstrate the desired activity, function, pharmacokinetics, and and / or undergo multiple evaluations and tests in the hope that they possess other desired characteristics. The system begins with the screening of a random group, often with one or more necessary Therefore, what is needed is a combination of computational, chemical, and and methods for developing engineered peptides that incorporate elements of biological design. Summary of the Invention

[0003] In some aspects, engineered peptides are provided herein, and the engineered peptides have molecular weights between 1 kDa and 10 kDa, contain up to 50 amino acids, and are spatially related. A combination of topological constraints on the target, one or more of which are derived from the reference target. Thus, 10%–98% of the amino acids in the engineered peptides are constrained by one or more reference targets. and amino acids that satisfy one or more of the constraints from the reference target have a distance of less than 8.0 Å from the reference target. They have a root mean square deviation (RSMD) structural identity of .

[0004] In some embodiments, amino acids that satisfy constraints from one or more reference targets are In some embodiments, they have between 10% and 90% sequence identity with the target. 2 ~3000Å 2 In certain embodiments, the surface area of the surface of the material is The combination includes constraints derived from at least two or at least five reference targets. In some embodiments, the combination of constraints includes one or more constraints that do not originate from the reference target. In some embodiments, the constraints from one or more non-reference targets include desired structural, Describe kinetic, chemical, or functional properties, or any combination thereof In still further embodiments, the one or more constraints are independently a biological response or a biological In some embodiments, the polypeptide is associated with a biological response or biological function. At least a portion of the atoms in the engineered peptide are in the beta-sheet or alpha-sheet. The target is topologically constrained by secondary structure elements in the reference target, such as helices.

[0005] In another aspect, provided herein is a method for selecting an engineered peptide, the method comprising: Identifying one or more topological characteristics of the reference target; Each position is then calculated to generate a set of spatially related topological constraints derived from the reference target. Designing spatially related constraints on topological properties; The spatially related topological features of the candidate peptides are compared with the spatially related topological features derived from the reference target. Comparing topological constraint combinations and Spatially related topological constraints that overlap with the set of spatially related topological constraints derived from the reference target. Selecting candidate peptides with related topological properties to generate engineered peptides. and,

[0006] In some embodiments, the overlap between each feature is independently calculated as a total topological constraint distance (TCD) , Topological Clustering Coefficient (TCC), Euclidean distance, Power distance, Soager distance distance, Canberra distance, Sorensen distance, Jaccard distance, Mahalanobis distance, Hamin Quantitative Estimate of L by one or more of the following parameters: In certain embodiments, the mean percent error (MPE) is 75% or less as determined by the The constraints are per-residue energy, per-residue interaction, per-residue variation, Atomic distances per residue, chemical descriptors per residue, solvent exposure per residue, amino acid sequence similarity, bioinformatic descriptors per residue, non-covalent bond propensity per residue, Phi / psi angle per group, van der Waals radius per residue, secondary structure per residue These are derived from structural propensity, amino acid adjacency per residue, or amino acid contact per residue. In some embodiments, the properties of one or more candidate peptides are determined by computer simulation. In still further embodiments, the one or more constraints are independently determined by a biological In some embodiments, the biological response or function is associated with At least a portion of the atoms in the engineered peptides that are relevant to the biological function are beta- - topologically constrained to secondary structure elements in the reference target, such as a sheet or alpha helix do.

[0007] In still further aspects, compositions comprising two or more selection-derived polypeptides are provided herein. Each polypeptide is independently a positive selection molecule containing one or more positive induction properties, or or a negative selection molecule containing one or more negative inductive properties, each type of property being independently Amino acid sequence, polypeptide secondary structure, molecular dynamics, chemical characteristics, biological function, immunogenicity , multispecificity of reference target(s), cross-species reference target reactivity, unwanted reference target(s) Selectivity of the desired reference target(s) relative to the desired reference target(s), sequence and / or structural homology Selectivity of reference target(s) within a family, reference targets with similar protein functions ( Selectivity of multiple targets, such as undesired targets with high sequence and / or structural homology Selectivity of distinct desired reference targets(s) from a larger family, distinct reference targets Selectivity for alleles or mutations, selectivity for chemical modifications at the distinct reference target residue level Selectivity, cell type selectivity, tissue type selectivity, tissue environment selectivity, reference standard tolerance to structural diversity of the target(s), tolerance to sequence diversity of the reference target(s) and tolerance to kinetic diversity of the reference target(s), At least one of the above polypeptides may be an engineered peptide as described herein. It is Do.

[0008] In some embodiments, at least one of the two or more polypeptides is a positively selected polypeptide. and at least one of the two or more polypeptides is a negative selection molecule. In some embodiments, at least one of the two or more polypeptides is a naturally occurring polypeptide. In certain embodiments, corresponding proteins include at least one shared characteristic type. at least one pair of positive selection molecules and negative selection molecules, The negative selection molecule comprises a negative characteristic.

[0009] In yet additional embodiments, a composition comprising two or more selection-inducing molecules described herein is Provided herein is a method for screening a library of binding molecules using subjecting the pool of candidate binding molecules to at least one selection round, Wound is A negative selection step is performed to screen at least a portion of the pool against the negative selection molecule. Tep and A positive selection step is performed to screen at least a portion of the pool in favor of a positive selection molecule. and The order of selection steps within each round, and the order of rounds, may differ from the alternative order. This results in the selection of a subset of the pool.

[0010] In some embodiments, the library of binding molecules is a phage library, or a B cell library. In some embodiments, the library is a cell library, such as a cell library or a T cell library. In some embodiments, the method comprises two or more selection rounds or three or more selection rounds. In embodiments, each round comprises a different set of selected molecules. At least two rounds use the same negative selection molecule, or the same positive selection molecule, or both. In some embodiments, the number of selection rounds is increased from one selection round before proceeding to the next selection round. The method involves analyzing a subset of the pool obtained from the analysis.

[0011] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Patent and Trademark Office upon request and payment of the necessary fee. This application can be understood by reference to the following description in conjunction with the accompanying drawings. Aspects of the present invention include the following: (1) An engineered peptide, the engineered peptide having a molecular weight of 1 kDa to 10 kDa and containing up to 50 amino acids, the engineered peptide comprising: a combination of spatially related topological constraints, one or more of which are derived from a reference target; between 10% and 98% of the amino acids of the engineered peptide satisfy the constraints from the one or more reference targets; an engineered peptide, wherein the amino acids that satisfy the constraints from the one or more reference targets have a backbone root mean square deviation (RSMD) structural identity with the reference target of less than 8.0 Å; (2) the engineered peptide according to (1) above, wherein the amino acids satisfying the constraints from the one or more reference targets have 10% to 90% sequence identity with the reference targets; (3) the amino acids that satisfy the constraints from the one or more reference targets are within 30 Å 2 ~3000Å 2 the engineered peptide of (1) or (2) above, having a van der Waals surface area overlap with said reference of (4) The engineered peptide according to any one of (1) to (3) above, wherein the combination includes constraints derived from at least two reference targets; (5) The engineered peptide according to any one of (1) to (4) above, wherein the combination includes constraints derived from at least five reference targets; (6) The engineered peptide according to any one of (1) to (5) above, wherein the combination of constraints includes one or more constraints that are not derived from the reference target; (7) The engineered peptide of (6) above, wherein the one or more non-reference target-derived constraints describe desired structural, kinetic, chemical, or functional properties, or any combination thereof; (8) The engineered peptide according to any one of (1) to (7) above, wherein the constraints are independently selected from the group consisting of: interatomic distance, Atomic fluctuations, atomic energy, chemical descriptors, solvent exposure, amino acid sequence similarity, bioinformatics descriptors, non-covalent tendency, Phi angle, Psi angle, van der Waals radius, secondary structure tendency, amino acid contiguity, and amino acid contacts; (9) The engineered peptide according to any one of (1) to (8) above, wherein one or more constraints are independently atomic variations; (10) The engineered peptide according to any one of (1) to (9) above, wherein one or more constraints are independently chemical descriptors; (11) The engineered peptide according to any one of (1) to (10) above, wherein one or more constraints are independently interatomic distances; (12) The engineered peptide according to any one of (1) to (11) above, wherein one or more constraints are independently a secondary structure; (13) The engineered peptide according to any one of (1) to (12) above, wherein one or more constraints are independently a van der Waals surface; (14) The engineered peptide according to any one of (1) to (13) above, wherein one or more constraints independently relate to a biological response or biological function; (15) The engineered peptide according to any one of (1) to (14) above, which comprises one or more atoms associated with a biological response or biological function; (16) The engineered peptide according to any one of (1) to (15) above, which comprises one or more amino acids associated with a biological response or biological function; (17) The biological response or biological function is selected from the group consisting of gene expression, metabolic activity, protein expression, cell proliferation, cell death, cytokine secretion, kinase activity, epigenetic modification, cell death activity, inflammatory signaling, chemotaxis, tissue infiltration, immune cell lineage commitment, tissue microenvironment modification, immune synapse formation, IL-2 secretion, IL-10 secretion, growth factor secretion, interferon gamma secretion, transforming growth factor beta secretion, immunoreceptor tyrosine-based activation motif activity, immunoreceptor tyrosine-based inhibitory motif activity, antibody-dependent cellular cytotoxicity, complement-dependent cytotoxicity, biological pathway agonism, and biological pathway. the engineered peptide according to any one of (14) to (16) above, selected from the group consisting of antagonistic action, biological pathway redirection, kinase cascade modification, protein degradation pathway modification, protein homeostasis pathway modification, protein folding / pathway, post-translational modification pathway, metabolic pathway, gene transcription / translation, mRNA degradation pathway, gene methylation / acetylation pathway, histone modification pathway, epigenetic pathway, immune-dependent clearance, opsonization, hormone signaling, integrin pathway, membrane protein signaling, ion channel flux, and g-protein coupled receptor response; (18) The reference target comprises one or more atoms associated with a biological response or biological function; The engineered peptide according to (15) above, wherein the atomic variation of the one or more atoms in the engineered peptide associated with a biological response or biological function overlaps with the atomic variation of the one or more atoms in the reference target associated with a biological response or biological function; (19) The engineered peptide according to (18) above, wherein the overlap is a root mean square dot product (RMSIP) greater than 0.25; (20) The engineered peptide according to (19) above, wherein the overlap has a root mean square dot product (RMSIP) greater than 0.75. (21) The engineered peptide according to any one of (18) to (20) above, wherein at least a portion of the atoms in the engineered peptide associated with a biological response or biological function are topologically constrained to secondary structure elements in the reference target; (22) The engineered peptide according to (21) above, wherein the secondary structure element is a beta-sheet; (23) The engineered peptide according to (21) above, wherein the secondary structure element is an alpha helix; (24) The engineered peptide according to (21), wherein the secondary structure element is a turn, the turn comprising 2 to 7 residues and comprising at least one inter-residue hydrogen bond; (25) The engineered peptide according to (21) above, wherein the secondary structure element is a coil, and the coil comprises 2 to 20 residues; (26) The engineered peptide according to (25), wherein the coil does not contain any inter-residue hydrogen bonds; (27) The engineered peptide according to any one of (21) to (26), wherein at least a portion of the atoms in the engineered peptide associated with a biological response or function are topologically constrained to a combination of two or more secondary structure elements independently selected from the group consisting of a beta-sheet, an alpha-helix, a turn, and a coil; (28) The engineered peptide according to any one of (1) to (27) above, wherein one or more spatially related topological constraints are interatomic distances; (29) The engineered peptide according to any one of (1) to (28) above, wherein one or more spatially related topological constraints are atomic energies; (30) The engineered peptide according to (29) above, wherein each atomic energy is independently a pairwise attractive energy between two atoms, a pairwise repulsive energy between two atoms, an atomic-level solvation energy, a pairwise charge attractive energy between two atoms, a pairwise hydrogen bond attractive energy between two atoms, or a non-covalent bond energy; (31) The engineered peptide according to any one of (1) to (30) above, wherein one or more spatially related topological constraints are chemical descriptors; (32) The engineered peptide according to (31) above, wherein each chemical descriptor is independently hydrophobicity, polarity, volume, net charge, logP, high performance liquid chromatography retention, or van der Waals radius; (33) The engineered peptide according to any one of (1) to (32) above, wherein one or more spatially related topological constraints are bioinformatic descriptors; (34) The engineered peptide according to (33) above, wherein each bioinformatic descriptor is independently BLOSUM similarity, pKa, zScale, Cruciani property, Chidera factor, VHSE scale, ProtFP, MS-WHIM score, T scale, ST scale, transmembrane propensity, protein-buried region, helix propensity, sheet propensity, coil propensity, turn propensity, immunogenic propensity, antibody epitope occurrence, or protein interface occurrence; (35) The engineered peptide according to any one of (1) to (34) above, wherein one or more spatially related topological constraints are solvent exposure; (36) The engineered peptide according to any one of (1) to (35) above, wherein at least one of the constraints derived from the one or more reference targets is a GPCR extracellular domain; (37) The engineered peptide according to any one of (1) to (36) above, wherein at least one of the constraints derived from the one or more reference targets is an ion channel extracellular domain; (38) The engineered peptide according to any one of (1) to (37) above, wherein at least one of the constraints derived from the one or more reference targets is a protein-protein or peptide-protein interface bond; (39) The engineered peptide according to any one of (1) to (38) above, wherein at least one of the constraints derived from the one or more reference targets is derived from a polymorphic region of the target; (40) The engineered peptide according to any one of (1) to (39) above, comprising one or more atoms associated with a biological response or function, each of said one or more atoms independently selected from the group consisting of carbon, oxygen, nitrogen, hydrogen, sulfur, phosphorus, sodium, potassium, zinc, manganese, magnesium, copper, iron, molybdenum, and nickel; (41) The engineered peptide according to any one of (1) to (40) above, comprising one or more amino acids associated with a biological function or biological response, each of which is independently a naturally occurring proteinogenic amino acid, a naturally occurring non-proteinogenic amino acid, or a chemically synthesized non-natural amino acid; (42) The engineered peptide according to any one of (1) to (41) above, wherein the engineered peptide has at least one structural difference compared to the reference target; (43) The engineered peptide of (42), wherein the at least one structural difference is independently selected from the group consisting of sequence, number of amino acid residues, total number of atoms, total hydrophilicity, total hydrophobicity, total positive charge, total negative charge, one or more secondary structures, shape factors, Zernike descriptors, van der Waals surface, structure graph nodes and edges, volume surface, electrostatic potential surface, hydrophobic potential surface, local diameter, local surface features, backbone model, charge density, hydrophilic density, surface-to-volume ratio, amphiphilicity density, and surface roughness; (44) The engineered peptide according to (16) above, wherein the difference in one or more secondary structures is the presence of one or more additional secondary structure elements in the engineered peptide compared to the reference target, each additional secondary structure element being independently selected from the group consisting of an alpha helix, a beta-sheet, a loop, a turn, and a coil; (45) The engineered peptide according to any one of (1) to (44) above, wherein 10% to 90% of the amino acids satisfy topological constraints derived from one or more non-reference targets; (46) The engineered peptide according to (45), wherein the topological constraints derived from the one or more non-reference targets reinforce a pre-specified function; (47) Non-reference-derived topological constraints reinforce or stabilize secondary structure elements in the reference-derived fraction of the peptide. non-reference-derived topological constraints enforce atomic variations in the reference-derived fraction of the peptide; Non-reference derived topological constraints modify the overall peptide hydrophobicity. Non-reference-derived topological constraints alter peptide solubility. Non-reference-derived topological constraints modify the peptide net charge. Non-reference derived topological constraints allow detection in labeled or label-free assays. Non-reference derived topological constraints allow detection in in vitro assays. Non-reference derived topological constraints allow detection in in vivo assays. Non-reference-derived topological constraints enable capture from complex mixtures. Non-reference-derived topological constraints enable enzymatic processing. Non-reference-derived topological constraints enable cell membrane permeability. non-reference-derived topological constraints allow binding to secondary targets, and / or An engineered peptide according to (46) above, wherein non-reference derived topological constraints modify immunogenicity; (48) A method for selecting an engineered peptide, comprising: Identifying one or more topological characteristics of the reference target; designing spatially related constraints for each topological feature to generate a combination of spatially related topological constraints derived from the reference target; comparing the spatially related topological properties of the candidate peptide with the set of spatially related topological constraints derived from the reference target; and selecting candidate peptides having spatially related topological properties that overlap with the combination of spatially related topological constraints derived from the reference target to generate the engineered peptide. (49) The method of (48), wherein the overlap between each feature is 75% or less mean percent error (MPE) as determined independently by one or more of Total Topological Constraint Distance (TCD), Topological Clustering Coefficient (TCC), Euclidean distance, Power distance, Soergel distance, Canberra distance, Sorensen distance, Jaccard distance, Mahalanobis distance, Hamming distance, Quantitative Estimate of Likeness (QEL), or Chain Topological Parameter (CTP); (50) The method according to (48) or (49) above, wherein one or more constraints are derived from per-residue energy, per-residue interaction, per-residue variation, per-residue interatomic distance, per-residue chemical descriptor, per-residue solvent exposure, per-residue amino acid sequence similarity, per-residue bioinformatic descriptor, per-residue noncovalent bonding propensity, per-residue phi / psi angle, per-residue van der Waals radius, per-residue secondary structure propensity, per-residue amino acid adjacency, or per-residue amino acid contact; (51) The method according to any one of (48) to (50) above, wherein the properties of one or more candidate peptides are determined by computer simulation; (52) The method according to (51) above, wherein the computer simulation includes molecular dynamics simulation, Monte Carlo simulation, coarse-grained simulation, Gaussian network model, machine learning, or any combination thereof; (53) The method according to any one of (48) to (52) above, wherein the properties of one or more candidate peptides are determined by experimental characterization; (54) The method according to any one of (48) to (53) above, wherein the amino acids satisfying the constraints derived from the one or more reference targets have 10% to 90% sequence identity with the reference targets; (55) The amino acid satisfying the constraints from the one or more reference targets is within 30 Å 2 ~3000Å 2 The method according to any one of (48) to (54) above, having a van der Waals surface area overlap with the reference of (56) The method according to any one of (48) to (55) above, wherein the combination includes constraints derived from at least two reference targets; (57) The method according to any one of (48) to (56) above, wherein the combination includes constraints derived from at least five reference targets; (58) The method according to any one of (48) to (57) above, wherein the combination of constraints includes one or more constraints that are not derived from the reference target; (59) The method according to (58) above, wherein the one or more non-reference target-derived constraints describe desired structural, kinetic, chemical, or functional properties, or any combination thereof; (60) The method according to any one of (48) to (59) above, wherein the constraints are independently selected from the group consisting of: interatomic distance, Atomic fluctuations, atomic energy, chemical descriptors, solvent exposure, amino acid sequence similarity, bioinformatics descriptors, non-covalent tendency, Phi angle, Psi angle, van der Waals radius, secondary structure tendency, amino acid contiguity, and amino acid contacts; (61) The method according to any one of (48) to (60) above, wherein one or more constraints are independently atomic variations; (62) The method according to any one of (48) to (61) above, wherein one or more constraints are independently chemical descriptors; (63) The method according to any one of (48) to (62) above, wherein one or more constraints are independently interatomic distances; (64) The method according to any one of (48) to (63) above, wherein one or more constraints are independently a secondary structure; (65) The method according to any one of (48) to (64) above, wherein one or more constraints are independently van der Waals surfaces; (66) The method according to any one of (48) to (65) above, wherein one or more constraints independently relate to a biological response or a biological function; (67) The method according to any one of (48) to (66) above, wherein the engineered peptide comprises one or more atoms associated with a biological response or biological function; (68) The method according to any one of (48) to (66) above, wherein the engineered peptide comprises one or more amino acids associated with a biological response or biological function; (69) The biological response or biological function is selected from the group consisting of gene expression, metabolic activity, protein expression, cell proliferation, cell death, cytokine secretion, kinase activity, epigenetic modification, cell death activity, inflammatory signaling, chemotaxis, tissue infiltration, immune cell lineage commitment, tissue microenvironment modification, immune synapse formation, IL-2 secretion, IL-10 secretion, growth factor secretion, interferon gamma secretion, transforming growth factor beta secretion, immunoreceptor tyrosine-based activation motif activity, immunoreceptor tyrosine-based inhibitory motif activity, antibody-dependent cellular cytotoxicity, complement-dependent cytotoxicity, biological pathway agonism, and biological the method according to any one of (66) to (68), wherein the therapeutic agent is selected from the group consisting of biological pathway antagonism, biological pathway redirection, kinase cascade modification, protein degradation pathway modification, protein homeostasis pathway modification, protein folding / pathway, post-translational modification pathway, metabolic pathway, gene transcription / translation, mRNA degradation pathway, gene methylation / acetylation pathway, histone modification pathway, epigenetic pathway, immune-dependent clearance, opsonization, hormone signaling, integrin pathway, membrane protein signaling, ion channel flux, and g-protein coupled receptor response; (70) The reference target comprises one or more atoms associated with a biological response or biological function; 66. The method of claim 66, wherein the atomic variation of the one or more atoms in the engineered peptide associated with a biological response or biological function overlaps with the atomic variation of the one or more atoms in the reference target associated with a biological response or biological function; (71) The method according to (70) above, wherein the overlap is a root mean square dot product (RMSIP) greater than 0.25; (72) The method according to (71) above, wherein the overlap has a root mean square dot product (RMSIP) greater than 0.75; (73) The method according to any one of (67) to (69), wherein at least a portion of the atoms in the engineered peptide associated with a biological response or biological function are topologically constrained to secondary structure elements in the reference target; (74) The method according to (73) above, wherein the secondary structure element is a beta-sheet; (75) The method according to (73) above, wherein the secondary structure element is an alpha helix; (76) The method according to (73) above, wherein the secondary structure element is a turn, the turn comprising 2 to 7 residues and comprising at least one inter-residue hydrogen bond; (77) The method according to (73) above, wherein the secondary structure element is a coil, and the coil comprises 2 to 20 residues; (78) The method according to (73) above, wherein the coil does not contain inter-residue hydrogen bonds; (79) The method according to any one of (67) to (69), wherein at least a portion of the atoms in the engineered peptide associated with a biological response or function are topologically constrained to a combination of two or more secondary structure elements independently selected from the group consisting of a beta-sheet, an alpha-helix, a turn, and a coil; (80) The method according to any one of (48) to (79) above, wherein one or more spatially related topological constraints are interatomic distances; (81) The method according to any one of (48) to (80) above, wherein one or more spatially related topological constraints are atomic energies; (82) The method according to (81) above, wherein each atomic energy is independently a pairwise attractive energy between two atoms, a pairwise repulsive energy between two atoms, an atomic-level solvation energy, a pairwise charge attractive energy between two atoms, a pairwise hydrogen bond attractive energy between two atoms, or a non-covalent bond energy; (83) The method according to any one of (48) to (82) above, wherein one or more spatially related topological constraints are chemical descriptors; (84) The method according to (83) above, wherein each chemical descriptor is independently hydrophobicity, polarity, volume, net charge, logP, high performance liquid chromatography retention, or van der Waals radius; (85) The method according to any one of (48) to (84) above, wherein one or more spatially related topological constraints are bioinformatic descriptors; (86) The method according to (85), wherein each bioinformatic descriptor is independently BLOSUM similarity, pKa, zScale, Cruciani property, Chidera factor, VHSE scale, ProtFP, MS-WHIM score, T scale, ST scale, transmembrane propensity, protein-buried region, helix propensity, sheet propensity, coil propensity, turn propensity, immunogenic propensity, antibody epitope occurrence, or protein interface occurrence; (87) The method according to any one of (48) to (86) above, wherein one or more spatially related topological constraints are solvent exposure; (88) The method according to any one of (48) to (87) above, wherein at least one of the constraints derived from the one or more reference targets is a GPCR extracellular domain; (89) The method according to any one of (48) to (88) above, wherein at least one of the constraints derived from the one or more reference targets is an ion channel extracellular domain; (90) The method according to any one of (48) to (89) above, wherein at least one of the constraints derived from the one or more reference targets is a protein-protein or protein-peptide interface bond; (91) The method according to any one of (48) to (90) above, wherein at least one of the constraints derived from the one or more reference targets is derived from a polymorphic region of the target; (92) The method according to any one of (48) to (91) above, wherein the engineered peptide comprises one or more atoms associated with a biological response or biological function, each of the one or more atoms being independently selected from the group consisting of carbon, oxygen, nitrogen, hydrogen, sulfur, phosphorus, sodium, potassium, zinc, manganese, magnesium, copper, iron, molybdenum, and nickel; (93) The method according to any one of (48) to (92) above, wherein the engineered peptide comprises one or more amino acids associated with a biological function or biological response, each of which is independently a naturally occurring proteinogenic amino acid, a naturally occurring non-proteinogenic amino acid, or a chemically synthesized non-natural amino acid; (94) The method according to any one of (48) to (93) above, wherein the engineered peptide has at least one structural difference compared to the reference target; (95) The method of (94) above, wherein the at least one structural difference is independently selected from the group consisting of sequence, number of amino acid residues, total number of atoms, total hydrophilicity, total hydrophobicity, total positive charge, total negative charge, one or more secondary structures, shape factor, Zernike descriptor, van der Waals surface, structural graph nodes and edges, volume surface, electrostatic potential surface, hydrophobic potential surface, local diameter, local surface features, backbone model, charge density, hydrophilic density, surface-to-volume ratio, amphiphilicity density, and surface roughness; (96) The method according to (95) above, wherein the difference in the one or more secondary structures is the presence of one or more additional secondary structure elements in the engineered peptide compared to the reference target, each additional secondary structure element being independently selected from the group consisting of an alpha helix, a beta-sheet, a loop, a turn, and a coil; (97) The method according to any one of (48) to (96) above, wherein 10% to 90% of the amino acids of the engineered peptide satisfy topological constraints derived from one or more non-reference targets; (98) The method according to (97) above, wherein the topological constraints derived from one or more non-reference targets reinforce a pre-specified function; (99) Non-reference-derived topological constraints reinforce or stabilize secondary structure elements in the reference-derived fraction of said peptide; non-reference-derived topological constraints enforce atomic variation in the reference-derived fraction of the peptide; Non-reference-derived topological constraints modify the overall peptide hydrophobicity, Whether non-reference-derived topological constraints alter peptide solubility; Whether non-reference-derived topological constraints alter peptide net charge; Non-reference derived topological constraints allow for detection in labeled or label-free assays, Whether non-reference-derived topological constraints enable detection in in vitro assays; Whether non-reference-derived topological constraints enable detection in in vivo assays; Whether non-reference-derived topological constraints enable capture from complex mixtures; Whether non-reference-derived topological constraints allow enzymatic processing Whether non-reference-derived topological constraints enable cell membrane permeability; Non-reference-derived topological constraints allow binding to secondary targets, or Whether non-reference-derived topological constraints alter immunogenicity or or any combination thereof; (100) A composition comprising two or more selection-inducing polypeptides, each polypeptide independently being a positive selection molecule comprising one or more positive induction properties, or a negative selection molecule comprising one or more negative induction properties, each property type independently being: amino acid sequence, polypeptide secondary structure, molecular dynamics, Chemical characteristics, biological function, immunogenicity, Multispecificity of reference target(s), Cross-species reference target reactivity, selectivity of the desired reference target(s) over the undesired reference target(s); selectivity of the reference target(s) within a sequence and / or structurally homologous family; selectivity of reference target(s) with similar protein function; Selectivity of distinct desired reference target(s) from a larger family of undesired targets with high sequence and / or structural homology; Selectivity for distinct reference target alleles or mutations; Selectivity for chemical modification at the level of distinct reference target residues; cell type selectivity, Selectivity for tissue type, selectivity to the organizational environment; tolerance to structural diversity of the reference target(s); Tolerance to sequence variability of the reference target(s), and tolerance to kinetic diversity of the reference target(s); a composition wherein at least one of the two or more polypeptides is the engineered peptide described in (1) above; (101) The composition according to (100), wherein at least one of the two or more polypeptides is a positive selection molecule and at least one of the two or more polypeptides is a negative selection molecule; (102) The composition according to (100) or (101), wherein at least one of the two or more polypeptides is a natural protein; (103) The composition according to any one of (100) to (102) above, comprising at least one pair of corresponding positive selection molecules and negative selection molecules that share at least one shared characteristic type, wherein the positive selection molecules share the positive characteristic and the negative selection molecules share the negative characteristic; (104) A method for screening a library of binding molecules using the composition according to (100), comprising subjecting a pool of candidate binding molecules to at least one selection round, Each selection round comprises a negative selection step in which at least a portion of the pool is screened against negatively selected molecules; a positive selection step of screening at least a portion of said pool in favor of positively selected molecules; a method in which the order of selection steps within each round, and said order of rounds results in the selection of a subset of said pool different from an alternative order; (105) The method according to (104) above, wherein the library of binding molecules is a phage library; (106) The method according to (105) above, wherein the library of binding molecules is a cell library; (107) The method according to (106) above, wherein the library of binding molecules is a B cell library; (108) The method according to (106) above, wherein the library of binding molecules is a T cell library; (109) The method according to any one of (104) to (108) above, comprising two or more selection rounds; (110) The method according to any one of (104) to (109) above, comprising three or more selection rounds; (111) The method according to (109) or (110) above, wherein each round involves a different set of selected molecules; (112) The method according to (109) or (110) above, wherein at least two rounds contain the same negative selection molecule, or the same positive selection molecule, or both; (113) The method according to any one of (109) to (112) above, comprising analyzing the subset of the pool obtained from a selection round before proceeding to the next selection round; (114) The method according to (113) above, wherein subset pool analysis determines the set of positively and / or negatively selected molecules to be used in one or more subsequent selection rounds; (115) The method according to (113) or (114) above, wherein each subset pool analysis is independently selected from the group consisting of peptide / protein biosensor binding, peptide / protein ELISA, peptide library binding, cell extract binding, cell surface binding, cell activity assay, cell proliferation assay, cell death assay, enzyme activity assay, gene expression profile, protein modification assay, Western blot, and immunohistochemistry; (116) The method according to any one of (113) to (115) above, wherein the positively, negatively, or both positively and negatively selected molecules used in one or more subsequent selection rounds are determined by statistical / informatics scoring of subset pool analysis or machine learning training; (117) The method according to any one of (109) to (116) above, wherein the subset pool obtained from a selection round is modified before moving to the next selection round; (118) The method according to (117) above, wherein the subset pool analysis determines the positively, negatively, or both positively and negatively selected molecules to be used in one or more subsequent selection rounds, and the subset pool is modified before moving on to the next selection round; (119) The method according to (117) or (118) above, wherein each modification is independently selected from the group consisting of gene mutation, gene depletion, gene enrichment, chemical modification, and enzymatic modification; [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 provides a schematic illustrating the construction of an exemplary combination of three spatially related topological constraints for use in the selection of engineered peptides described herein.

