Methods for selecting binding reagents

Abstract

Provided herein are methods and systems for selecting affinity reagents that bind to desired peptide epitopes in a variety of sequence contexts. The method relies on obtaining a peptide library and using the peptide library to select affinity reagents, wherein each peptide has a sequence αXβ, wherein X is a desired peptide epitope, wherein α and β each comprise amino acids.

Description

[0001] Cross-references

[0002] This application claims the benefit of U.S. Provisional Application No. 62 / 547,699, filed on August 18, 2017, which is incorporated herein by reference. Background Art

[0003] Selection methods for generating binding agents are generally designed to select binding agents with high affinity and specificity for a single epitope or protein. For certain applications, it may be useful to select binding agents that bind multiple epitopes or to characterize the binding mode of binding agents that are not specific for a single protein or epitope. Summary of the invention

[0004] The present disclosure provides methods and systems for selecting and characterizing affinity reagents. In some embodiments, the present disclosure provides methods for selecting affinity reagents to bind to epitopes in a variety of sequence environments. The methods and systems described herein can also be used to characterize the binding mode of affinity reagents, and the influence of the sequence environment on the binding of affinity reagents to epitopes. In addition, the methods and systems described herein can be used to characterize and select affinity reagents that bind across multiple epitopes, such as binding across multiple epitopes with the same length. By demonstrating promiscuity on multiple epitopes of the same length, a specific affinity reagent can be used to identify a sequence as comprising at least one epitope in a plurality of epitopes to which the specific affinity reagent binds.

[0005] One aspect of the present invention provides a method for selecting an affinity reagent that binds to a desired peptide epitope in a variety of sequence environments. The method includes obtaining a peptide library, each peptide having a sequence αXβ and a length k, wherein X is a desired peptide epitope of length m. The method also includes exposing the peptide library to a variety of specific affinity reagents. The method also includes determining which peptides in the peptide library are bound by the specific affinity reagent. In addition, the method also includes selecting the specific affinity reagent using the peptide library if the specific affinity reagent binds to more than a threshold number of peptides in the peptide library.

[0006] One aspect of the present invention provides a method for selecting an affinity agent for binding to a desired peptide epitope in a variety of sequence environments. The method includes obtaining a peptide library and selecting an affinity agent using the peptide library, each peptide having a sequence αXβ, wherein X is a desired peptide epitope, and wherein α and β each comprise an amino acid. In some cases, α and β each consist of one amino acid. In some cases, at least one of α and β comprises a connector. In some cases, at least one of α and β comprises a modification. In some cases, multiple affinity agents are selected using the methods described herein. In some cases, the desired peptide epitope X is 2 to 7 amino acids. In some cases, the selected affinity agent is an aptamer. In some cases, the selected affinity agent is an antibody. In some cases, the method further includes screening the selected affinity agent for multiple peptides having the same length as the desired epitope to characterize the binding of the selected affinity agent. In some cases, the multiple peptides having the same length as the desired epitope represent 90% of all possible variations of peptides having the same length as the desired epitope. In some cases, the method further includes characterizing the combination of the selected affinity reagent by screening the selected affinity reagent for multiple peptides, wherein each peptide in the multiple peptides comprises a desired epitope and one or more flanking residues. In some cases, the method further includes characterizing the combination of the selected affinity reagent by screening the selected affinity reagent for a second multiple peptide, wherein each peptide in the multiple peptides comprises a second epitope identified in the characterization step, and one or more flanking residues. In some cases, the method further includes characterizing the combination of the selected affinity reagent by screening the selected affinity reagent for multiple subsets (subset) of multiple peptides shorter than the desired epitope. In some cases, the method further includes characterizing the combination of the selected affinity reagent by screening the selected affinity reagent for a group of proteins containing the triamino acid epitope. In some cases, if the selected affinity reagent is combined with more than about 10%, 20%, 30%, 40%, 60%, 75% or 90% of the protein, the affinity reagent is retained. In some cases, the method further comprises characterizing the selected affinity reagent by screening the selected affinity reagent against a panel of proteins that do not contain the three amino acid epitope. In some cases, the selected affinity reagent is retained if it binds to less than about 15%, 10%, 5%, 1%, 0.1% or 0.01% of the proteins.