[0013] [Figure 2] Figure 2 provides a schematic of the steps involved in several exemplary methods for determining reference-derived spatially relevant topological constraints and their use in the selection of engineered peptides (mesoscale molecules, MEMs).

[0014] [Figure 3A] 3A-3C provide schematic diagrams illustrating the selection of a group of engineered peptides using the methods described herein. Figure 3A illustrates the extraction of spatially relevant topological information for interfaces of interest in a reference and its use in defining topological constraints for use in the selection of engineered peptides. [Figure 3B] Figure 3B provides a schematic detailing the in silico screen steps, showing how mismatched candidates are discarded while phase-matched candidates are retained. [Figure 3C] Figure 3C shows the top 12 candidate engineered peptides identified.

[0015] [Figure 4A] 4A-4B provide a second set of schematic diagrams illustrating the selection of groups of different engineered peptides based on different sets of reference parameters using the methods described herein. Figure 4A illustrates the extraction of spatially related phase information and the construction of a phase matrix. [Figure 4B] Figure 4B provides a list of the top eight candidate engineered peptides, selected by comparing the candidates with topological constraints in silico.

[0016] [Figure 5] FIG. 5 is a schematic outlining the design of an exemplary programmable in vitro selection using the engineered peptides described herein, and also using native proteins as positive (T) or negative (X) selection molecules.

[0017] [Figure 6A]Figures 6A-6H provide an overview of the selection of five engineered peptides and their use in a programmable in vitro selection protocol for phage panning. Figure 6A shows the selection of VEGF as a reference target and the identification of portions of VEGF from which spatially relevant topological information was derived and used to construct a combination of spatially relevant topological constraints (Step 1). This combination was then used in in silico screening of candidate engineered peptides to identify positive and negative selection molecules (Step 2). Selected candidates were further screened in silico for stabilizing crosslinking options. Once identified and stabilized engineered peptides were obtained, they were then used to construct a programmable in vitro selection protocol for phage panning. [Figure 6B] FIG. 6B shows the analysis and identification of spatially relevant topological constraints based on the reference target (a portion of VEGF) used in the selection of engineered peptides. [Figure 6C] Figures 6C, 6D, and 6E show the construction of each of the first, second, and third candidate engineered peptides, as well as the derivation of parameters for comparison with the constraint combinations created in Figure 6B. [Figure 6D] Same as above [Figure 6E] Same as above [Figure 6F] Figure 6F lists the mean percent error (MPE) of each MEM compared to the reference target, and their rank based on the MPE. [Figure 6G] Figure 6G shows how an additional set of constraints was added to the combination based on the reference target. [Figure 6H] In Figure 6H, this additional constraint set is used to evaluate candidate MEM1. The MPE for this comparison was 36.6%.

[0018] [Figure 7-1]Figure 7A is a ribbon diagram of VEGF, using the reference section to select engineered peptides (R82-H90). Figure 7B is a ribbon diagram of five candidate engineered peptides selected based on constraints created from the target reference in Figure 7A. The sequences and root mean squared RMSIPs are listed in Table 1. [Figure 7-2] Figure 7D shows the two eigenvectors describing the two most dominant motions of the epitope in the reference target. The x, y, and z components of the 10 Ca atoms in the epitope, as well as the eigenvalues of the eigenvectors, are tabulated. The structure shows the projection of each Ca atom in the epitope along eigenvector 1 (arrow) and eigenvector 2 (arrow). Eigenvectors are orthonormal by definition. [Figure 7-3] Figure 7E shows the eigenvectors describing the most dominant motions (modes) in the epitopes of the reference target (left) and the MEM (right). The structure of the MEM superimposed on the epitope is shown, along with the MEM variant ID and RMSIP. [Figure 7-4] Figure 7F provides eigenvectors describing the second major motion (mode) in the epitope of the reference target (left) and the MEM (right). The structure of the MEM superimposed on the epitope is shown, along with the MEM variant ID and RMSIP. [Figure 7-5] Figure 7G provides the structure of the reference target and MEM with their associated projections along the three most principal motions (modes, eigenvectors 1-3) in relation to their position in the dot product matrix used to compute the RMSIP. The RMSIP equation used is shown for reference.

[0019] [Figure 8] Figure 8 shows the structural ensemble and coordinate covariance matrices of the reference target (top) and MEM (bottom) generated from experimental data or computer simulations. The epitope is the darker area in the upper right corner of the reference target.

[0020] [Figure 9]Figure 9 is an overview of the in vitro programmable selection design using four engineered peptides (also called mesoscale engineered molecules, or MEMs) for positive or negative selection. The atomic motion and phase scores of the MEMs are included for reference. The sequences are provided as SEQ ID NOs: 1-4.

[0021] [Figure 10-1] 10A-D are graphs of binding biosensor assays using different engineered peptides from FIG. 9 against bevacizumab. [Figure 10-2] Same as above

[0022] [Figure 11] Figure 11 shows a description of eight different panning programs, seven of which include engineered peptides as one or more selection molecules, and eight of which use conventional native proteins for selection. A naive Hu scFv library was panned separately in each program.

[0023] [Figure 12A] Figures 12A and 12B are VEGF ELISA response graphs comparing VEGF binding responses to binding partners selected using the different panning programs described in Figure 11. As shown in Figure 12A, MEM-programmed in vitro selection does not significantly reduce full-length target binding propensity with certain MEM program inputs, but not with all inputs. Bars indicate means; significant difference between P12 and P7: p-value < 0.0001. [Figure 12B] As shown in Figure 12B, MEM-programmed in vitro selection favors putative epitope-selective clones in a statistically significant manner: bars indicate means; P12 vs. P6: p-value 0.024, P12 vs. P9: p-value 0.0004, P12 vs. P10: p-value 0.049.

[0024] [Figure 13A]13A-13H are graphs showing the binding of sMEM engineered peptides versus VEGF (reference) for binding partners selected using the different panning programs described in FIG. [Figure 13B] Same as above [Figure 13C] Same as above [Figure 13D] Same as above [Figure 13E] Same as above [Figure 13F] Same as above [Figure 13G] Same as above [Figure 13H] Same as above

[0025] [Figure 14A] 14A-14I are graphs showing binding of binding partners selected using different panning programs described in FIG. 11 in a VEGF cross-blocking assay using dose-response competition with bevacizumab (0 nM, 67 pM, 670 pM, 6.7 nM). [Figure 14B] Same as above [Figure 14C] Same as above [Figure 14D] Same as above [Figure 14E] Same as above [Figure 14F] Same as above [Figure 14G] Same as above [Figure 14H] Same as above [Figure 14I] Same as above

[0026] [Figure 15] FIG. 15 is a graph of the distinct clones with confirmed cross-blocking properties obtained from each of the different selection programs outlined in FIG.

[0027] [Figure 16] FIG. 16 is a summary of binding, cross-blocking, CDR sequences, and germline usage for all Fabs generated from the selection program outlined in FIG.

[0028] [Figure 17] 17 and 18 are the ELISA binding results for all of the Fabs listed in FIG. [Figure 18] Same as above

[0029] [Figure 19] Figure 19 shows bevacizumab blockade propensity scores (0 nM, 67 pM, 670 pM, 6.7 nM) for random clones versus those selected from the selection program outlined in Figure 11. ELISA Z-score (sMEM + VEGF-iMEM) + bevacizumab blockade Z-score.

[0030] [Figure 20] FIG. 20 summarizes the cross-blocking enrichment for random-to-uniform selection of clones from the entire panning program described in FIG.

[0031] [Figure 21] Figure 21 is a schematic diagram showing how next-generation sequencing samples of selected clones were prepared. Individual heavy and light chain sequences in the constant portion of the expression vector were cloned using 2x250 paired-end sequencing runs. The ends were then joined and the reads were annotated (e.g., using PyIg). The reads obtained from the clones selected using each selection program are shown in the bar graph.

[0032] [Figure 22] Figure 22 shows the clonality analysis (number of distinct antibodies) of the different panning rounds, and the normalized Shannon analysis.

[0033] [Figure 23] FIG. 23 shows the clonality of the different screening programs described in FIG.

[0034] [Figure 24A]Figures 24A-24L are germline usage heat maps and dimensionality reduction plots analyzing how different screening rounds and programs for round 1 (Figures 24A-24D), round 2 (Figures 24E-24H), and round 3 (Figures 24I-24L) shape the diversity of the resulting selected pools. [Figure 24B] Same as above [Figure 24C] Same as above [Figure 24D] Same as above [Figure 24E] Same as above [Figure 24F] Same as above [Figure 24G] Same as above [Figure 24H] Same as above [Figure 24I] Same as above [Figure 24J] Same as above [Figure 24K] Same as above [Figure 24L] Same as above

[0035] [Figure 25A] Figures 25A-25B summarize the clones (S# on the x-axis) isolated from each selection program and their binding to VEGF and the engineered peptide sMEM. [Figure 25B] Same as above

[0036] [Figure 26] Figure 26 summarizes the enrichment rates of unique mAb hits obtained from each round of each program that were confirmed to bind VEGF and cross-block bevacizumab, and that were not identified by conventional panning (Program 12) without engineered peptides.

[0037] [Figure 27] Figure 27 summarizes the enrichment kinetics of mAb hits obtained from a conventional panning program (12) that were confirmed to bind VEGF but were not putative epitope-selective mAb hits.

[0038] [Figure 28] Figure 28 summarizes the binding to sMEM or VEGF of different clones obtained from different panning programs.

[0039] [Figure 29] Figure 29 is a schematic diagram of a second exemplary set of programmed in vitro selection protocols targeting a proposed therapeutic epitope reference site on PD-L1. Spatially related topological constraints were derived from this putative site and used to screen in silico for engineered peptides with properties that overlap with the combined constraints. These were then used in selection rounds in phage panning of a naive Hu scFv library.

[0040] [Figure 30] Figure 30 provides the modeled structures and peptide sequences of three engineered peptides selected according to the schematic diagram in Figure 29. The sequences are provided as SEQ ID NOs: 5-7.

[0041] [Figure 31A] Figures 31A-31D show the interatomic distance and amino acid descriptor matrices derived from the reference (Figure 31A) and engineered peptides sMEM (Figure 31B), nMEM (Figure 31C), and iMEM (Figure 31D). Compared to the reference phase, the average error rates of the sMEM, nMEM, and iMEM phases were 3.58%, 0.84%, and 19.3%, respectively. [Figure 31B] Same as above [Figure 31C] Same as above [Figure 31D] Same as above

[0042] [Figure 31E] Figures 31E-31G are biosensor binding graphs showing the binding between the engineered peptides described in Figure 30 and avelumab. The KD for nMEM binding with avelumab was 43.4 uM. [Figure 31F] Same as above [Figure 31G] Same as above

[0043] [Figure 32A] 32A-32C are biosensor binding graphs showing the binding between the engineered peptides described in FIG. 30 and durvalumab. [Figure 32B] Same as above [Figure 32C] Same as above

[0044] [Figure 33] Figure 33 summarizes the differences between a programmed in vitro selection panning program using one or more of the engineered peptides described in Figure 30 and a traditional panning method using native protein (C1). The engineered peptides sMEM, nMEM, and iMEM in Figure 30 are sMEM #1, sMEM #5, and iMEM in Figure 33.

[0045] [Figure 34] Figure 34 is a graph and summary of the PD-L1 ELISA binding responses for clones selected using each of the panning programs described in Figure 33.

[0046] [Figure 35] FIG. 35 is a graph and summary of the ELISA binding responses to sMEM #1 for clones selected using each of the panning programs described in FIG.

[0047] [Figure 36] FIG. 36 is a graph and summary of the ELISA binding responses to nMEM #5 for clones selected using each of the panning programs described in FIG.

[0048] [Figure 37]Figure 37 is a graph and summary of the ELISA epitope selectivity responses to PD-L1 and sMEM #1 for clones selected using each of the panning programs described in Figure 33.

[0049] [Figure 38] Figure 38 is a graph and summary of the ELISA epitope selectivity responses to PD-L1 and nMEM #5 for clones selected using each of the panning programs described in Figure 33.

[0050] [Figure 39A] Figures 39A-39U compare the different ELISA binding responses of Figures 34-38, showing the selectivity of the binding partners selected using the different programs. [Figure 39B] Same as above [Figure 39C] Same as above [Figure 39D] Same as above [Figure 39E] Same as above [Figure 39F] Same as above [Figure 39G] Same as above [Figure 39H] Same as above [Figure 39I] Same as above [Figure 39J] Same as above [Figure 39K] Same as above [Figure 39L] Same as above [Figure 39M] Same as above [Figure 39N] Same as above [Figure 39O] Same as above [Figure 39P] Same as above [Figure 39Q] Same as above [Figure 39R] Same as above [Figure 39S] Same as above [Figure 39T] Same as above [Figure 39U] Same as above

[0051] [Figure 40] Figure 40 is a table summarizing the anti-PD-L1 panning ELISA hit identification criteria used to analyze clones obtained from the selection program described in Figure 33.

[0052] [Figure 41A] Figures 41A-41C compare the different ELISA binding responses to sMEM #1 and nMEM#5 compared to PD-L1 (Figures 41A and 42B, respectively), and sMEM #1 compared to nMEM#5 (Figure 41C), for binding partners selected using the different panning programs described in Figure 33. [Figure 41B] Same as above [Figure 41C] Same as above

[0053] [Figure 42-1] 42A-42F compare the different ELISA responses and confirmed Tx mAb X blockers for all of the programs described in FIG. [Figure 42-2] Same as above [Figure 42-3] Same as above

[0054] [Figure 43] Figure 43 summarizes the 23 distinct clones from the program described in Figure 33 identified from the cross-blocking hits and their sequences.

[0055] [Figure 44] FIG. 44 is a chart of the confirmed cross-blocking distinct clones obtained from each of the programs described in FIG.

[0056] [Figure 45A]Figure 45A is a graph of the blocking propensity of randomly selected clones obtained from each of the programs described in Figure 33. Blocking was assessed as clonal blockade of PD-L1 binding to avelumab or durvalumab. Blocking propensity was assessed as ELISA Z-score (sMEM1 + sMEM5 + PD-L1-iMEM) + MAX (avelumab blocking Z-score, durvalumab blocking Z-score).

[0057] [Figure 45B] Figures 45B and 45C summarize the blocking trends of clones obtained from the different programs evaluated in Figure 45 A. The shaded inputs in Figure 45C were obtained using a conventional selection approach using native proteins. [Figure 45C] Same as above

[0058] [Figure 46] Figure 46 is a summary of the cross-blocking enrichment observed in pools of clones obtained using the program described in Figure 33 compared to the control (traditional approach).

[0059] [Figure 47] FIG. 47 is an example of a phase matrix that can be used in the selection of engineered peptides described herein.

[0060] [Figure 48] Figure 48 is an example of a topologically constrained chemical descriptor vector that can be used in the selection of engineered peptides described herein.

[0061] [Figure 49] FIG. 49 is an exemplary L×2 phi / psi matrix that can be used in the selection of engineered peptides described herein.

[0062] [Figure 50] Figure 50 is an exemplary SxSxM matrix of secondary structure interaction descriptors that may be used in the selection of engineered peptides described herein.

[0063] [Figure 51] FIG. 51 is an exemplary diagram showing exemplary engineered peptide clusters and TCC vectors that can be used in the selection of engineered peptides described herein.

[0064] [Figure 52] FIG. 52 is an exemplary L×M topological constraint matrix that can be used in the selection of engineered peptides described herein.

[0065] [Figure 53] FIG. 53 is an exemplary secondary structure index and lookup table that can be used in the selection of engineered peptides described herein.

[0066] [Figure 54] Figure 54 is another representation of the data obtained from the VEGF panning programs. S1 refers to anti-VEGF panning program 6, S2 refers to anti-VEGF panning program 13, and C is the conventional full-length VEGF program.

[0067] [Figure 55] Figure 55 is another representation of the data provided in Figure 24I. S1 refers to anti-VEGF panning program 6, S2 refers to anti-VEGF panning program 13, and C is the conventional full-length VEGF program.

[0068] [Figure 56] Figure 56 is another representation of the data provided in Figure 26. S1 refers to anti-VEGF panning program 6, S2 refers to anti-VEGF panning program 13, and C is the conventional full-length VEGF program.

[0069] [Figure 57A]Figures 57A-57E are graphs of VEGF (solid grey line) and cross-blocking (dotted line) binding data for clones on selected epitopes from programmed in vitro selection. [Figure 57B] Same as above [Figure 57C] Same as above [Figure 57D] Same as above [Figure 57E] Same as above

[0070] [Figure 58A] Figures 58A-58C are graphs of VEGF binding data for selected clones outside the epitope from full-length in vitro selection. [Figure 58B] Same as above [Figure 58C] Same as above

[0071] [Figure 59A] Figures 59A-59B summarize antibody clone hits CDR loop sequence diversity for anti-VEGF programmed in vitro selection (red) and conventional in vitro selection (gray). [Figure 59B] Same as above

[0072] [Figure 60]Figure 60 shows sequence alignments of clones selected using the programmable in vitro selection methods described herein using exemplary engineered peptides described herein. The top row is an alignment of heavy chain sequences of clones on the top five epitopes selected across all programmed in vitro selection programs, the second row is an alignment of heavy chain sequences of clones outside the top five epitopes selected using a traditional approach using VEGF and BSA as selection molecules, the third row is an alignment of light chain sequences of clones on the top five epitopes selected across all programmed in vitro selection programs, and the bottom row is an alignment of light chain sequences of clones selected using a traditional approach with VEGF and BSA.

[0073] [Figure 61] FIG. 61 is a schematic diagram of an exemplary method for engineered polypeptide design.

[0074] [Figure 62] FIG. 62 is a schematic diagram of an exemplary method of using a machine learning model for engineered polypeptide design. DETAILED DESCRIPTION OF THE INVENTION

[0075] Methods for selecting mesoscale engineered peptides and methods for preparing peptides containing the engineered peptides Compositions containing the engineered peptides and methods of using the engineered peptides are provided herein. For example, methods of using engineered peptides in the in vitro selection of antibodies are described herein. Provided.

[0076] The engineered peptides of the present disclosure are between 1 kDa and 10 kDa and are referred to herein as "metabolic peptides." Engineered peptides of this size are, in some embodiments, , protein-like functionality, large theoretical space for candidate selection, cell permeability, as well as and / or may have particular advantages such as structural and kinetic variability.

[0077] The methods provided herein involve the application of multiple spatially related topological constraints, some of which may include: (which may be derived from a reference target) and constructing a set of constraints. , comparing candidate peptides with the combination and having constraints that overlap with the combination. and selecting candidates. Different aspects of the engineered peptides may be tailored to suit the intended use, desired function, or other desired function. Furthermore, in some embodiments, all of the constraints may be included in the combination depending on the characteristics of the Approximately 1000 nucleotides need not be derived from the reference target. Through such methods, in some embodiments, The selected engineered peptides may simply be variations of a reference target (e.g., a single reference peptide). rather than a desired function (as may be obtained through mutagenesis or incremental modification). The peptides have overall characteristics that differ from the reference peptide while retaining their structural properties and / or important substructures. It may have a structure.

[0078] The present invention includes a method for programmable in vitro selection using one or more engineered peptides. Further provided herein are methods of using such engineered peptides, including: Such selection can be used, for example, in the identification of antibodies.

[0079] These methods and engineered peptides are described in more detail below.

[0080] I. Methods for selecting engineered peptides In some aspects, provided herein are methods for selecting engineered peptides, teeth, Identifying one or more topological characteristics of the reference target; For each topological feature, a spatially related set of constraints is generated to generate a combination of constraints derived from the reference target. Designing related constraints; Spatially related topological properties of candidate peptides are compared with combinations derived from reference targets To do, possess spatially related topological properties that overlap with the set of constraints derived from the reference target and selecting candidate peptides that are

[0081] In some embodiments, one or more additional spatially related positions not derived from the reference target are included. The relative constraints are included in the combination.

[0082] a. Spatially related topological constraints The engineered peptides described herein are designed to conform to the topological constraints with which they are spatially related. The combination is selected based on how closely it matches the Such combinations (or In a tensor, each constraint is independently described in three-dimensional space (e.g., spatially related The combination of these constraints in three-dimensional space can be used to achieve different desired properties, e.g. and providing a representational "map" of their desired levels (if applicable) relative to their locations. This map may, in some embodiments, be linear or otherwise based on a given amino acid skeleton. It is not based on the criteria and therefore does not guarantee that the desired combination will be met as described. For example, in some embodiments, A "gap" is a region of space where a given constraint can be satisfactorily satisfied by two adjacent amino acids. In some embodiments, these amino acids may be directly linked (e.g., 2 two consecutive amino acids), while in other embodiments, the amino acids are not directly linked to each other. can be joined in space by peptide folding (e.g., consecutive amino acids The separate constraints themselves are not necessarily structure-based, but may be based on, for example, chemical descriptors. and / or functional descriptors. In some embodiments, the constraints may include a desired secondary structure. In certain embodiments, each constraint independently comprises a structural descriptor, such as a structure or amino acid residue. is selected.

[0083] For example, Figure 1 shows a schematic diagram illustrating the construction of a representative set of spatially related topological constraints. The three constraints in Figure 1 are alignment, nearest neighbor distance, and atomic motion. and atomic motion are combined in one graphic. As shown, some constraints The coordinates are mapped independently of the backbone position (e.g., atomic motion of a particular side chain) and Therefore, much more is possible than simply varying one or more positions on a reference scaffold. Three different constraints and their spatial recording allow various structural configurations to be attempted. The descriptions are combined into a matrix (e.g., a tensor), and then a set of candidate peptides is combined into this matrix. The peptides are compared to the original peptide combination to identify new engineered peptides that meet the desired criteria. In some embodiments, one or more additional non-referentially derived constraints may also be included in the combination. Comparison of candidate peptides with defined combinations can include, for example, determining the desired combination. To evaluate the constraints of each candidate peptide against the This may be done using in silico methods. The candidates with overlapping levels are synthesized using standard peptide synthesis methods known to those skilled in the art. can be evaluated.

[0084] In some embodiments, the combination of constraints is at least 3, at least 4, or at least At least 5, at least 6, at least 7, at least 8, at least 9, at least 10, At least 11, at least 12, 3-12, 3-10, 3-8, 3-6, or 3, or or 4, 5, or 6 independently selected spatially related topological constraints One or more of the constraints are derived from a reference target. In some embodiments, each of the constraints is derived from the reference target. In other embodiments, at least one constraint is derived from the reference target. However, the remaining constraints are not derived from the reference target. For example, in some embodiments, 1 to 9 0 constraints, 1-7 constraints, 1-5 constraints, or 1-3 constraints are derived from the reference target. , 1 to 9 constraints, 1 to 7 constraints, 1 to 5 constraints, or 1 to 3 constraints are It does not originate from a specific source.

[0085] Once a set of constraints is constructed, a set of candidate peptides is compared to that set. To identify one or more new engineered peptides that meet the desired criteria. In embodiments, at least 5, at least 10, at least 15, at least 20, or at least At least 25, 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 125, At least 150, at least 175, at least 200, or at least 250 or more The candidate peptides are compared in combination to identify one or more new engineered peptides that meet the desired criteria. In some embodiments, for example, more than 250 candidate peptides are identified. peptides, over 300 candidate peptides, over 400 candidate peptides, over 500 100 candidate peptides, more than 600 candidate peptides, or more than 750 candidate peptides In some embodiments, the candidate peptide is compared to the combination of constraints. To evaluate the overlap (if any) of the topological properties of In some embodiments, one or more candidate peptides are also compared to a reference target, The overlap (if any) of the topological features of the candidate peptide with the reference target topological features is assessed. In some embodiments, the engineered peptides are more than 5, more than 10, , More than 20, More than 30, More than 40, More than 50, More than 60 , more than 70, more than 80, more than 90, or more than 100 distinct Identified from computational samples of peptides and topological property simulations, The engineered peptides are selected, and the selected engineered peptides are compared to the reference target. , has the highest overlap of topological properties among the entire sample population.

[0086] The spatially related components used to construct the desired combination (e.g., the desired tensor) The associated topological constraints may each be independently selected from a wide group of possible properties. These include, for example, structural, kinetic, chemical, or functional properties, or The constraints may include constraints describing any combination of

[0087] Structural constraints include, for example, interatomic distances, amino acid sequence similarity, solvent exposure, phi angles, and The amino acid sequence may include the rhombohedral structure, secondary structure, or amino acid contacts, or any combination thereof. do.

[0088] Dynamic constraints include, for example, atomic fluctuations, atomic energies, van der Waals radii, and The atomic energies may include, for example, amino acid proximity, or non-covalent bonding tendencies. pairwise attractive energy between two atoms, pairwise repulsive energy between two atoms, Molecular level solvation energy, pairwise charge attraction energy between two atoms, Pairwise hydrogen bond attractive energy between molecules, or non-covalent bond energy, or Any combination of these may be included.

[0089] Chemical properties can include, for example, chemical descriptors. Such chemical descriptors include, for example, For example, hydrophobicity, polarity, atomic volume, atomic radius, net charge, logP, HPLC retention, foundation Lewaals radius, charge pattern, or H-bonding pattern, or any combination thereof This may include a combination of

[0090] Functional properties include, for example, bioinformatic descriptors, biological responses, or biological functions. Bioinformatic descriptors can include, for example, BLOSUM similarity, pKa, zScale, Luciani trait, Chidera factor, VHSE scale, ProtFP, MS-WHIM score , T-scale, ST-scale, transmembrane tendency, protein-buried region, helix tendency , sheet propensity, coil propensity, turn propensity, immunogenic propensity, antibody epitope occurrence, and / or or protein interface development, or any combination thereof.

[0091] In some embodiments, designing constraints involves determining the energy per residue, interactions, per-residue fluctuations, per-residue interatomic distances, per-residue chemical descriptors, Per-residue solvent exposure, per-residue amino acid sequence similarity, per-residue bioinformatic descriptors , non-covalent bond tendency per residue, phi / psi angle per residue, van der Waals angle per residue Wahl's radius, secondary structure propensity per residue, amino acid adjacency per residue, or In some embodiments, these properties are is a subset of all residues in the reference target or a subset of all residues for all combinations of constraints. In some embodiments, one or more different The same properties are used for one or more different residues. One or more properties are used for a subset of residues, and at least one different property is used for different In some embodiments, one or more constraints are used to design a subset of residues. One or more of the properties used for the analysis are determined by computer simulation. Suitable computer simulation methods include, for example, molecular dynamics simulation. Monte Carlo simulation, coarse-grained simulation, Gaussian network The methods may include machine learning, machine learning, or any combination thereof.

[0092] In some embodiments, multiple constraints are selected from one category, for example: In some embodiments, the combination is a combination of two biological responses that are independently of one type of biological response. In some embodiments, two or more constraints may independently be of a type In certain embodiments, two or more constraints independently represent a type of chemical structure. In other embodiments, the combination does not include overlapping categories of constraints.

[0093] In some embodiments, the one or more constraints are independently a biological response or a biological In some embodiments, the constraints are related to the spatially defined atoms ( level constraints, or spatially defined shape / area / volume level constraints (some a characteristic shape / area / volume etc. that can be filled by several different atomic compositions, or Spatially defined dynamic level constraints (satisfied by several different atomic compositions) (e.g., a characteristic dynamic or set of dynamics that can be detected).

[0094] In some embodiments, one or more constraints are related to a biological function or biological response. For example, in some embodiments, One or more constraints may be imposed on the extracellular domain of a G protein-coupled receptor (GPCR), or on an ion channel. In some embodiments, the 1 One or more constraints arise from protein-protein interfacial bonding. , one or more constraints are protein, such as the MHC-peptide or GPCR-peptide interface In certain embodiments, such proteins or peptides are The atoms or amino acids constrained in the peptide structure are those related to biological function or biological response. In some embodiments, such a protein or peptide is an atom or amino acid. The atoms or amino acids in the engineered peptide that are constrained to the peptide structure are derived from the reference target. In some embodiments, one or more constraints are atoms or amino acids that derived from polymorphic regions (e.g., regions subject to allelic variation between individuals).

[0095] In some embodiments, the biological response or function is determined by gene expression, metabolic activity, or other biological processes. , protein expression, cell proliferation, cell death, cytokine secretion, kinase activity, epigenetics Modification of cytotoxicity, cell death activity, inflammatory signaling, chemotaxis, tissue infiltration, immune cell lineage commitment, and tissue Tissue microenvironment modification, immune synapse formation, IL-2 secretion, IL-10 secretion, growth factor secretion, Interferon gamma secretion, transforming growth factor beta secretion, immunoreceptor tyrosine-based Activation motif activity, immunoreceptor tyrosine-based inhibitory motif activity, antibody-dependent cellular cytotoxicity Injury, complement-dependent cytotoxicity, biological pathway agonist action, biological pathway antagonist action for biological pathway redirection, kinase cascade modification, protein degradation pathway modification, protein Quality homeostasis pathway modification, protein folding / pathway, post-translational modification pathway, metabolic pathway, gene transcription / Translation, mRNA degradation pathway, gene methylation / acetylation pathway, histone modification pathway, epitaxial Genetic pathways, immune-dependent clearance, opsonization, hormone signaling, Integrin pathway, membrane protein signaling, ion channel flux, and g- Protein-coupled receptor response.

[0096] In some embodiments, one or more atoms associated with a biological function or biological response are carbon, oxygen, nitrogen, hydrogen, sulfur, phosphorus, sodium, potassium, zinc, manganese, The metal is selected from the group consisting of magnesium, copper, iron, molybdenum, and nickel. In embodiments, the atoms are selected from the group consisting of oxygen, nitrogen, sulfur, and hydrogen.