[0007] In some cases, the selected affinity agent is screened against a library of random peptides of length 2 to 10 amino acids, and further comprises: identifying peptides bound by the selected affinity agent; and retaining the affinity agent if the selected affinity agent 1) binds less than or equal to the number of epitopes determined by equation 20*(k-2), where k is the length of the central epitope, and 2) binds at least 10% of all longer peptides containing the epitope. Since affinity agents are selected based on their ability to bind to a specific central epitope of length k, it should be noted that the percentage of the proteome that can bind to a specific affinity agent generally decreases as the length k of the specific central epitope increases. Although there are some embodiments in which it is beneficial to select affinity agents that bind to a high percentage of the proteome, thereby having the benefit of identifying affinity agents that may have a smaller central epitope length k, it may also be advantageous to identify and select more specific affinity agents even if they may cover a smaller percentage of the proteome. In particular, it may be beneficial to use the selected affinity agents in pools designed for different scenarios of protein identification. For example, if the selected affinity reagents are able to rapidly discriminate between a small number of candidate proteins that have been narrowed down using other methods, then these selected affinity reagents may be highly advantageous for use in a pooled set of affinity reagents.

[0008] In some cases, the affinity reagent is not retained unless the selected affinity reagent binds at least 15%, 20%, 30%, 40%, 50%, 75%, 90% or 95% of all longer peptides comprising the epitope. In some cases, the method further comprises characterizing the binding of the affinity reagent by screening the affinity reagent against a peptide library, the peptide library comprising peptides shorter than the desired epitope, peptides of the same length as the desired epitope, and peptides longer than the desired epitope; and identifying the sequence of the bound peptides, thereby characterizing the binding of the affinity reagent. In some cases, the affinity reagent is fixed to one or more substrates, such as a solid support, and wherein the random peptide library is passed through the fixed affinity reagent. In some cases, the affinity reagent is fixed, and in other cases, the target is fixed. In some cases, the solid support can be a bead. One aspect of the invention provides a method for selecting an affinity agent from a group of affinity agents, comprising screening the group of affinity agents against a library of random peptides of 2 to 10 amino acids in length; identifying peptides bound by each affinity agent in the group of affinity agents; and selecting an affinity agent that 1) binds to less than 20 different epitopes and 2) binds to at least 10% of all longer peptides containing the epitope. In some cases, the affinity agent is not selected unless the affinity agent binds to at least 15%, 20%, 30%, 40%, 50%, 75%, 90% or 95% of all longer peptides containing the epitope. One aspect of the invention provides an affinity agent that specifically binds to an amino acid epitope, does not bind to more than 19 other amino acid epitopes, and binds to at least 10% of the sequence in the form of αXβ, where X is the desired epitope and α and β are any amino acid residues. In some cases, the affinity agent binds to at least 0.25%, 0.5%, 0.75%, 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, 15%, 20%, 30%, 40%, 50%, 75%, or 90% of the sequence of the αXβ form.

[0009] One aspect of the invention provides an affinity reagent that specifically binds to a tri-amino acid epitope, does not bind to any other tri-amino acid epitope, and binds to a desired epitope with substantially similar affinity, regardless of the flanking sequences around the desired epitope. In some cases, the affinity reagent does not bind to a subset of the epitopes. Another aspect of the invention provides an affinity reagent that preferentially binds to a known set of tri-amino acid epitopes, for which the priority has been determined relative to other epitopes, and which is affected by the flanking residues.

[0010] Incorporation by reference

[0011] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The novel features of the present invention are particularly set forth in the appended claims. The features and advantages of the present invention will be better understood by referring to the following detailed description of illustrative embodiments utilizing the principles of the present invention and the accompanying drawings, in which:

[0013] Figure 1 An immobilized target for selecting an affinity agent, according to some embodiments, and an exemplary list of peptides comprising the target are shown.

[0014] Figure 2 A flow chart is shown of an affinity reagent selection process according to some embodiments.

[0015] Figure 3A An exemplary flow chart illustrating the screening step in the affinity reagent characterization process according to some embodiments.

[0016] Figure 3B Shown is an array of 5-mer αXβ peptides used to perform a screening step to determine the effect of flanking sequences on binding of an affinity agent to epitope AAA, according to some embodiments.

[0017] Figure 4 A computer control system programmed or otherwise configured to implement the methods provided herein is shown.