[0097] One of the constraints is one or more amino acids associated with a biological function or biological response. and / or the engineered peptide is associated with a biological function or biological response. In some embodiments, the one or more amino acids are independently There are 20 natural amino acids that make proteins, 20 natural amino acids that do not make proteins, and and unnatural amino acids. In some embodiments, the unnatural amino acids In certain embodiments, one or more amino acids are chemically synthesized. In another embodiment, one or more amino acids are selected from the naturally occurring amino acids that make up proteins. In yet a further embodiment, one or more naturally occurring amino acids are selected from non-protein-forming amino acids. In yet further embodiments, one or more of the amino acids are selected from unnatural amino acids. There are 20 natural amino acids that make up proteins, and 20 natural amino acids that do not make up proteins. The amino acid is selected from a combination of an amino acid, an amino acid, and an unnatural amino acid.

[0098] The combination of constraints used to select the engineered peptides described herein The combination includes at least one constraint derived from a reference target, but in some embodiments, One or more constraints of the combination do not originate from the reference target. Thus, in certain embodiments, The selected engineered peptides contain one or more properties that are not shared with the reference target.

[0099] In some embodiments, one or more constraints derived from a reference target and used in combination describes the inverse of the property observed in the reference target. Thus, for example, the reference target The charge-related constraints may be derived from the reference target and may have a particular pattern of charges. The constraints described are similar but neutral or negative charge patterns. In some embodiments, one or more inverse constraints are derived from a reference target and are included in the combination. Such counter-constraints may be used, for example, to determine the control sequence for a particular assay or panning method. as a child or as a negative in the programmable in vitro selection methods described herein These may be useful in selecting engineered peptides as selection molecules for the In some embodiments, the combination of spatially defined topological constraints may include one or more In some embodiments, one or more non-reference-derived topological constraints are included. The constraints either reinforce or stabilize one or more secondary structure elements or reinforce atomic perturbations. , modifying the peptide's overall hydrophobicity, modifying the peptide's solubility, or modifying the peptide's overall charge. or to allow detection in labeled or label-free assays, or in vivo. allow for detection in vitro assays or allow for detection in in vivo assays whether it allows capture from complex mixtures, allows enzymatic processing, or is cell membrane permeable. enable binding to a secondary target, or alter immunogenicity. In an embodiment, one or more non-reference derived topological constraints may be used in place of a reference target derived constraint (or (followed by the selected peptide) to restrict one or more atoms or amino acids. For example, in some embodiments, the combination of constraints is based on two sets of constraints derived from a reference target. The combination of constraints may also include secondary structure (e.g., additional hydrogen bonds, or hydrophobic interactions). secondary interactions, or side chain stacking, or salt bridges, or disulfide bonds) The structural element includes a constraint that stabilizes the structural element, and the stabilizing constraint is not present in the reference target. In some embodiments, the combination of constraints (or the subsequent selection of peptides) The constraints are: Also, the atomic variation in at least a portion of the atoms or amino acids derived from the target reference is enforced. In some embodiments, one or more constraints are present in the target reference. The above non-referentially derived constraints are inverse constraints. For example, in some embodiments, two of the constraints Combinations of these are constructed to select engineered peptides with opposite properties. In some such embodiments, the first set of constraints may include one or more constraints derived from the reference target. and one or more constraints that do not originate from the reference target, and the second combination of constraints is , the same one or more constraints derived from the reference target, and the first combination of non-reference target constraints Contains one or more of the reverses.

[0100] d. Reference target Any suitable reference target may be used to generate one or more reference targets for use in the methods provided herein. In some embodiments, the reference target may be: In other embodiments, the reference target is a full-length naturally occurring protein. In yet a further embodiment, the reference target is a non-naturally occurring protein, or a portion thereof. be.

[0101] For example, in some embodiments, the reference target is a cell surface receptor or a transmembrane protein. Protein, or signaling protein, or multiprotein complex, or protein- In some embodiments, the reference target is a peptide complex, or a portion thereof. A part of a target protein, which is a protein that causes a disease in an organism such as a human. In some embodiments, the protein of interest is involved in processes such as cancer growth. or involved in metastasis or inflammatory disorders, and the reference target is a putative epitope of interest. Thus, in some embodiments, the present invention provides a method for the preparation of a protein comprising: The method described herein is for selecting one or more engineered peptides that can serve as immunogens. The target protein may be used to generate antibodies to the target protein. Examples of proteins that can be PD-1, PD-L1, CD25, IL2, M IF, CXCR4, or VEGF. Thus, in some embodiments, The reference targets are PD-1, PD-L1, CD25, IL2, MIF, CXCR4, or is VEGF, or a portion thereof, such as an epitope. The methods provided herein involve the use of antibodies specific to a protein that is an immunogen and from which a target reference is derived. one or more engineered peptides that can be used to generate one or more antibodies that specifically bind In still further embodiments, the method provided herein may be used to select The method may be used to select one or more engineered peptides, The engineered peptides are expressed on one or more of the target proteins, such as antibodies or Fab-displaying phages. The method may be used to select one or more binding partners.

[0102] c.Compare constraints In some embodiments, one or more constraints (e.g., reference or non-reference origin) are: Molecular simulations (e.g., molecular dynamics) or laboratory measurements (e.g., NMR) , or a combination of them. Once the constraints are obtained and combined, the candidate The engineered peptides may, in some embodiments, be synthesized using computational protein design. In some embodiments, the peptide space is generated using a Other methods of sampling are then used to obtain the parameters of the selected constraints. To do this, dynamics simulations may be performed on candidate engineered peptides. A covariance matrix of atomic fluctuations is generated for the reference target, and the covariance matrix is calculated for the candidate manipulated These covariance matrices are generated for each residue in the peptide and used to determine overlap. Each covariance matrix, i.e., one covariance matrix for the reference target and one for the candidate Eigenvectors and eigenvalues for one covariance for each of the engineered peptides To compute , a principal component analysis is performed and the eigenvector with the largest eigenvalue is selected. The torque is maintained.

[0103] The eigenvectors are the first, second, and third eigenvalues observed in the set of simulated molecular structures. Describe the 3rd, 4th, and Nth principal motions. Although it is undesirable, if the candidate engineered peptide moves like the reference target, its intrinsic vector The eigenvectors of the target will be similar to those of the reference target. The components (3D vectors centered on each CA atom) are lined and point in the same direction. ) corresponds to an exemplary eigenvector comparison between a reference target and a candidate engineered peptide. are shown in Figures 7D-7G.

[0104] In some embodiments, the eigenvectors between the candidate engineered peptide and the reference target This similarity is computed using the dot product of the two eigenvectors. The dot product value is , is 0 if the two eigenvectors are at 90 degrees to each other, or If the rules point in exactly the same direction, then it is 1. Without wishing to be bound by theory, The ordering of eigenvectors is based on their eigenvalues, and the eigenvalues are used in molecular dynamics (MD ) simulations sample the underlying energy landscape of those different molecules. Due to the stochastic nature of the interaction, the results may not necessarily be identical between two different molecules, A dot product between several individually ranked eigenvectors is required in some embodiments. (e.g., the eigenvectors of the engineered peptides multiplied by the eigenvectors 2, 3, 4, etc. of the reference target) Furthermore, molecular motion is complex and involves more than two (or even several) principal Therefore, in some embodiments, the candidate operating modes may be The dot products between all pairs of eigenvectors in the selected peptide and the reference target are computed. This results in a matrix of dot products, the dimension of which depends on the number of eigenvectors being analyzed. For example, for 10 eigenvectors, the inner product matrix is 10 x 10 This matrix of dot products is the root mean square value of the dot products of 100 (in the case of 10 × 10). can be reduced to a single value by computer calculation, which is the root mean square The equation for RMSIP is shown in Figure 7G. From this comparison, One or more candidate engineered peptides that have similarity to a defined combination of constraints are identified. be selected.

[0105] e. Additional steps In some embodiments, the selection of one or more engineered peptides comprises one or more additional For example, in some embodiments, the candidate engineered peptides may be Similarity to a defined set of spatially related topological constraints, as described in the specification and then one or more analyses to determine one or more additional characteristics. and undergo one or more structural modifications to impart or enhance desired properties. For example, in some embodiments, the selected candidates are analyzed through molecular dynamics simulations or the like. and analyzed to determine the overall stability of the molecule and / or its propensity for a particular folding structure. In some embodiments, a desired level of stability or a desired folded state is determined. One or more modifications may be made to the engineered peptide to impart or enhance desired structural preferences. Such modifications can include, for example, one or more cross-links (such as disulfide bonds), salt bonds, or the like. Introduction of bridges, hydrogen bonding interactions, or hydrophobic interactions, or any combination thereof It may include insertion.

[0106] The methods provided herein involve obtaining one or more desired compounds, such as a desired binding interaction or activity. further comprising assaying one or more selected engineered peptides for a property. Any suitable assay may be used, as needed, to measure the desired property. obtain.

[0107] II. Selected Engineered Peptides In other aspects, engineered peptides, such as those selected through the methods described herein, Provided herein are engineered peptides of the formula: Peptides have a molecular weight between 1 kDa and 10 kDa and contain up to 50 amino acids. In embodiments, the engineered peptide is between 2 kDa and 10 kDa, , 3kDa~10kDa, 4kDa~10kDa, 5kDa~10kDa, 6kDa~1 0kDa, 7kDa~10kDa, 8kDa~10kDa, 9kDa~10kDa, 1k Da~9kDa, 1kDa~8kDa, 1kDa~7kDa, 1kDa~6kDa, 1kDa Da to 5kDa, 1kDa to 4kDa, 1kDa to 3kDa, or 1kDa to 2kDa In certain embodiments, the engineered peptide has a molecular weight of up to 45 amino acids. , up to 40 amino acids, up to 35 amino acids, up to 30 amino acids, up to 25 Amino acids, maximum 20 amino acids, at least 5 amino acids, at least 10 amino acids amino acids, at least 15 amino acids, at least 20 amino acids, at least 25 amino acids, at least 30 amino acids, at least 35 amino acids, or at least also contains 40 amino acids.

[0108] In certain embodiments, the engineered peptides are composed of a combination of spatially related topological constraints. and one or more of the constraints are from a reference target. Any of the constraints may be used in combination in some embodiments. In terms of morphology, 10%–98% of the amino acids in the engineered peptide are derived from one or more reference targets. (e.g., if the engineered peptide contains 50 amino acids, the length must be between 5 and 49. amino acids satisfy constraints from one or more reference targets). 20%-98%, 30%-98%, 40%-98%, 5% of the amino acids in the engineered peptide 0%~98%, 60%~98%, 70%~98%, 80%~98%, 90%~98%, 1 0%~90%, 10%~80%, 10%~70%, 10%~60%, 10%~50%, 1 0%-40%, 10%-30%, or 10%-20% of the control group is derived from one or more reference targets In still further embodiments, one or more sequences satisfying constraints from one or more reference targets are provided. The amino acids above are within 8.0 Å, 7.5 Å, 7.0 Å, and 6.5 Å of the reference target. , less than 6.0 Å, less than 5.5 Å, or less than 5.0 Å root mean square deviation (RSMD) structure In some embodiments, the engineered peptide has a molecular weight of between 1 kDa and 10 kDa. It has a molecular weight of 100 Da, contains up to 50 amino acids, and is a set of spatially related topological constraints. and one or more of the constraints are derived from a reference target, and the engineered peptide Between 10% and 98% of the amino acids in the target satisfy the constraints from one or more reference targets, and one or more Amino acids that satisfy the constraints from the reference target have a backbone mean square deviation (rms) of less than 8.0 Å from the reference target ( RSMD) structural homology.

[0109] In some embodiments, engineered peptides that satisfy constraints derived from one or more reference targets. amino acids with 10% to 90% sequence identity with the reference target, 20% to 90% sequence identity, 30%-90% sequence identity, 40%-90% sequence identity, 50%-90% sequence identity Sex, 60%-90% sequence identity, 70%-90% sequence identity, or 80%-90% In some embodiments, the sequence homology satisfies the constraints from one or more reference targets. The amino acid is 30Å 2 ~3000Å 2 , or 100 Å 2 ~3000Å 2 , or 25 0Å 2 ~3000Å 2 , or 500 Å 2 ~3000Å 2 , or 750 Å 2 ~3000 Å 2 , or 1000 Å 2 ~3000Å 2 , or 1250 Å 2 ~3000Å 2 ,or 1500Å 2 ~3000Å 2 , or 1750 Å 2 ~3000Å 2 , or 2000 Å 2 ~3000Å 2 , or 2250 Å 2 ~3000Å 2 , or 2500 Å 2 ~3000Å 2 , or 2750 Å 2 ~3000Å 2 having a van der Waals surface area overlap with the reference do.

[0110] The combinations of constraints that the engineered peptides satisfy are 2 or more, 3 or more, 4 or more, 5 or more. The combination may include constraints from two or more, six or more, or seven or more reference targets. As described elsewhere in the disclosure, the reference target may contain one or more constraints that do not originate from the reference target. These referential constraints, and non-referential constraints (if any), are independent of each other. The structural, kinetic, chemical, or functional properties described herein or The constraints may be any of those described herein, such as any combination of

[0111] In some embodiments, the engineered peptide has at least Such structural differences include, for example, differences in sequence, number of amino acid residues, atom number, etc. total number of secondary structures, total hydrophilicity, total hydrophobicity, total positive charge, total negative charge, one or more secondary structures, shape factor, Nike descriptors, van der Waals surfaces, structural graph nodes and edges, volumetric surfaces, electrostatics Potential surface, hydrophobic potential surface, local diameter, local surface features, skeleton model, charge density, hydrophilic density Differences in surface roughness, surface-to-volume ratio, amphiphilic density, or surface roughness, or any combination thereof. In some embodiments, a combination of one or more characteristics (such as those described herein) may be included. Differences in the characteristics of a target (e.g., two or more characteristics) compared to the characteristics of a reference target, if applicable to the type of characteristic At least 10%, at least 20%, at least 30%, at least 40%, at least At least 50%, at least 60%, at least 70%, at least 80%, at least 90% , at least 100%, or more than 100%. For example, in some embodiments, the difference is the total number of atoms, and the engineered peptides have at least 10% fewer atoms than the reference target. At least 20%, or at least 30% more atoms, or at least 1% more atoms than the reference target Some examples have 0%, at least 20%, or at least 30% fewer atoms. In an embodiment, the difference is a total positive charge, and the total positive charge of the engineered peptide is less than that of the reference target. At least 10%, at least 20%, at least 30%, at least 40%, or less In other embodiments, the manipulated pair The total positive charge of the peptide is at least 10%, at least 20%, at least 30%, at least 40%, or at least 50% smaller (e.g., less positive ).

[0112] In some embodiments, the combination of spatially defined topological constraints is Thus, in some embodiments, the manipulation comprises one or more secondary structure elements that are not present in the The generated peptide contains one or more secondary structure elements that are not present in the reference target. In this embodiment, the combined and / or engineered peptides are 1 secondary structure element, 2 secondary structure elements, 3 secondary structure elements, 4 secondary structure elements In some embodiments, each secondary structure element comprises five or more secondary structure elements. , independently selected from the group consisting of a helix, a sheet, a loop, a turn, and a coil. In some embodiments, each secondary structure element that is not present in the reference target is independently: α-helix, β-bridge, β-strand, 3 10 Helix, π-helix, It may be a turn, loop, or coil.

[0113] In some embodiments, the engineered peptides are capable of modifying a biological response or function. containing one or more related atoms, or one or more amino acids, or a combination thereof. In some embodiments, the biological response or function can be a response to a signal, such as gene expression, metabolic activity, or other signaling. Sex, protein expression, cell proliferation, cell death, cytokine secretion, kinase activity, epigenetics Tick modification, cell killing activity, inflammatory signaling, chemotaxis, tissue infiltration, immune cell lineage commitment, Tissue microenvironment modification, immune synapse formation, IL-2 secretion, IL-10 secretion, growth factor secretion, Interferon gamma secretion, transforming growth factor beta secretion, immunoreceptor tyrosine-based Activation motif activity, immunoreceptor tyrosine-based inhibitory motif activity, antibody-dependent cell Injury, complement-dependent cytotoxicity, biological pathway agonist action, biological pathway antagonist Action, biological pathway redirection, kinase cascade modification, protein degradation pathway modification, protein Protein homeostasis pathway modification, protein folding / pathway, post-translational modification pathway, metabolic pathway, gene transcription transcription / translation, mRNA degradation pathway, gene methylation / acetylation pathway, histone modification pathway, Genetic pathways, immune-dependent clearance, opsonization, hormone signaling , integrin pathways, membrane protein signaling, ion channel flux, and g - protein-coupled receptor response.

[0114] In certain embodiments, the reference target is a biological response or biological function (as described herein). The engineered peptide contains one or more atoms related to a specific biological function, such as a peptide that is associated with a specific target molecule. containing one or more atoms associated with an answer or biological function (such as those described herein). The atomic variation of that atom in the engineered peptide is compared with the atomic variation of that atom in the reference target. Thus, for example, in some embodiments, the atoms themselves are different atoms. However, their atomic variations overlap. In other embodiments, the atoms are the same atoms and their In still further embodiments, the atoms are independently the same or In some embodiments, overlap is measured by a root mean square dot product (R) greater than 0.25. In some embodiments, the overlap is greater than 0.3, greater than 0.35, Greater than, greater than 0.4, greater than 0.45, greater than 0.5, greater than 0.55 , greater than 0.6, greater than 0.65, greater than 0.7, greater than 0.75, Greater than 0.8, greater than 0.85, greater than 0.9, or greater than 0.95 In certain embodiments, the RMSIP is calculated by:

number

[0115] In some embodiments, the engineered peptides are capable of modifying a biological response or function. Related atoms or amino acids (or combinations thereof) At least a portion of the amino acids or combinations are derived from the reference target and are included in the engineered peptide. The specific constraints on the set of atoms or amino acids in the domain and the set in the reference target are given by a matrix. In some embodiments, the matrix is an L×L matrix. In a further embodiment, the matrix is an S×S×M matrix. is the I / P angle matrix

[0116] For example, in some embodiments, engineered proteins associated with a biological response or function are The atomic variations of atoms or amino acids in the selected peptide are described by an L × L matrix. The atom or amino acid portion is derived from the reference target, and the atomic variation in the reference target of the portion is , described by an L×L matrix. In some embodiments, the adjacency (amino The α, β ... For all matrix elements (i, j) of the L × L atomic fluctuation or adjacency matrix of the engineered peptide, The mean percent error (MPE) across the 2000-20 ... , the reference target atomic variation or the corresponding (i, j) element in the adjacency matrix is 75% or less. In some embodiments, the MPE is a function of the proportion of engineered peptides derived from a reference target. In this case, the ratio is less than 70%, less than 65%, less than 60% of the corresponding element in the reference target matrix. less than 55%, less than 50%, less than 45%, or less than 40%. In some embodiments, L is the number of amino acid positions, and the atomic variation matrix elements (i , j) values are the i-th and j-th interatomic distances, respectively, when the (i, j) interatomic distance is 7 Å or less. The sum of intramolecular atomic variations for amino acids, or the (i, j) interatomic distance is 7 Å is zero if (i, j) is greater than 1 or if (i, j) is on the diagonal. In some embodiments, the interatomic distances are expressed as atomic fluctuation matrix elements, instead of multipliers of 0 or 1. (i, j). In certain embodiments, the i-th and j The atomic fluctuations and distances are determined by molecular simulations (e.g., molecular dynamics) and / or or laboratory measurements (e.g., NMR). In an embodiment, L is the number of amino acid positions, and the value of the adjacency matrix element (i, j) is the number of amino acid positions. If the distance between the i-th and j-th amino acids is less than 7 Å, the intramolecular atoms If the interatomic distance is greater than 7 Å, or if (i, j) is a diagonal If it is on a line, it is zero. Alternatively, in some embodiments, the interatomic distance is 0 or or a multiplier of 1, can serve as a weighting factor for the adjacency matrix element (i, j). In certain embodiments, the distance between the i-th and j-th atoms is determined by molecular simulation ( can be determined by molecular dynamics) and / or laboratory measurements (e.g., NMR) do.

[0117] In certain embodiments, atoms or atoms in the engineered peptide that are relevant to a response or function are The amino acids are then combined with topologically constrained chemical descriptor vectors and engineered peptides derived from the reference target. The proportion of chemical descriptors should have an average error of less than 75% relative to the reference described by the same chemical descriptor. Each i-th element in the chemical descriptor vector is assigned an amino acid position index. In some embodiments, the MPE corresponds to an engineered peptide derived from a reference target. The proportion of references described by the same chemical descriptor was less than 70%, less than 65% Full, less than 60%, less than 55%, less than 50%, less than 45%, or less than 40%. Typical vectors are shown in FIG.

[0118] In yet a further embodiment, the matrix is an L×2 phi / psi matrix and is manipulated The atoms or amino acids in a peptide that are relevant for response or function are engineered from the reference target. The percentage of peptides analyzed had an MPE of less than 75% relative to the reference phi / psi angle matrix. where L is the number of amino acid positions, and the phi and psi values are of dimension (L, 1 ) and (L,2). In some embodiments, the MPE is a manipulation derived from a reference target. The percentage of peptides generated was less than 70% and 65% of the reference phi / psi angle matrix. %, less than 60%, less than 55%, less than 50%, less than 45%, or less than 40%. In some embodiments, the phi / psi value is determined by molecular simulation (e.g., molecular dynamics). dynamics), knowledge-based structure prediction, or laboratory measurements (e.g., NMR). An exemplary L×2 phi / psi matrix is shown in FIG.

[0119] In some embodiments, the matrix is an S×S×M secondary structure element interaction matrix, The atoms or amino acids in the peptides that are relevant for response or function are derived from the reference target. The proportion of engineered peptides that matched the reference secondary structure element relationship matrix was less than 75%. where S is the number of secondary structure elements and M is the number of interaction descriptors. In some embodiments, the MPE is an engineered peptide derived from a reference target. The ratio of the secondary structure element relationship matrix to the reference secondary structure element relationship matrix is less than 70%, less than 65%, and less than 60%. The interaction descriptors are: full, less than 55%, less than 50%, less than 45%, or less than 40%. , e.g., hydrogen bonding, hydrophobic packing, van der Waals interactions, ionic interactions , covalent crosslinks, chirality, orientation, or distance, or any combination thereof In the secondary structure element interaction matrix index, (i, j, m) is the i-th and j-th secondary structure element. The m-th interaction descriptor value between the next structural elements. An exemplary S×S×M matrix is shown in Figure 50. is shown.

[0120] The mean percent error (MPE) for the different matrices described herein is calculated by: It can be calculated,

number

[0121] In some embodiments, the engineered peptide has an M of less than 75% compared to a reference target. In certain embodiments, the engineered peptide has a 70% PE as compared to the reference target. Less than, less than 65%, less than 60%, less than 55%, less than 50%, less than 45%, or less than 40% In some embodiments, the MPE is a total topological constraint distance (TCD) of , Topological Clustering Coefficient (TCC), Euclidean distance, Power distance, Soager distance distance, Canberra distance, Sorensen distance, Jaccard distance, Mahalanobis distance, Hamin The correlation is determined by the metric distance, quantitative estimation of likelihood (QEL), or chain topology parameter (CTP). It is determined.

[0122] a. Secondary structure elements In some embodiments, at least a portion of the engineered peptide comprises one or more secondary In some embodiments, the engineered peptide is topologically constrained by structural elements. Atoms or amino acids associated with a biological response or biological function are organized into one or more secondary structure elements. In some embodiments, the secondary structure elements are independently topologically constrained to In some embodiments, the bilayer is a turn, a helix, a turn, a loop, or a coil. The secondary structural elements are independently α-helices, β-bridges, β-strands, and 3 10 helicopter In certain embodiments, the amino acid sequence is a helix, a π-helix, a turn, a loop, or a coil. At least a portion of the engineered peptide is topologically constrained by one or more secondary structure elements. In some embodiments, at least one of the engineered peptides is present in the reference target. The sequence is topologically constrained in part by the combination of secondary structure elements, each of which independently The amino acid sequence is selected from the group consisting of a chain, a helix, a turn, a loop, and a coil. In some embodiments, each element may independently be an α-helix, a β-bridge, a β-strand, or , 3 10 Helix, π-helix, turn, loop, and coil. It is selected.

[0123] In some embodiments, the secondary structural elements are parallel or antiparallel sheets. In some embodiments, the sheet secondary structure comprises two or more residues. The sheet secondary structure comprises 50 residues or less. The structure contains 2 to 50 residues. The sheets can be parallel or antiparallel. In an embodiment, the parallel sheet secondary structure is formed by two parallel strands i, j (i and j strands). N-termini in opposite orientations of the strands), and those with a hydrogen bonding pattern of residues i:j. In some embodiments, the antiparallel sheet secondary structure can also be described as an antiparallel double stranded structure. two strands i and j (the N-termini of the i and j strands in the same orientation), and residues i:j i:j+1 hydrogen bonding pattern. In this state, strand orientation and hydrogen bonding are determined by knowledge-based simulations or analytical methods. It can be determined by molecular dynamics simulations and / or laboratory measurements.

[0124] In some embodiments, the secondary structure element is a helix. A helix is a right-handed In some embodiments, the helix is 2.5 to 6.0. with residues per turn (residues / turn) values, and pitches between 3.0 Å and 9.0 Å. In some embodiments, the residues / turns and pitches are determined by knowledge-based simulations. or determined by molecular dynamics simulations and / or laboratory measurements.

[0125] In some embodiments, the secondary structure element is a turn. The turn comprises 2 to 7 residues and one or more inter-residue hydrogen bonds. In certain embodiments, the turn comprises two, three, or four inter-residue hydrogen bonds. The turn is knowledge-based simulation or molecular dynamics simulation, and / or determined by laboratory measurements.

[0126] In still further embodiments, the secondary structural element is a coil. The IL contains 2-20 residues and zero predicted inter-residue hydrogen bonds. In an embodiment, these coil parameters are determined using knowledge-based simulation or analysis. determined by molecular dynamics simulations and / or laboratory measurements.

[0127] In yet a further embodiment, the engineered peptide comprises one or more precursors derived from a reference target. Some of the atoms or amino acids have secondary structure. In embodiments, these atoms or amino acids are associated with a biological response or biological function. In some embodiments, the secondary structure of atoms or amino acids in the engineered peptides The engineered motif vector is a vector that measures the proportion of engineered peptides derived from the reference target. have a cosine similarity greater than 0.25 to the target secondary structure motif vector, The length of the vector is the number of secondary structure motifs, and the value at the ith vector position is the number of lookups. Define the identity of secondary structure motifs (e.g., helices, sheets) derived from the table. In some embodiments, each motif comprises two or more amino acids. In this embodiment, the motif may be, for example, an α-helix, a β-bridge, a β-strand, a 3 10 Helices, π-helices, turns, and loops. Cosine similarity is the ratio of the proportion of engineered peptides derived from the reference target to the proportion of engineered peptides derived from the reference target. For secondary structure motif vectors, the values are greater than 0.3, greater than 0.35, and greater than 0.4. Exemplary secondary structure indices and lookups are: A lookup table is provided in Figure 53. Cosine similarity can be calculated by: ,

number

[0128] In some embodiments, one or more atoms of an engineered peptide derived from a reference target or amino acids are matched to the corresponding reference target atoms or In some embodiments, the engineered amino acid sequence derived from the reference target may be compared to the The total TCD of a peptide atom or amino acid is calculated relative to the TCD distance of the corresponding atom in the reference target. The two intramolecular topological constraints are within + / - 75% of the pairwise distance between them. In some embodiments, the manipulated The atoms or amino acids in a peptide are associated with a biological function or response. In some embodiments, the pairwise distance between the i-th and j-th atoms or amino acids is are based on molecular simulations (e.g., molecular dynamics) and / or laboratory measurements (e.g., An exemplary method for calculating the total topological constraint distance (TCD) is The equation is:

number

[0129] In some embodiments, one or more atoms of an engineered peptide derived from a reference target or amino acids are mapped to the corresponding reference target atoms or In some embodiments, the engineered peptide atom or The CTP of an amino acid is + / - relative to the CTP of the corresponding atom or amino acid in the reference target. 50%, and the intrastrand topological interactions are at pairwise distances of 7 Å or less. In embodiments, the atoms or amino acids in the engineered peptides being compared are selected from the group consisting of: In some embodiments, the i-th and j-th pairs are related to a biological function or response. The wise distances are, in some embodiments, calculated using molecular simulations (e.g., molecular dynamics). and / or can be determined by laboratory measurements (e.g., NMR). An exemplary equation for is:

number

[0130] In some embodiments, one or more atoms of an engineered peptide derived from a reference target or amino acids are compared with the corresponding reference target atoms or amino acids using a quantitative evaluation of likelihood (QEL). can be compared to an amino acid. In some embodiments, the engineered peptide atom or The QEL of an amino acid is + / - the QEL of the corresponding atom or amino acid in the reference target. In some embodiments, the atoms in the engineered peptides being compared are or amino acids associated with a biological function or biological response. An exemplary equation for is:

number

[0131] In some embodiments, one or more atoms of an engineered peptide derived from a reference target or amino acids are calculated using the topological clustering coefficient (TCC) vector and the mean error rate (MP E) can be used to compare the corresponding reference target atoms or amino acids. In this form, the TCC vector and MPE are the positions of the corresponding atoms or amino acids in the reference target. Each element (i) of the vector is the amino acid at the i-th position. is the topological clustering coefficient, and the intramolecular cluster is within 7 Å from the i-th amino acid position. It is defined by the following interaction edge distance and two edges: ij, jl. In some embodiments, the atoms or amino acids in the engineered peptides being compared are , associated with a biological function or biological response. In some embodiments, In some embodiments, the jth and lth edge distances are calculated using the molecular simulation (e.g., molecular dynamics) and / or laboratory measurements (e.g., NMR) An exemplary equation for evaluating the topological clustering coefficient of the ith location is: Below,

number

[0132] In yet a further embodiment, one or more atoms of an engineered peptide derived from a reference target Or amino acids are expressed as L × M topological constraint matrices, and Euclidean distance, power distance, and so on. Lager distance, Canberra distance, Sorensen distance, Jaccard distance, Mahalanobis distance , or the mean percent error (MPE) of the Hamming distance across all M dimensions, The L×M matrix element (l, m) is the l-th where L is the number of amino acid positions, and M is is the number of distinct topological constraints. In some embodiments, the engineered peptide L×M matrix The MPE of the corresponding reference target atom or amino acid matrix is less than 75%. In some embodiments, the MPE is less than 70%, less than 65%, less than 60%, less than 55%, In some embodiments, the engineered The atoms or amino acids in the selected peptides are associated with a biological function or biological response. An exemplary LxM matrix is shown in Figure 52.

[0133] III. Programmable in vitro selection In other embodiments, a series of programmed selection steps is used to select binding partners. In so doing, methods of using the engineered peptides described herein are further described herein. At least one selection step is provided to identify potential binding partners with the engineered peptide. This involves evaluating the interaction of pools of toner.

[0134] In some embodiments, two or more selection molecules are used to guide the selection of binding molecules. In some embodiments, the method comprises: and subjecting the pool to at least one selection round, each round comprising subjecting at least one of the pool to at least one selection round. at least one negative selection step, in which a portion is screened against a negative selection molecule; and at least a portion of the pool is screened for positive selection molecules. In some embodiments, the method comprises at least two steps of: rounds, at least 3 rounds, at least 4 rounds, at least 5 rounds, At least 5 rounds, at least 6 rounds, at least 7 rounds, at least 8 rounds At least nine rounds, at least ten rounds, or more, with each round The steps independently undergo at least one negative selection step and at least one positive selection step. In some embodiments, each round independently comprises two or more negative selection steps. The method includes one or more positive selection steps, or two or more positive selection steps, or a combination thereof. provides an exemplary schematic illustrating three selection rounds, with the first and third rounds The first round contains two or more negative selection steps, and the second round contains two or more positive selection rounds. As shown in the scheme, two negative selection molecules (baits) are Three negatively selected molecules are used in the first round, and three negatively selected molecules are used in the third round. One positively selected molecule is used in the first round.

[0135] In some embodiments, the method comprises two or more rounds, wherein each negative and positive selection molecule is In other embodiments, the same negative selection molecule or the same positive selection molecule are independently selected. , or a combination thereof may be used in two or more rounds. For example, FIG. During this time, the same negative selection molecule used in round 1 is used again in round 3, with additional A third negatively selected molecule is also included in round 3. The order of the negative and positive selection steps can be In certain embodiments, they may be selected independently within each selection round. In some embodiments, the method comprises one or more selection rounds, each round comprising: In another embodiment, the method comprises first a negative selection step and then a positive selection step. It involves one or more selection rounds, each round consisting of an initial positive selection step followed by a negative selection step. In yet a further embodiment, the method comprises one or more selection rounds, Each round independently includes a negative selection step and a positive selection step, In the method, the negative selection step can be independently performed before or after the positive selection step. This is after the game.

[0136] Such methods of selection may be directed towards a particular desired property, such as binding specificity or binding affinity. To derive a library of candidate binding molecules, and negative (-) steps are used. Multiple steps using both positive and negative selection molecules are used. By using a combination of nucleotide sequences, the pool of candidates is balanced to favor desirable and undesirable traits. Furthermore, in some embodiments, Now, the order of each step within each round, and the order of rounds relative to each other, is a function of the choice Thus, for example, in some embodiments: A method involving one round of (+) selection followed by (-) selection is where the (-) selection is first. This results in a different final pool of candidates than if (+) selection were to follow. Applied to methods involving rounds, the order of selection steps is such that the same positive and negative selection molecules are Even if used in its entirety, it may result in a different final pool of selected candidates.

[0137] In some embodiments, a selection molecule is used that has the opposite properties of another selection molecule. were identified using positive selection molecules (or excluded due to negative selection molecules), for example. (i) candidate binding partners possessing the desired trait ( to ensure that animals are identified (or excluded) for undesirable traits It can be useful for removing binding partners that are bound through irrelevant interactions. To do this, the selected molecule is stripped of residues / structures that convey the desired (or undesired) trait. Reverse selection molecules having similar or identical structures and properties to the positive selection molecules can be used. If interaction with a particular charge pattern in a selected molecule is desired, the charge pattern is provided. Reverse negative selectivity, where the residues providing the selectivity are replaced with uncharged residues and / or residues of the opposite charge. Thus, for a particular selected molecule, multiple different corresponding The reverse selection molecule may be possible.

[0138] In the selection methods provided herein, at least one of the selection molecules is a molecule described herein. In some embodiments, two or more engineered peptides are In some embodiments, each engineered peptide is independently positive or negative. is a negative selection molecule. In certain embodiments, each selection molecule used in one or more selection rounds The selected molecules are independently engineered peptides. In other embodiments, the engineered peptides At least one molecule that is not a peptide is used as the selection molecule. Such a selected molecule may include, for example, a naturally occurring polypeptide, or a portion thereof. In other embodiments, the one or more selected molecules that are not engineered peptides may be, for example, It may also include non-naturally occurring polypeptides, or portions thereof. For example, in some embodiments In one embodiment, one or more selection molecules (e.g., positive selection molecules or negative selection molecules) are selected from immunogens, antibodies, , cell surface receptor, or transmembrane protein, or signal transduction protein, or is a multiprotein complex, or a peptide-protein complex, or any part thereof or any combination thereof. In some embodiments, one or more selected molecules The children were PD-1, PD-L1, CD25, IL2, MIF, CXCR4, or VEG F, or any part thereof, or antibodies to any of these (bevacizumab) These include rheumatoid arthritis (rheumatoid arthritis), rheumatoid arthritis (e.g., ...

[0139] The positive and negative traits that are being selected for or against at each step are selected from a range of traits. may be selected and adjusted depending on the desired characteristics of the final binding molecule or molecules obtained. Such desired characteristics may depend, for example, on the intended use of the one or more binding molecules. For example, in some embodiments, antibody candidates are identified using the methods provided herein by: Advantages include one or more positive properties, such as high specificity, and one or more negative properties, such as cross-reactivity. What is considered a positive trait in one situation may be a negative trait in another. It should be understood that a Thus, the positively selected molecules in a series of selection rounds may in some embodiments be in different series of selection rounds or in the selection of different types of binding molecules or the same type of binding molecule but in a selection for a different purpose, as a negative selection molecule. It is possible.

[0140] In some embodiments, each selection feature is independently an amino acid sequence, a polypeptide secondary Multiple characteristics of structure, molecular dynamics, chemical features, biological function, immunogenicity, and reference target(s) Cross-species, cross-species reference target reactivity, desired reference target versus undesired reference target(s) Selectivity, sequence and / or structural homology of a family of reference targets (multiple) selectivity of reference target(s) with similar protein function, high sequence and and / or from a larger family of structurally homologous unwanted targets. Selectivity of the desired reference target(s), for distinct reference target alleles or mutations Selectivity, selectivity for chemical modification at the distinct reference target residue level, selectivity for cell type, Selectivity for tissue type, selectivity for tissue environment, selectivity for structural diversity of reference target(s) tolerance to sequence variability of the reference target(s), as well as tolerance to sequence variability of the reference target(s). In some embodiments, each of the selected In other embodiments, the two or more selection properties are different types. For example, in some embodiments, two or more selectable The selective property is the polypeptide secondary structure, and one is the selection for the desired polypeptide secondary structure. One is a positive selection and the other is a negative selection against undesired polypeptide secondary structures. In some embodiments, the two or more selection properties are selectivity for a cell type and positive selection A selective property is a property that is selective for a particular desired cell type, whereas a negative selective property is a property that is selective for a particular desired cell type. In some embodiments, two or more, three or more, four or more, The more, five or more, or six or more selective characteristics are of the same type.

[0141] In yet another aspect, a composition comprising two or more selection-derived polypeptides is provided herein. and each polypeptide independently binds to a positive selection molecule containing one or more positive induction properties, or is a negative selection molecule that contains one or more negative inductive properties. Such properties are In embodiments, amino acid sequences, polypeptide secondary structures, molecular dynamics, chemical characteristics, biological Function, immunogenicity, multispecificity of reference target(s), cross-species reference target reactivity, undesirable Selectivity, sequence and / or sequence of the desired reference target(s) relative to the undesired reference target(s). Selectivity of reference target(s) within a structurally homologous family, with similar protein function The desired target has high sequence and / or structural homology to the reference target(s). selectivity of distinct desired reference target(s) from a larger family of unrelated targets; Selectivity for distinct reference target alleles or mutations, distinct reference target residue level chemistry Selectivity for modification, cell type, tissue type, tissue environment Selectivity, tolerance to structural diversity of the reference target(s), sequence diversity of the reference target(s) from the group consisting of tolerance to kinetic diversity of the reference target(s). can be selected.

[0142] Thus, in a further aspect, the selection inducer compositions described herein are used to identify binding moieties. Provided herein is a method for screening a library of antibodies, wherein each selection round comprises a protease inhibitor. a negative selection step of screening at least a portion of the molecules against the negative selection molecule; a positive selection step in which at least a portion of the pool is screened in favor of positively selected molecules; The order of selection steps within each round, and the order of rounds, may include alternative orderings. This results in the selection of a subset of the pool that is different from the

[0143] In some embodiments, the compositions of the selection-derived polypeptides described herein, as well as The binding partners being evaluated using the screening methods described herein are , a phage library, e.g., a Fab-containing phage library, or a cell library , for example, a B cell library or a T cell library.

[0144] In some embodiments of the screening methods provided herein, the method comprises screening two or more Includes three or more, four or more, five or more, six or more, or seven or more selection rounds. In some embodiments where there is more than one round, each round involves the selection of a different selection molecule. In other embodiments where there are more than two rounds, the set includes at least two rounds. The rounds contain the same negative selection molecule, the same positive selection molecule, or both.

[0145] In some embodiments of the screening method, the method comprises: , and analyzing a subset of the pools. In certain embodiments, each subset pool The analysis was performed independently on peptide / protein biosensor binding, peptide / protein E LISA, peptide library binding, cell extract binding, cell surface binding, cell activity assay , cell proliferation assay, cell death assay, enzyme activity assay, gene expression profile, Protein modification assays, Western blots, and immunohistochemistry. In some embodiments, gene expression profiles are obtained using next generation sequencing, etc. In some embodiments, the statistical analysis includes full sequence repertoire analysis of a subset pool of and / or information scoring or machine learning training to identify one or more selection labels. evaluate one or more subsets of the pools in the field.

[0146] In some embodiments, the identity of positively and / or negatively selected molecules for subsequent rounds is determined by the The uniformity and / or order of analyzing a subset pool from one selection round In some embodiments, statistical and / or information scoring is used. , or using machine learning training to select one or more of the pools in one or more selection rounds. Evaluate a subset to determine the number of participants in a subsequent round (e.g., the next round, or within the program). the identity and / or identity of the positively and / or negatively selected molecules for further rounds, etc. determines the order.

[0147] In still further embodiments, the selection method comprises: This includes modifying the subset pool obtained from the target. Such modifications include, for example, , genetic mutations in subset pools, genetic depletion of subset pools (e.g., during selection) (selecting a subset of a subset pool to proceed with) Gene enrichment (e.g., increasing the size of the pool), at least chemical modification of at least a portion of the subset pool, or enzymatic modification of at least a portion of the subset pool; Any combination thereof may be included. In some embodiments, statistical and / or Uses information scoring or machine learning training to evaluate and select subset pools. determining one or more modifications to make to the modified subset pool before proceeding further. In certain embodiments, such statistical and / or information scoring, or Use machine learning training to identify positively and / or negatively selected molecules for subsequent selection rounds. Determine identity and / or order.

[0148] Any suitable assay can be used to identify binding partners with the selected molecule at each step. In some embodiments, binding can be assessed by, for example, binding to a binding partner. The activity of the antibody is directly assessed by directly detecting the label of the antibody. Such labels include, for example, In other embodiments, the fluorescent label may include a fluorescent label such as a fluorophore or a fluorescent protein. For example, binding can be assessed indirectly using a sandwich assay. In the assay, the binding partner binds to the molecule of choice, and then a secondary labeled reagent binds to the selected molecule. A secondary labeling reagent is added to label the bound binding partner. This secondary labeling reagent is then detected. Examples of sandwich assay components include anti-His tag antibodies or His tag-specific antibodies. His-tag binding partner, labeled streptavidin, detected with a specific fluorescent probe or biotin-labeled binding partners detected with labeled avidin, or anti-binding partners Examples include unlabeled binding partners detected with a toner antibody.

[0149] In some embodiments, the binding partners selected at each step may be any number of Available detection methods are used to identify them based on binding signals or dose responses. These detection methods include, for example, imaging, fluorescence activated cell sorting (FACS), mass spectrometry, or In some embodiments, a hit threshold (e.g., a signal (for the null median) is defined, and any signal above that , flagged as a putative hit motif.

[0150] IV. Use of Engineered Peptides to Generate Antibodies Engineered peptides provided herein and identified by the methods provided herein The method can be used, for example, to generate one or more antibodies. The antibody may be a monoclonal or polyclonal antibody. In this embodiment, antibodies generated by immunizing an animal with an immunogen are used herein. As provided herein, the immunogen is an engineered peptide as provided herein. In certain embodiments, the animal is a human, rabbit, mouse, hamster, monkey, etc. In embodiments, the monkey is a cynomolgus monkey, a macaque monkey, or a rhesus monkey. Immunizing an animal with a peptide can be achieved by, for example, administering the peptide and optionally an adjuvant. The method may include administering to the animal at least one dose of a composition comprising the compound. In some embodiments, generating antibodies from an animal includes isolating antibody-expressing B cells. Some embodiments involve fusing B cells with myeloma cells to express antibodies. In some embodiments, the engineered peptide is a nucleotide sequence that is a sequence of a polypeptide that is a nucleotide sequence that is a hybridoma. Antibodies generated using the antibody are known to cross-react with humans and monkeys, e.g., cynomolgus monkeys. This can be done.

[0151] The description provided herein describes numerous example configurations, methods, parameters, etc. However, such descriptions are not intended as limitations on the scope of the present disclosure. It should be appreciated that the present disclosure is not intended to be limiting and is instead provided as a description of exemplary embodiments. do.

[0152] Illustrative Embodiments Embodiment I-1. An engineered peptide, wherein the engineered peptide is between 1 kDa and 1 100 kDa molecular weight, containing up to 50 amino acids, and an engineered peptide a combination of spatially related topological constraints, one or more of which are related to the reference target It is a constraint of origin, 10%–98% of the amino acids in the engineered peptides are constrained by one or more reference targets. Fulfill, Amino acids satisfying one or more of the constraints from the reference target have a backbone alignment of less than 8.0 Å with the reference target. Engineered peptides with root mean square deviation (RSMD) structural homology.

[0153] Embodiment I-2. Amino acids satisfying constraints from one or more reference targets are compared to the reference targets. 1. The engineered peptide of embodiment I-1, having a sequence homology of 1% to 90%.

[0154] Embodiment I-3. Amino acids satisfying constraints from one or more reference targets are within 30 Å 2 ~30 00Å 2 having a van der Waals surface area overlap with the reference of embodiment I-1 or I- 2. The engineered peptide according to claim 2.

[0155] Embodiment I-4. An embodiment in which the combination includes constraints from at least two reference targets The engineered peptide according to any one of aspects I-1 to I-3.

[0156] Embodiment I-5. An embodiment in which the combination includes constraints from at least five reference targets The engineered peptide according to any one of aspects I-1 to I-4.

[0157] Embodiment I-6. The combination of constraints includes one or more constraints that do not originate from the reference target. An engineered peptide according to any one of embodiments I-1 to I-5.

[0158] Embodiment I-7. One or more non-reference target-derived constraints are used to provide desired structural, kinetic, or chemical Embodiment I- describes a therapeutic or functional property, or any combination thereof. 6. The engineered peptide according to claim 6.

[0159] Embodiment I-8. Embodiment I-1, wherein the constraints are independently selected from the group consisting of: 1. An engineered peptide according to any one of claims 1 to 1-7. interatomic distance, Atomic fluctuations, atomic energy, chemical descriptors, solvent exposure, amino acid sequence similarity, bioinformatics descriptors, non-covalent tendency, Phi angle, Psi angle, van der Waals radius, secondary structure tendency, amino acid contiguity, and Amino acid contact.

[0160] Embodiment I-9. Embodiments I-1 to I-I, wherein one or more constraints are, independently, atomic variations. -8. The engineered peptide according to any one of claims 1 to 8.

[0161] Embodiment I-10. Embodiment I-1, Wherein One or More Constraints Are, Independently, Chemical Descriptors 10. The engineered peptide according to any one of claims 1 to 1-9.

[0162] Embodiment I-11. The embodiment I-1, wherein one or more constraints are, independently, interatomic distances. 1. The engineered peptide of any one of claims 1 to 1-10.

[0163] Embodiment I-12. Embodiment I-1 to I-2, wherein one or more constraints are, independently, secondary structure. 1-11. An engineered peptide according to any one of claims 1-11.

[0164] Embodiment I-13. An embodiment in which one or more constraints are, independently, van der Waals surfaces The engineered peptide according to any one of embodiments I-1 to I-12.

[0165] Embodiment I-14. One or more constraints independently affect a biological response or biological function. A related engineered peptide according to any one of embodiments I-1 to I-13.

[0166] Embodiment I-15. Comprising one or more atoms associated with a biological response or biological function , The engineered peptide according to any one of embodiments I-1 to I-14.

[0167] Embodiment I-16. One or more amino acids associated with a biological response or biological function The engineered peptide of any one of embodiments I-1 to I-15, comprising

[0168] Embodiment I-17. The biological response or function is a function of gene expression, metabolic activity, Protein expression, cell proliferation, cell death, cytokine secretion, kinase activity, epigenetics Modification, cell death activity, inflammatory signaling, chemotaxis, tissue infiltration, immune cell lineage commitment, tissue microenvironment Environmental modification, immune synapse formation, IL-2 secretion, IL-10 secretion, growth factor secretion, interleukin-1 Feron gamma secretion, transforming growth factor beta secretion, and immunoreceptor tyrosine-based activation Motif activity, immunoreceptor tyrosine-based inhibitory motif activity, antibody-dependent cellular cytotoxicity, complement Body-dependent cytotoxicity, biological pathway agonist action, biological pathway antagonist action, Biological pathway redirection, kinase cascade modification, protein degradation pathway modification, proteostasis Sexual pathway modification, protein folding / pathway, post-translational modification pathway, metabolic pathway, gene transcription / translation , mRNA degradation pathway, gene methylation / acetylation pathway, histone modification pathway, epigenetic Tick pathway, immune-dependent clearance, opsonization, hormone signaling, and inte Glycine pathway, membrane protein signaling, ion channel flux, and g-proteins Any of embodiments I-14 to I-16, wherein the protein-coupled receptor response is selected from the group consisting of 1. An engineered peptide according to claim 1.

[0169] Embodiment I-18. The reference target is one or more associated biological responses or biological functions containing atoms of One or more atoms in an engineered peptide that are associated with a biological response or biological function The atomic variation of one or more atoms in the reference target that is associated with a biological response or biological function. The engineered peptide of embodiment I-15, wherein the atomic variations of

[0170] Embodiment I-19. The overlap is greater than 0.25 root mean square dot product (RMSIP). An engineered peptide according to embodiment I-18.

[0171] Embodiment I-20. The overlap has a root mean square dot product (RMSIP) greater than 0.75. The engineered peptide of embodiment I-19, comprising:

[0172] Embodiment I-21. In engineered peptides associated with a biological response or biological function at least a portion of the atoms of the reference target are topologically constrained to secondary structure elements in the reference target; The engineered peptide according to any one of aspects I-18 to I-20.

[0173] Embodiment I-22. The method of embodiment I-21, wherein the secondary structure element is a beta-sheet. engineered peptides.

[0174] Embodiment I-23. The compound of embodiment I-21, wherein the secondary structure element is an alpha helix. The engineered peptides described.

[0175] Embodiment I-24. The secondary structure element is a turn, the turn comprising 2 to 7 residues The engineered peptide of embodiment I-21, comprising at least one inter-residue hydrogen bond. Do.

[0176] Embodiment I-25. The secondary structure element is a coil, and the coil contains 2 to 20 residues. 1-22. The engineered peptide of embodiment I-21.

[0177] Embodiment I-26. The coil of embodiment I-25, wherein the coil does not contain inter-residue hydrogen bonds. Engineered peptides.

[0178] Embodiment I-27. In engineered peptides associated with a biological response or biological function At least a portion of the atoms of the and a combination of two or more secondary structure elements independently selected from the group consisting of: The engineered peptide according to any one of embodiments I-21 to I-26, subject to the following restrictions:

[0179] Embodiment I-28. An embodiment in which one or more spatially related topological constraints are interatomic distances. The engineered peptide according to any one of embodiments I-1 to I-27.

[0180] Embodiment I-29. One or more spatially related topological constraints are atomic energies , The engineered peptide according to any one of embodiments I-1 to I-28.

[0181] Embodiment I-30. Each atomic energy is independently a pairwise attractive energy between two atoms. energy, pairwise repulsion energy between two atoms, atomic level solvation energy, Pairwise charge attraction energy between two atoms, pairwise hydrogen bond attraction between two atoms or non-covalent bond energy. peptide.

[0182] Embodiment I-31. An embodiment in which one or more spatially related topological constraints are chemical descriptors. The engineered peptide according to any one of embodiments I-1 to I-30.

[0183] Embodiment I-32. Each chemical descriptor independently comprises hydrophobicity, polarity, volume, net charge, lo gP, high performance liquid chromatography retention, or van der Waals radius. The engineered peptide according to embodiment I-31.

[0184] Embodiment I-33. One or more spatially related topological constraints are bioinformatic descriptors , The engineered peptide according to any one of embodiments I-1 to I-32.

[0185] Embodiment I-34. Each bioinformatic descriptor independently includes BLOSUM similarity, pKa, z Scale, Cruciani trait, Chidera factor, VHSE scale, ProtFP, MS- WHIM score, T scale, ST scale, transmembrane tendency, protein-buried area, Helix propensity, sheet propensity, coil propensity, turn propensity, immunogenic propensity, antibody epitope The engineered peptide of embodiment I-33, wherein the peptide is a protein-interface-occurring or protein-interface-occurring peptide. .

[0186] Embodiment I-35. An embodiment in which one or more spatially related topological constraints is solvent exposure. The engineered peptide of any one of Forms I-1 to I-34.

[0187] Embodiment I-36. At least one of the constraints derived from the one or more reference targets is a GPC The engineered antibody according to any one of embodiments I-1 to I-35, wherein the R extracellular domain is peptide.

[0188] Embodiment I-37. At least one of the constraints from the one or more reference targets is an ion The method according to any one of embodiments I-1 to I-36, wherein the channel extracellular domain The peptides were

[0189] Embodiment I-38. At least one of the constraints from one or more reference targets is tamper-evident. Protein-protein or peptide-protein interfacial bonding, 37. The engineered peptide of any one of claims 37 to 37.

[0190] Embodiment I-39. At least one of the constraints from one or more reference targets is The engineered peptide of any one of embodiments I-1 to I-38, derived from a polymorphic region. Chid.

[0191] Embodiment I-40. A compound comprising one or more atoms associated with a biological response or function. , each of the one or more atoms independently selected from carbon, oxygen, nitrogen, hydrogen, sulfur, phosphorus, sodium, Iron, potassium, zinc, manganese, magnesium, copper, iron, molybdenum, and nickel The engineered compound according to any one of embodiments I-1 to I-39, selected from the group consisting of: peptide.

[0192] Embodiment I-41. One or more amino acids associated with a biological function or biological response Each of the one or more amino acids independently constitutes a naturally occurring amino acid that makes up a protein. In one embodiment, the amino acid is a natural amino acid that does not form proteins, or a chemically synthesized non-natural amino acid. An engineered peptide according to any one of embodiments I-1 to I-40.

[0193] Embodiment I-42. The engineered peptide exhibits at least one structural alteration compared to a reference target. The engineered peptide according to any one of embodiments I-1 to I-41, having a nucleotide sequence similar to that of the engineered peptide of any one of embodiments I-1 to I-41.

[0194] Embodiment I-43. The at least one structural difference is, independently, in sequence, number of amino acid residues, Total number of atoms, total hydrophilicity, total hydrophobicity, total positive charge, total negative charge, one or more secondary structures, shape factor, Zernike descriptors, van der Waals surfaces, structural graph nodes and edges, volumetric surfaces, Electrostatic potential surface, hydrophobic potential surface, local diameter, local surface features, skeleton model, charge density, hydrophilicity a surface to volume ratio, an amphiphilic density, and a surface roughness. The engineered peptide according to embodiment I-42

[0195] Embodiment I-44. Differences in one or more secondary structures are engineered compared to a reference target. the presence of one or more additional secondary structure elements in the peptide, each additional secondary structure element being Independently, they consist of alpha helices, beta-sheets, loops, turns, and coils. 17. The engineered peptide according to embodiment I-16, wherein the engineered peptide is selected from the group consisting of:

[0196] Embodiment I-45. Between 10% and 90% of the amino acids are topologically derived from one or more non-reference targets The engineered peptide according to any one of embodiments I-1 to I-44, which meets the constraints.

[0197] Embodiment I-46. One or more non-reference target-derived topological constraints are used to define pre-specified functions. The engineered peptide of embodiment I-45, which enhances

[0198] Embodiment I-47. Non-reference-derived topological constraints may reinforce secondary structure elements in the reference-derived fraction of peptides. It stabilizes Non-reference-derived topological constraints enforce atomic variations in the reference-derived fraction of peptides. Non-reference derived topological constraints modify the overall peptide hydrophobicity. Non-reference-derived topological constraints alter peptide solubility. Non-reference-derived topological constraints modify the peptide net charge. Non-reference-derived topological constraints facilitate detection in labeled or label-free assays. make it possible, Non-reference derived topological constraints allow detection in in vitro assays. Non-reference derived topological constraints allow detection in in vivo assays. Non-reference-derived topological constraints enable capture from complex mixtures. Non-reference-derived topological constraints enable enzymatic processing. Non-reference-derived topological constraints enable cell membrane permeability. non-reference-derived topological constraints allow binding to secondary targets, and The engineered polypeptide of embodiment I-46, wherein the non-reference-derived topological constraints modify immunogenicity. Selected peptides. Embodiment I-48. A method for selecting engineered peptides, comprising: Identifying one or more topological characteristics of the reference target; Each of the three sets of spatially related topological constraints derived from the reference target is generated. Designing spatially related constraints on topological properties; The spatially related topological features of the candidate peptides are compared with the spatially related topological features derived from the reference target. and comparing it with the combination of topological constraints that Spatially overlapping with the set of spatially relevant topological constraints derived from the reference target. Selecting candidate peptides with relevant topological properties to generate engineered peptides A method including:

[0199] Embodiment I-49. The overlap between each feature is independently calculated as the total topological constraint distance (TCD), TCC, Euclidean distance, power distance, Soerger distance, Chambler distance, Sorensen distance, Jaccard distance, Mahalanobis distance, Hamming distance , quantitative estimation of likelihood (QEL), or chain topology parameter (CTP) and a mean percent error (MPE) of 75% or less as determined by The method described below.

[0200] Embodiment I-50. One or more constraints include per-residue energy, per-residue interactions, for, per residue variation, per residue interatomic distance, per residue chemical descriptor, per residue solvent exposure, per-residue amino acid sequence similarity, per-residue bioinformatic descriptors, per-residue non-covalent bonding tendency per residue, phi / psi angle per residue, van der Waals half angle per residue diameter, secondary structure tendency per residue, amino acid adjacency per residue, amino acid adjacency per residue The method of embodiment I-48 or I-49, wherein the strain is derived from a strain of a plant.

[0201] Embodiment I-51. The properties of one or more candidate peptides are determined by computer simulation. The method of any one of embodiments I-48 to I-50, wherein the method is determined by

[0202] Embodiment I-52. The computer simulation comprises a molecular dynamics simulation; Monte Carlo simulation, coarse-grained simulation, Gaussian network model, The method of embodiment I-51, comprising machine learning, or any combination thereof.

[0203] Embodiment I-53. The properties of one or more candidate peptides are determined by experimental characterization. The method of any one of embodiments I-48 to I-52, wherein

[0204] Embodiment I-54. Amino acids satisfying constraints from one or more reference targets are identified as identical to the reference targets. 1-53, having 0% to 90% sequence identity. How to do it.

[0205] Embodiment I-55. Amino acids satisfying constraints from one or more reference targets are within 30 Å 2 ~3 000Å 2 and embodiments I-48 to I-10, having van der Waals surface area overlap with reference to 54. A method according to any one of claims 1 to 54.

[0206] Embodiment I-56. The combination includes constraints from at least two reference targets. The method of any one of aspects I-48 to I-55.

[0207] Embodiment I-57. The combination includes constraints derived from at least five reference targets. The method of any one of aspects I-48 to I-56.

[0208] Embodiment I-58. The combination of constraints includes one or more constraints that do not originate from the reference target , The method according to any one of embodiments I-48 to I-57.

[0209] Embodiment I-59. One or more non-reference target-derived constraints are used to achieve desired structural, kinetic, chemical, or functional properties. Embodiment I, which describes a biological or functional property, or any combination thereof. -58.

[0210] Embodiment I-60. Embodiment I-6, wherein the constraints are independently selected from the group consisting of: 48 to I-59. interatomic distance, Atomic fluctuations, atomic energy, chemical descriptors, solvent exposure, amino acid sequence similarity, bioinformatics descriptors, non-covalent tendency, Phi angle, Psi angle, van der Waals radius, secondary structure tendency, amino acid contiguity, and Amino acid contact.

[0211] Embodiment I-61. Embodiment I-48 wherein one or more constraints are, independently, atomic variations ~The method according to any one of I-60.

[0212] Embodiment I-62. Embodiment I-4 wherein one or more constraints are, independently, chemical descriptors 8 to I-61.

[0213] Embodiment I-63. Embodiment I-4 wherein one or more constraints are, independently, interatomic distances. 8 to I-62.

[0214] Embodiment I-64. The embodiment I-48 in which one or more constraints are, independently, secondary structure. 1. The method according to any one of claims 1 to 1-63.

[0215] Embodiment I-65. An embodiment in which one or more constraints are, independently, van der Waals surfaces The method according to any one of embodiments I-48 to I-64.

[0216] Embodiment I-66. One or more constraints independently affect a biological response or biological function. Related methods according to any one of embodiments I-48 to I-65.

[0217] Embodiment I-67. The engineered peptide is associated with a biological response or biological function. The method of any one of embodiments I-48 to I-66, comprising one or more atoms selected from the group consisting of methyl, ...

[0218] Embodiment I-68. The engineered peptide is associated with a biological response or biological function. The method according to any one of embodiments I-48 to I-66, comprising one or more amino acids

[0219] Embodiment I-69. The biological response or function is a function of gene expression, metabolic activity, Protein expression, cell proliferation, cell death, cytokine secretion, kinase activity, epigenetics Modification, cell death activity, inflammatory signaling, chemotaxis, tissue infiltration, immune cell lineage commitment, tissue microenvironment Environmental modification, immune synapse formation, IL-2 secretion, IL-10 secretion, growth factor secretion, interleukin-1 Feron gamma secretion, transforming growth factor beta secretion, and immunoreceptor tyrosine-based activation Motif activity, immunoreceptor tyrosine-based inhibitory motif activity, antibody-dependent cellular cytotoxicity, complement Body-dependent cytotoxicity, biological pathway agonist action, biological pathway antagonist action, Biological pathway redirection, kinase cascade modification, protein degradation pathway modification, proteostasis Sexual pathway modification, protein folding / pathway, post-translational modification pathway, metabolic pathway, gene transcription / translation , mRNA degradation pathway, gene methylation / acetylation pathway, histone modification pathway, epigenetic Tick pathway, immune-dependent clearance, opsonization, hormone signaling, and inte Glycine pathway, membrane protein signaling, ion channel flux, and g-proteins Any of embodiments I-66 to I-68, wherein the protein-coupled receptor response is selected from the group consisting of The method described in one.

[0220] Embodiment I-70. The reference target is one or more associated biological responses or biological functions containing atoms of One or more atoms in an engineered peptide that are associated with a biological response or biological function The atomic variation of one or more atoms in the reference target that is associated with a biological response or biological function. The method of embodiment I-66, wherein the atomic variation of

[0221] Embodiment I-71. The overlap is a root mean square dot product (RMSIP) greater than 0.25. The method of embodiment I-70.

[0222] Embodiment I-72. The overlap has a root mean square dot product (RMSIP) greater than 0.75. The method of embodiment I-71, comprising:

[0223] Embodiment I-73. In engineered peptides associated with a biological response or biological function at least a portion of the atoms of the reference target are topologically constrained to secondary structure elements in the reference target; The method according to any one of aspects I-67 to I-69.

[0224] Embodiment I-74. The method of embodiment I-73, wherein the secondary structure element is a beta-sheet. How to do it.

[0225] Embodiment I-75. The compound of embodiment I-73, wherein the secondary structure element is an alpha helix. The method described.

[0226] Embodiment I-76. The secondary structure element is a turn, the turn comprising 2 to 7 residues , the method of embodiment I-73, comprising at least one inter-residue hydrogen bond.

[0227] Embodiment I-77. The secondary structure element is a coil, and the coil contains 2 to 20 residues. The method of embodiment I-73.

[0228] Embodiment I-78. The method of embodiment I-73, wherein the coil does not contain inter-residue hydrogen bonds. method.

[0229] Embodiment I-79. In engineered peptides associated with a biological response or biological function At least a portion of the atoms of the and a combination of two or more secondary structure elements independently selected from the group consisting of: The method of any one of embodiments I-67 to I-69, subject to the restrictions.

[0230] Embodiment I-80. An embodiment wherein one or more spatially related topological constraints are interatomic distances. The method according to any one of embodiments I-48 to I-79.

[0231] Embodiment I-81. One or more spatially related topological constraints are atomic energies , The method of any one of embodiments I-48 to I-80.

[0232] Embodiment I-82. Each atomic energy is independently a pairwise attractive energy between two atoms. energy, pairwise repulsion energy between two atoms, atomic level solvation energy, Pairwise charge attraction energy between two atoms, pairwise hydrogen bond attraction between two atoms The method of embodiment I-81, wherein the bond is energy, or non-covalent energy.

[0233] Embodiment I-83. An embodiment in which one or more spatially related topological constraints are chemical descriptors. The method according to any one of embodiments I-48 to I-82.

[0234] Embodiment I-84. Each chemical descriptor independently comprises hydrophobicity, polarity, volume, net charge, lo gP, high performance liquid chromatography retention, or van der Waals radius. The method according to claim 1-83.

[0235] Embodiment I-85. One or more spatially related topological constraints are bioinformatic descriptors , The method of any one of embodiments I-48 to I-84.

[0236] Embodiment I-86. Each bioinformatic descriptor independently includes BLOSUM similarity, pKa, z Scale, Cruciani trait, Chidera factor, VHSE scale, ProtFP, MS- WHIM score, T scale, ST scale, transmembrane tendency, protein-buried area, Helix propensity, sheet propensity, coil propensity, turn propensity, immunogenic propensity, antibody epitope The method of embodiment I-85, wherein the protein is generated or a protein interface is generated.

[0237] Embodiment I-87. An embodiment in which one or more spatially related topological constraints is solvent exposure. The method of any one of aspects I-48 to I-86.

[0238] Embodiment I-88. At least one of the constraints derived from the one or more reference targets is a GPC The method of any one of embodiments I-48 to I-87, wherein the R extracellular domain.

[0239] Embodiment I-89. At least one of the constraints from the one or more reference targets is an ion The method of any one of embodiments I-48 to I-88, which is a channel extracellular domain. Law.

[0240] Embodiment I-90. At least one of the constraints from the one or more reference targets is tamper-evident. Protein-protein or protein-peptide interfacial bonding, embodiments I-48 to I-49 -89.

[0241] Embodiment I-91. At least one of the constraints from one or more reference targets is The method of any one of embodiments I-48 to I-90, wherein the polymorphic region is derived from the polymorphic region.

[0242] Embodiment I-92. The engineered peptide is associated with a biological response or biological function. each of the one or more atoms independently selected from carbon, oxygen, nitrogen, hydrogen, , sulfur, phosphorus, sodium, potassium, zinc, manganese, magnesium, copper, iron, molyb Any of embodiments I-48 to I-91, wherein the metal is selected from the group consisting of: arsenic, arsenic, and nickel. or one of the methods described above.

[0243] Embodiment I-93. The engineered peptide is associated with a biological function or biological response. Each of the one or more amino acids independently contributes to the production of a protein. Natural amino acids that form proteins, natural amino acids that do not form proteins, or chemically synthesized non-natural amino acids The method of any one of embodiments I-48 to I-92, wherein the amino acid is an amino acid.

[0244] Embodiment I-94. The engineered peptide exhibits at least one structural alteration compared to a reference target. The method of any one of embodiments I-48 to I-93, having a differential.

[0245] Embodiment I-95. The at least one structural difference is, independently, in sequence, number of amino acid residues, Total number of atoms, total hydrophilicity, total hydrophobicity, total positive charge, total negative charge, one or more secondary structures, shape factor, Zernike descriptors, van der Waals surfaces, structural graph nodes and edges, volumetric surfaces, Electrostatic potential surface, hydrophobic potential surface, local diameter, local surface features, skeleton model, charge density, hydrophilicity a surface to volume ratio, an amphiphilic density, and a surface roughness. The method according to embodiment I-94

[0246] Embodiment I-96. One or more secondary structural differences are engineered relative to a reference target. the presence of one or more additional secondary structure elements in the peptide, each additional secondary structure element being Independently, they consist of alpha helices, beta-sheets, loops, turns, and coils. The method of embodiment I-95, wherein the patient is selected from the group consisting of:

[0247] Embodiment I-97. Between 10% and 90% of the amino acids of the engineered peptide are one or more non- any one of embodiments I-48 to I-96, satisfying topological constraints from a reference target. How to do it.

[0248] Embodiment I-98. One or more non-reference target-derived topological constraints are used to define a pre-specified function. The method of embodiment I-97, wherein the method is enhanced.

[0249] Embodiment I-99. Non-reference-derived topological constraints enforce or suppress secondary structure elements in the reference-derived fraction of peptides. Stabilize it or Whether non-reference-derived topological constraints reinforce atomic variation in the reference-derived fraction of peptides; Non-reference-derived topological constraints modify the overall peptide hydrophobicity, Non-reference-derived topological constraints may alter peptide solubility; Non-reference-derived topological constraints modify the peptide net charge or Non-reference-derived topological constraints can be detected in labeled or label-free assays. or Whether non-reference-derived topological constraints enable detection in in vitro assays; Whether non-reference-derived topological constraints enable detection in in vivo assays; Whether non-reference-derived topological constraints enable capture from complex mixtures; Whether non-reference-derived topological constraints allow enzymatic processing Whether non-reference-derived topological constraints enable cell membrane permeability; Non-reference-derived topological constraints allow binding to secondary targets, or Whether non-reference-derived topological constraints alter immunogenicity or or any combination thereof.

[0250] Embodiment I-100. A composition comprising two or more selection-derived polypeptides, wherein each polypeptide The peptides may be positive selection molecules that independently contain one or more positive induction properties, or one or more a negative selection molecule comprising negative induction properties, each type of property independently: amino acid sequence, polypeptide secondary structure, molecular dynamics, Chemical characteristics, biological function, immunogenicity, Multispecificity of reference target(s), Cross-species reference target reactivity, selectivity of the desired reference target(s) over the undesired reference target(s); selectivity of the reference target(s) within a sequence and / or structurally homologous family; selectivity of reference target(s) with similar protein function; A larger family of undesired targets with high sequence and / or structural homology selectivity of distinct desired reference target(s) from the Selectivity for distinct reference target alleles or mutations; Selectivity for chemical modification at the level of distinct reference target residues; cell type selectivity, Selectivity for tissue type, selectivity to the organizational environment; tolerance to structural diversity of the reference target(s); Tolerance to sequence variability of the reference target(s), and tolerance to kinetic diversity of the reference target(s); At least one of the two or more polypeptides is selected from the group consisting of the two or more polypeptides ... The composition is a peptide.

[0251] Embodiment I-101. At least one of the two or more polypeptides is a positive selection molecule. and at least one of the two or more polypeptides is a negative selection molecule. The composition according to embodiment I-100.

[0252] Embodiment I-102. At least one of the two or more polypeptides is a naturally occurring protein. The composition of embodiment I-100 or I-101, wherein the composition is a protein.

[0253] Embodiment I-103. Corresponding positive selectivity components including at least one shared characteristic type. at least one pair of a positive selection molecule and a negative selection molecule, wherein the positive selection molecule comprises a positive characteristic; The method of any one of embodiments I-100 to I-102, wherein the negative selection molecule comprises a negative characteristic. The composition described above.

[0254] Embodiment I-104. A library of binding molecules is prepared using the composition described in embodiment I-100. A method for screening a pool of candidate binding molecules, comprising: passing the pool of candidate binding molecules through at least one selection loop; each selection round comprising: A negative selection step is performed to screen at least a portion of the pool against the negative selection molecule. Tep and A positive selection step is performed to screen at least a portion of the pool in favor of a positive selection molecule. and The order of selection steps within each round, and the order of rounds, may differ from the alternative order. The method results in the selection of a subset of the pool.

[0255] Embodiment I-105. The embodiment in which the library of binding molecules is a phage library The method described in I-104.

[0256] Embodiment I-106. The library of binding molecules is a cell library. 105. The method according to claim 105.

[0257] Embodiment I-107. Embodiment I, wherein the library of binding molecules is a B-cell library -106.

[0258] Embodiment I-108. Embodiment I, wherein the library of binding molecules is a T cell library -106.

[0259] Embodiment I-109. Embodiments I-104 to I-10, which include two or more selection rounds. 8. A method according to any one of claims 1 to 8.

[0260] Embodiment I-110. Embodiments I-104 to I-10, comprising three or more selection rounds. 9. The method according to any one of claims 1 to 9.

[0261] Embodiment I-111. The embodiment I-1, wherein each round comprises a different set of selected molecules 09 or I-110.

[0262] Embodiment I-112. At least two rounds use the same negative selection molecule or the same positive selection molecule. The method of embodiment I-109 or I-110, comprising a selective molecule, a nucleotide sequence, or both.

[0263] Embodiment I-113. Proportions resulting from one selection round are used before proceeding to the next selection round. Any one of embodiments I-109 to I-112, comprising analyzing a subset of rules. The method described in the first paragraph.

[0264] Embodiment I-114. Subset pool analysis is used in one or more subsequent selection rounds determining a set of positive and / or negative selection molecules to be used in the assay, How to do it.

[0265] Embodiment I-115. Each subset pool analysis independently determines a peptide / protein balance. Biosensor binding, peptide / protein ELISA, peptide library binding, cell extraction Substrate binding, cell surface binding, cell activity assay, cell proliferation assay, cell death assay, enzyme Activity assays, gene expression profiles, protein modification assays, Western blots and immunohistochemistry, The method described below.

[0266] Embodiment I-116. Positive, negative, or positive and negative markers used in one or more subsequent selection rounds Both positive and negative selected molecules are analyzed by statistical / informatics scoring or machine learning for subset pool analysis. According to any one of embodiments I-113 to I-115, determined by training. method.

[0267] Embodiment I-117. A subset pool obtained from one selection round is used in a next selection round. The method according to any one of embodiments I-109 to I-116, wherein the method is modified before moving to the next round. How to post.

[0268] Embodiment I-118. Subset pool analysis is used in one or more subsequent selection rounds Determine which molecules will be positively, negatively, or both positively and negatively selected and move on to the next selection round. The method of embodiment I-117, wherein the subset pool is modified before

[0269] Embodiment I-119. Each modification independently represents a gene mutation, a gene depletion, a gene enrichment, a In embodiments I-117 or I-118, the compound is selected from the group selected from the group consisting of chemical modifications, and enzymatic modifications. -118. [Example]

[0270] The following examples are merely illustrative and are not intended to limit in any way any aspect of the present disclosure. It is not intended to be limiting.

[0271] Example 1: Selection of engineered peptides using a VEGF epitope as a reference target As shown in Figures 6A and 7A, putative therapeutic epitopes of VEGF were identified using engineered Identified as a reference target for peptide selection, and determined interatomic distances and amino acid descriptor phases. (Figure 6B). The interatomic distances and amino acid descriptor phases of the reference target were calculated by dynamics simulation. The covariance matrix of atomic variations was generated for the epitopes in the reference target. The different engineered peptide candidates are then subjected to computational protein design (e.g., We generate a model using Zetta and perform dynamics simulations on the candidates to determine the interatomic distances. The phases of the amino acid descriptors were determined (Figures 6C-6E). The percent difference (MPE) was compared (Figure 6G-6H). MPE values are calculated as the reference phase versus candidate 1 phase: 6. 03%, Reference Phase vs. Candidate 2 Phase: 6.00%, and Reference Phase vs. Candidate 3 Phase: 22.8% It was.

[0272] For evaluation of one candidate engineered peptide, additional constraints on the combination, atomic variations, The high-dimensional topological similarity between this candidate and the VEGF-derived reference target was added (Figures 6G-6H). When comparing gender, the MPE was 36.6%.

[0273] Example 2: Selection of engineered peptides using a VEGF epitope as a reference target Using the same reference target identified in Example 1 above, a second sequence of engineered peptides was prepared. A set of candidate engineered peptides was created using computational protein design ( Generated using, for example, Rosetta or other peptide space sampling methods, Dynamics simulations were performed on the candidates. The covariance matrix of the atomic fluctuations was calculated using the reference target energy. for the epitope and for residues in the candidate that correspond to residues in the epitope of the reference target This was generated.

[0274] Each covariance matrix, i.e., one covariance matrix for the reference target and one for each of the candidates, To compute the eigenvectors and eigenvalues for a single covariance, Perform a principal component analysis and retain only the eigenvectors with the largest eigenvalues (Figure 8). The vectors are the first, second, third, and fourth observed vectors in the set of simulated molecular structures. If the candidate behaves like the reference epitope, The eigenvectors of the target will be similar to the eigenvectors of the reference target (epitope). The similarity of vectors is determined by their components (each CA atom) being aligned and pointing in the same direction. The eigenvector between the candidate and the reference target corresponds to the eigenvector (3D vector centered at the target) (Fig. 7D-7G). This similarity between the vectors was computed using the dot product of the two eigenvectors. The product value is 0 if the two eigenvectors are at 90 degrees to each other, or If the vectors point in exactly the same direction, it is 1.

[0275] The ordering of eigenvectors is based on their eigenvalues, and the eigenvalues are The probability that the simulation samples the underlying energy landscape of those different molecules Due to their chemical nature, they may not necessarily be identical between two different molecules, so multiple individual The dot product between separately ranked eigenvectors was required (e.g., the eigenvectors of the reference target). The candidate eigenvectors are multiplied by vectors 2, 3, 4, etc.1). However, molecular motion is complex and involves more than two (or several) It may involve the primary / important mode of movement.

[0276] To solve these two problems, we first consider all pairs of eigenvectors in the candidate and reference targets. This results in a matrix of dot products, the dimensions of which are analyzed. is determined by the number of eigenvectors used, and for 10 eigenvectors, the dot product matrix is , 10 × 10. This matrix of dot products can be used to compute the root mean square value of the dot products. This is the root mean square dot product (RMSIP). .

[0277] Principal component analysis (PCA) calculates a 3L × 3L dimensional coordinate covariance matrix (where L is the number of atoms) as follows: Convert to eigenvectors, Φ (reference target), and Ψ (MEM), and eigenvalues Λ. The set Φ is a set of N eigenvectors φ for the reference target. i and the set Ψ is and N eigenvectors ψ j and the eigenvectors are represented by their associated eigenvalues. Within each of these sets, the eigenvector with the largest eigenvalue is the eigenvector with the largest eigenvalue. It accounts for the largest part of the covariates. To compare the movement similarity between the reference target and the MEM, , each φ i and ψ j Compute the dot product of the eigenvectors. i and ψ j Unique base The root mean square of all dot product combinations of vectors is the RMSIP relative to the reference target (RMSIP). The overall similarity of the motion of the candidate engineered peptide (MEM) is provided (Figure 8).

[0278] RMSIP results from five candidate engineered peptides versus the VEGF reference epitope were These data are shown in Table 1. Sampling from full simulation of 1000 candidates generated using Zetta design Among the 1000 candidates, XTR-1000-T0 achieved the lowest Rosetta (static structural) energy (lower is better), but had intermediate RMSIP kinetic agreement. Candidates XTR-1000-B1 and B2 had the highest kinetic match scores (e.g., For example, their motions are best compared with the motion of a reference target computed by RMSIP. (These candidates were also closely matched.) Candidates XTR-1000-W1 and W2 were selected from these 1000 candidates. The RMSIP dynamic range in the dataset was 0.772 to 0.545. The VEGF reference epitope was shown to be consistent with the VEGF-specific phenotype, and had the lowest kinetic match score. The aligned candidate structures are shown in Figure 7B. [Table 1]

[0279] Example 3: Phage synthesis using engineered peptides for putative VEGF epitopes Programmed in vitro selection The three engineered peptides described in Example 1 and additional peptides made following similar procedures were An additional fourth engineered peptide was used in a series of phage panning procedures. The peptides are shown in Figure 9. Two of the peptides were positively selected by the positive selection molecules (uMEM and s MEM), and two were negative selection molecules (iMEM2 and iMEM1). EM peptides were highly phase-referenced, whereas uMEM peptides were less phase-referenced. The two iMEM peptides are in phase-reference agreement with zero, which is indicative of no other binding interaction than the desired one. Therefore, to select against binding partners that bind to sMEM or uMEM, Included as an inverted version of EM and uMEM. Biosensor assays were used. Analysis of the biotin-conjugated peptide confirmed binding to bevacizumab, which was significantly higher than binding to the reference target. The predicted phases were based on the similarity of the candidate phases.

[0280] Octet / biosensor screening: Affinity of different engineered peptides A single-cycle kinetic assay design was used to evaluate the activity on an Octet Red 384 instrument. The peptides were evaluated separately and linked to the streptavidin biosensor via a biotin linker. Immobilized on streptavidin-coated chip. The analyte was washed onto the sensor tip and analyzed into peptides. The binding of the molecule in the sample was recorded. For this assay, the analyte was 0.19 μM to 1. Each assay was performed in duplicate. Controls were also , buffer (to control for sensor drift) alone, and purified Ig from human ND serum. A separate control of G (to control for nonspecific IgG binding) was run.

[0281] Seven different panning programs were designed, each containing three rounds. Each program includes a positive selection step and a negative selection step (Figure 11). At least one engineered peptide was used as a target. VEGF, and BSA as a negative target to select against nonspecific binding) Conventional selection was also included. 738 clones were selected for ELISA after three panning rounds. were selected for analysis.

[0282] The panning protocol starts with a human naive scFv library and pans in solution. The selected molecules were bound to biotin (but still in solution). For the nucleotide sequence, the starting pool is first combined in solution with the negative selection molecule and then streptavidin. A putavidin-coated substrate (e.g., magnetic beads) is applied to the mixture to allow for negative selection. Therefore, any phage in the pool that bound to the negative selection molecule also The remaining solution was removed and the flow-through fraction was The flow-through fraction was combined with the positive selection molecule, allowed to bind, and then In this step, a streptavidin-coated solid substrate was applied to the mixture. The remaining non-binding phages were removed while the binding phages were retained. E. coli was then transfected with the eluted phages using a 30 minute incubation. Transfect the transfected cells into a 500 kDa genomic DNA matrix for next-generation sequencing and analysis. Divide the phage for DNA isolation and then use in subsequent panning rounds For each panning program, negative selection was performed first, followed by positive selection in each round. The second choice was made.

[0283] The candidates obtained from each of the seven panning programs and the conventional panning method were then The pool was analyzed for response to VEGF and sMEM positive selection molecules (corrected iMEM). Binding to full-length VEGF and the putative epitope sMEM was analyzed using ELISA. The analysis of these ELISA tests is shown in Figures 12A-12B and 13A-13H. These results are shown in Figure 1. In vitro selection programs using engineered peptides However, this did not reduce the propensity to bind full-length VEGF and did not increase the putative epitopes in the clones tested. The candidate pool was divided into two groups: bevacizumab:V Blockade of EGF binding was also tested in a cross-blocking ELISA assay (0 nM, 67p (dose-response competition with bevacizumab at 6.7 nM, 670 pM, and 6.7 nM). These results The confirmed cross-blocking results obtained from each program are shown in Figures 14A-14I and Table 2. The total number of fragment clones is summarized in Figure 15. These were fragments obtained using engineered peptides. A programmable in vitro selection program was used to obtain engineered peptides. A complete clone library that cross-blocks bevacizumab and shares the same reference target epitope This demonstrates that we were able to isolate clones from the [Table 2]

[0284] Clones that exhibited cross-blocking behavior were sequenced via Sanger sequencing and It was found that 10 distinct clones were identified. The results from the in vitro selection are shown in Table 3A. Those obtained through conventional selection with A are listed in Table 3B. All Fabs generated for testing were analyzed for binding, cross-blocking, CDR sequences, and Figures 17 and 18 summarize the genetic and germline usage of the genes listed in Tables 3A and 3B. ELISA binding results for Fab are shown. These were obtained using a programmable in vitro selection The selection of antibodies exploits antibody CDR loop diversity and Ig germline utilization, unlike conventional panning. Demonstrate that the induction was performed in a manner that [Table 3] [Table 4] [Table 5]

[0285] The selection pool was scored using the following equation: Blockage trend = SUM(X Blockage gradient, (sMEM + VEGF) - iMEM), where X Blockage gradient Diagram, sMEM, and VEGF are robust Z-scores.

[0286] Scoring rationale: Through significant (by robust z-score) negative gradients, screening is performed. If a blocking response is observed, the blocking tendency is determined by the z-score for VEGF binding and the x-blocking slope. The blocking trends are summarized in Figure 19 and the table below. [Table 6]

[0287] Compared to a conventional program (which only used VEGF and BSA as selection molecules) Using uniform and random sampling of all in vitro selection programs, different The selection program is evaluated for cross-blocking enrichment compared to a control (conventional) program; At least four programs using engineered peptides have demonstrated the enrichment potential summarized in Figure 20. The statistical test for cross-blocking enrichment was the Kruskal-Wallis test as follows: It was. 1. A random, uniform sample of 96 clones from all panning programs , measuring cross-blocking activity 2. Rank cross-blocking across all 96 clones 3. Run a Kruskal-Wallis test to compare the cross-blocked runs per program against the control. Calculate the average of 4. x Intercept Enrichment = 100% * (Program Intercept Mean Rank - Control Mean Rank) / ( Control average rank)

[0288] Clones were also subjected to next-generation sequencing (NGS) to identify CDR loops at the genome level. Figure 21 provides a schematic overview of NGS sample preparation. Specifically, individual heavy and light chain sequences are cloned in the constant portion of an expression vector. Samples were prepared by 2 x 250 paired-end sequencing runs. The reads were combined and annotated using tools such as PyIg.

[0289] The sequences were analyzed to determine whether two unique sequences were identified by a sequencing error called "clonality." As shown in Figure 22, the normalized sequence A summary of clonality for each round of each program is shown in Figure 23. can be.

[0290] Classical panning approaches using only full-length protein (VEGF) limit diversity (Program 12), but the engineered peptides are programmed to The genomic approach focuses on repertoire diversity at least twice as efficiently. 24A-24L are Round 1 (Figures 24A-24D) and Round 2 (Figures 24E-24H) Different screening rounds for round 1, 2, and round 3 (Figures 24I-24L) resulted in pairings, analyzing how they shape the diversity of the resulting selected pool. Frequency comparison and dimension chart.

[0291] In vitro selection of engineered peptides (MEMs) was performed as a first selection loop. In the United States, a different approach with higher germline diversity compared to traditional approaches was Using sMEM-based in vitro selection, complete antibody clonotypes are isolated. More diverse light chain germline usage in round 1 compared to full-length antigen and uMEM The MEM-based in vitro selection program provides This results in different heavy chain germline usage in round 2. The order and identity of the MEMs used in the MHC dictates heavy chain germline usage. EM-based in vitro selection program showed significantly higher chromatin expression in round 2 compared to full-length antigens MEMs used in in vitro selection programs resulting in different light chain germline usage. The order and identity of the nucleotides influences light chain germline usage. The selection program was different and more diverse in round 3 compared to full-length antigens. This results in optimal heavy chain germline usage. The order and identity influence heavy chain germline usage and diversity. The in vitro selection program differs in round 3 compared to the full-length antigen, and This results in more diverse light chain germline usage. The order and identity of the EM influences light chain germline usage and diversity.

[0292] Key points on how different phage panning programs focused on Fab hits Approximately, the following is provided in Figures 25A and 25B.

[0293] The (sMEM) on epitopes per panning round for each program shown in Figure 26 ) Graph summarizing VEGF hit frequencies is shown in Figure 1 for an in vitro selection protocol of engineered peptides. Col is a specific mAb hit that has been identified to bind to VEGF and block bevacizumab. Many of these hits were not identified using traditional approaches. Figure 27 shows the number of extra-epitope VEGF hits per panning round for each program. The frequency of VEGF binding was summarized, and the conventional program identified putative VEGF-binding proteins. Figure 28 shows that mAb hits were identified that were not pitope-selective mAb hits. Summarize the join.

[0294] Example 4: Phospholipase C19 (PAC) using engineered peptides for PD-L1 therapeutic epitopes Programmed in vitro selection of dipeptides The therapeutic epitope reference target sites identified on PD-L1 were used to develop a series of engineered The peptide (MEM) was synthesized as described in Example 2, as summarized in Figures 29-31D. These three engineered peptides, sM EM, nMEM (both positive selection molecules), and iMEM (negative selection molecules with opposite properties) The ability of the biosensor to detect and treat two anti-PD-L1 antibodies, avelumab and dextromethorphan, was investigated. Binding to urvalumab was assessed (both antibodies were shown to bind to the reference target epitope). The data are shown in Figures 31A-32C. A series of five different panning programs were used, as shown in Figure 33, for conventional selection. A control program using the molecules PD-L1 and BSA was designed and presented on phage. For screening the indicated naive human Ig scFv format library A similar panning protocol was used as described above in Example 3. For the matching program, negative selection was performed first and positive selection was performed second in each round. .

[0295] The results of the experiments performed on PD-L1 and the different engineered peptides were selected using each program. The ELISA responses of the resulting pools are summarized in Figures 34-38 and are shown for different protocols. The complete ELISA responses comparing the gram are provided in Figures 39A-39U. The resulting pool was divided into groups with different combinations of desired binding behaviors, as summarized in Figure 40. Different selection filter criteria were used for analysis. ELISA further subjected to cross-blocking assay A summary of the different clones selected from the results is provided in Tables 2A and 2B below. [Table 7] [Table 8]

[0296] These ELISA hits were assayed at 0 nM, 67 pM, 670 pM, and 6.7 nM. A dose-response PD-L1 competition analysis with avelumab or durvalumab in 3 Four putative cross-blocking clonal hits were identified. Blocking propensity was calculated as follows: EL ISA Z score (sMEM1 + sMEM5 + PD-L1 - iMEM) + MAX (Abel A summary of the results is provided in Table 3 below. can be. [Table 9]

[0297] ELISA responses are provided in Figures 42A-42F. Three independent clones were sequenced (via Sanger sequencing) and are listed in Figure 43 A summary of the number of distinct clones with cross-blocking hits across the panning program is shown in Figure 44. will be provided to.

[0298] These results were analyzed to determine whether any of the in vitro selection programs were associated with PD-L1:Abeta Determine whether random selection results in enrichment of clones that cross-block lumab / durvalumab Compared with the previous program (which only used PD-L1 and BSA as selection molecules), In comparison, clones from a uniform random sampling of all in vitro selection programs Based on ELISA and cross-blocking data using the ELISA, engineered peptides were used. At least two of the programs tested showed enrichment. The results and summary of clones are available. , as shown in Figures 45A-46 (the shaded input in Figure 45C is from conventional panning). The following rationale was used in the analysis:Scoring rationale:Significant (donkey If a blocking response is observed through a negative slope (by PD z score), the blocking tendency is -L1, MEM binding, and X-interception slope z-score combinations are used. The resulting X-blocking z-scores indicate that these Tx mAbs have slightly different epitopes on their surface. Therefore, the maximum z-score for avelumab versus durvalumab.

[0299] Example 5: Machine learning model for the selection of engineered peptides Use the reference target to identify topological characteristics of the reference target (sequence) and create a scaffold. The scaffold blueprint is then coded into the The sequence of amino acids in the resulting polypeptide is constrained to match the order of amino acids in the reference target. Sequence homology can be constrained to 100% (each amino acid in the reference target is (corresponding to one amino acid in the print) or sequence homology is, for example, 10 to 90 The scaffold blueprint can be a base sequence of the original scaffold. The spatially related topological properties are then converted into a vector representation (Figure 61, left) derived from the reference target. Combining and overlapping spatially related topological constraints to generate engineered peptides Each scaffold group may be used to generate candidate polypeptides. Lints are assigned labels based on the scoring of overlaps (Figure 61, right).

[0300] Machine learning (ML) models generate scaffold blueprints and corresponding scores. It can be trained with training data that includes expressions, e.g., one-dimensional vectors of numbers, alphanumeric characters, It can be a two-dimensional matrix of data, or a three-dimensional tensor of normalized values. In some cases, the representation is an ordered list of the number of intervening scaffold residue positions Since the target-residue order can be inferred from the target structure, such A representation may be used, thus specifying the amino acid identity of the target residue position. Scaffold Blueprint scores are calculated based on the number of scaffold blueprints that are created. Computational protein modeling (e.g., The scores can then be calculated using a computerized time series. It can be calculated based on the energy terms generated by protein modeling.

[0301] ML models include, for example, boosted decision tree algorithms, ensembles of decision trees, Extreme Gradient Boosting (XGBoost) models, Random Forests, Support Vector Machines After training, the ML model can then be , which is run to generate a set of prediction scores from a set of scaffold blueprints. If the predicted score is higher than the desired score, the scaffold corresponding to the predicted score is The blueprint is simulated by computational protein modeling. The ground truth score can be generated by The accuracy and prediction scores can be compared to determine whether to retrain the ML model. In the implantation of the training and execution steps, we combine the optimal / improved skills with the desired scores. 62 until the workload blueprint is predicted. The optimal / improved scaffold blueprint is then transformed into an engineered peptide. Exchange.

Claims

1. 1. A method for selecting engineered peptides, comprising: Identifying one or more topological characteristics of the reference target; designing spatially related constraints for each topological feature to generate a combination of spatially related topological constraints derived from the reference target; comparing the spatially related topological properties of the candidate peptide with the set of spatially related topological constraints derived from the reference target; and selecting candidate peptides having spatially related topological properties that overlap with the combination of spatially related topological constraints derived from the reference target to generate the engineered peptide.

2. the overlap between each feature is 75% or less mean percent error (MPE) as determined independently by one or more of Total Topological Constraint Distance (TCD), Topological Clustering Coefficient (TCC), Euclidean distance, Power distance, Soergel distance, Canberra distance, Sorensen distance, Jaccard distance, Mahalanobis distance, Hamming distance, Quantitative Estimate of Likeness (QEL), or Chain Topology Parameter (CTP); one or more constraints are derived from per-residue energy, per-residue interaction, per-residue variation, per-residue interatomic distance, per-residue chemical descriptor, per-residue solvent exposure, per-residue amino acid sequence similarity, per-residue bioinformatic descriptor, per-residue non-covalent bonding propensity, per-residue phi / psi angle, per-residue van der Waals radius, per-residue secondary structure propensity, per-residue amino acid adjacency, per-residue amino acid contact; or the properties of one or more candidate peptides are determined by computer simulation, wherein the computer simulation comprises molecular dynamics simulation, Monte Carlo simulation, coarse-grained simulation, Gaussian network model, machine learning, or any combination thereof; or The properties of one or more candidate peptides are determined by experimental characterization; The method of claim 1.

3. One or more constraints Independently, are atomic fluctuations; Independently, is it a chemical descriptor? Independently, are the interatomic distances; Independently, is it a secondary structure? independently, a van der Waals surface; or Independently, associated with a biological response or function; At least one of The method according to claim 1 or 2.

4. the engineered peptide is (A) one or more atoms associated with a biological response or biological function; or (B) one or more amino acids associated with a biological response or function; at least one of The method according to any one of claims 1 to 3, The biological response or biological function may be a signaling pathway, such as gene expression, metabolic activity, protein expression, cell proliferation, cell death, cytokine secretion, kinase activity, epigenetic modification, apoptotic activity, inflammatory signaling, chemotaxis, tissue infiltration, immune cell lineage commitment, tissue microenvironment modification, immune synapse formation, IL-2 secretion, IL-10 secretion, growth factor secretion, interferon gamma secretion, transforming growth factor beta secretion, immunoreceptor tyrosine-based activation motif activity, immunoreceptor tyrosine-based inhibitory motif activity, antibody-dependent cellular cytotoxicity, complement-dependent cytotoxicity, or a biological pathway. pathway agonism, biological pathway antagonism, biological pathway redirection, kinase cascade modification, protein degradation pathway modification, protein homeostasis pathway modification, protein folding / pathway, post-translational modification pathway, metabolic pathway, gene transcription / translation, mRNA degradation pathway, gene methylation / acetylation pathway, histone modification pathway, epigenetic pathway, immune-dependent clearance, opsonization, hormone signaling, integrin pathway, membrane protein signaling, ion channel flux, and g-protein coupled receptor response; the reference target comprises one or more atoms associated with a biological response or function; the atomic variation of the one or more atoms in the engineered peptide associated with a biological response or biological function overlaps with the atomic variation of the one or more atoms in the reference target associated with a biological response or biological function; or the overlap is a root mean square dot product (RMSIP) greater than 0.25 or a RMSIP greater than 0.75; At least a portion of the atoms in the engineered peptide associated with a biological response or function are topologically constrained to secondary structure elements in the reference target; or the secondary structure element is (a) beta-sheet; (b) alpha helix; (c) a turn containing 2 to 7 residues and containing at least one inter-residue hydrogen bond; (d) coils containing no inter-residue hydrogen bonds; and at least a portion of the atoms in the engineered peptide that are associated with a biological response or function are topologically constrained into a combination of two or more secondary structure elements independently selected from the group consisting of a beta-sheet, an alpha helix, a turn, and a coil; at least one structural difference in the one or more secondary structures is the presence of one or more additional secondary structure elements in the engineered peptide compared to the reference target, each additional secondary structure element independently selected from the group consisting of an alpha helix, a beta-sheet, a loop, a turn, and a coil; The topological constraints from the one or more non-reference targets enforce a pre-specified function; or The method, wherein the method is at least one of the following: whether non-reference-derived topological constraints reinforce or stabilize secondary structure elements in the reference-derived fraction of the peptide; whether non-reference-derived topological constraints reinforce atomic variation in the reference-derived fraction of the peptide; whether non-reference-derived topological constraints alter the overall peptide hydrophobicity; whether non-reference-derived topological constraints alter peptide solubility; whether non-reference-derived topological constraints alter the peptide total charge; whether non-reference-derived topological constraints allow for detection in labeled or label-free assays; whether non-reference-derived topological constraints allow detection in in vitro assays; whether non-reference-derived topological constraints allow for detection in in vivo assays; whether non-reference-derived topological constraints allow capture from complex mixtures; Do non-reference-derived topological constraints allow for enzymatic processing? Do non-reference-derived topological constraints enable cell membrane permeability? non-reference-derived topological constraints allow binding to secondary targets; or Non-reference-derived topological constraints alter immunogenicity;

5. whether one or more spatially related topological constraints are solvent exposed; at least one of the constraints from the one or more reference targets is a GPCR extracellular domain; at least one of the constraints from the one or more reference targets is an ion channel extracellular domain; at least one of the constraints derived from the one or more reference targets is a protein-protein or protein-peptide interface junction; or at least one of the constraints derived from the one or more reference targets is derived from a polymorphic region of the target; The method according to any one of claims 1 to 4, wherein the method is at least one of the following:

6. the engineered peptide is (A) one or more atoms associated with a biological response or function, wherein each of said one or more atoms is independently selected from the group consisting of carbon, oxygen, nitrogen, hydrogen, sulfur, phosphorus, sodium, potassium, zinc, manganese, magnesium, copper, iron, molybdenum, and nickel; or (B) one or more amino acids associated with a biological function or response, wherein each of said one or more amino acids is independently a naturally occurring proteinogenic amino acid, a naturally occurring non-proteinogenic amino acid, or a chemically synthesized non-natural amino acid; at least one of The method according to any one of claims 1 to 5, the engineered peptide has at least one structural difference compared to the reference target; the at least one structural difference is independently selected from the group consisting of sequence, number of amino acid residues, total number of atoms, total hydrophilicity, total hydrophobicity, total positive charge, total negative charge, one or more secondary structures, shape factors, Zernike descriptors, van der Waals surface, structure graph nodes and edges, volume surface, electrostatic potential surface, hydrophobic potential surface, local diameter, local surface features, backbone model, charge density, hydrophilic density, surface to volume ratio, amphiphilic density, and surface roughness; the amino acids that satisfy the constraints from the one or more reference targets have between 10% and 90% sequence identity with the reference targets; The amino acids that satisfy the constraints from the one or more reference targets are within 30 Å ２ ~3000 Å ２ having a van der Waals surface area overlap with said reference of the combination includes constraints from at least two reference targets; the combination includes constraints from at least five reference targets; the combination of constraints includes one or more constraints that do not originate from the reference target; the one or more non-reference target-derived constraints describe desired structural, kinetic, chemical, or functional properties, or any combination thereof; The constraints may be independently selected from the group consisting of: interatomic distance, Atomic fluctuations, atomic energy, chemical descriptors, solvent exposure, amino acid sequence similarity, bioinformatics descriptors, non-covalent tendency, Phi angle, Psi angle, van der Waals radius, secondary structure tendency, amino acid contiguity, and selected from amino acid contacts; topological constraints from the one or more non-reference targets enforce a pre-specified function; or whether non-reference-derived topological constraints reinforce or stabilize secondary structure elements in the reference-derived fraction of the peptide; whether non-reference-derived topological constraints reinforce atomic variation in the reference-derived fraction of the peptide; whether non-reference-derived topological constraints alter the overall peptide hydrophobicity; whether non-reference-derived topological constraints alter peptide solubility; whether non-reference-derived topological constraints alter the peptide total charge; whether non-reference-derived topological constraints allow for detection in labeled or label-free assays; whether non-reference-derived topological constraints allow detection in in vitro assays; whether non-reference-derived topological constraints allow for detection in in vivo assays; whether non-reference-derived topological constraints allow capture from complex mixtures; Do non-reference-derived topological constraints allow for enzymatic processing? Do non-reference-derived topological constraints enable cell membrane permeability? non-reference-derived topological constraints allow binding to secondary targets; or whether non-reference-derived topological constraints alter immunogenicity; or any combination thereof; The method.

Images