[0018] Figure 5 shows examples of targets embedded in longer sequences with different secondary structures. Figure 5A Shown are examples of peptides that form alpha helices with targets (shown as checkerboard boxes) embedded in different regions of the alpha helical peptide, according to some embodiments. Figure 5B Shown are examples of peptides that form beta sheets with targets (shown as black solid boxes) embedded in different regions of the beta sheet forming peptide, according to some embodiments.

[0019] Figure 6 Shown are target binding for several aptamers identified in an aptamer selection screen, according to some embodiments.

[0020] Figure 7 Shown is a preliminary characterization of the identified aptamers, according to some embodiments.

[0021] Figure 8 Shown is a preliminary characterization of the identified aptamers, according to some embodiments.

[0022] Fig. 9Mass spectrometry verification of synthesized peptides is shown, according to some embodiments.

[0023] Fig.10 An example of cluster amplification according to some embodiments is shown.

[0024] Fig.11 Shown is the binding of a fluorescently labeled peptide target to an aptamer cluster on a flow cell, according to some embodiments.

[0025] Fig.12 The binding affinities of several aptamers to peptide targets are shown, according to some embodiments. DETAILED DESCRIPTION

[0026] Across the life sciences, there is a strong interest in the development of affinity reagents that can bind to specific proteins, metabolites, cells, or cellular interfaces. Recently, affinity reagent selection techniques have been expanded to include unnatural nucleotides and amino acids. The goal of these approaches is to develop reagents that bind only to a given epitope.

[0027] Exclusivity of binding is considered a desirable property of an affinity reagent. Considerable effort has been made to ensure that affinity reagents bind only to one protein, with minimal binding to other proteins. Exceptions are antibodies directed against functionally important residues such as phosphotyrosine.

[0028] A particular challenge faced by affinity reagents can be environmental sensitivity. For example, an affinity reagent may bind perfectly to a core epitope, but may be biased to bind better or not at all depending on the flanking residues. For example, when generating an affinity reagent for the peptide n-Glu-Gln-Lys-Leu-Ile-Ser-Glu-Glu-Asp-Leu, the reagent may bind better if the n-terminal residue is Gly than if it is Ser.

[0029] Although it may generally be undesirable to generate an affinity reagent that is not specific for any one protein, there are specific use cases where it may be optimal for an affinity reagent to be specific for one or more peptides that may occur in many proteins.

[0030] As used herein, the term "epitope" may refer to a portion of a macromolecule such as a protein or peptide that is recognized by an affinity reagent. In some cases, an epitope may be a portion of a protein or peptide that is recognized by an antibody. In some cases, an epitope may be a portion of a protein or peptide that is recognized by an antibody fragment. In some cases, an epitope may be a portion of a protein or peptide that is recognized by an aptamer. In some cases, an epitope may be a portion of a protein or peptide that is recognized by a peptide.

[0031] As used herein, the term "antigenicity" may refer to the ability of a chemical structure (antigen, hapten, epitope, or amino acid sequence) to specifically bind to a group of specific products of adaptive immunity or to a class of affinity reagents. The term "antigenicity" may be used interchangeably with the term "aptagenicity" - the ability of a chemical structure to be recognized by an aptamer. The term "antigenicity" may also be used interchangeably with the term "aptagenicity", which generally refers to the ability of a chemical structure to be recognized by an affinity reagent.

[0032] Choice of affinity reagent

[0033] New affinity reagents can be generated by any method known in the art. The method of developing affinity reagents includes systematic evolution of ligands (SELEX) of exponential enrichment, phage display, yeast display, mammalian cell display, insect cell display, ribosome display, particle display, peptidomer (peptimer) evolution, peptidomer design and inoculation. In some instances, affinity reagents can be designed using structure-based drug design methods. Structure-based drug design (or direct drug design) utilizes the knowledge of the three-dimensional structure of the epitope of interest and the binding site of the affinity reagent.

[0034] In some cases, affinity reagents of the present invention can be selected for the ability to bind to the desired epitope independently of the sequence environment. In some embodiments, affinity reagents can be designed to bind to the desired epitope when the protein is in a denaturing environment. In some embodiments, affinity reagents of the present invention can be selected for the ability to bind to the desired epitope in the protein under folding or unfolding conditions. In some embodiments, denatured proteins can contain or produce microfolding within the protein. In some embodiments, affinity reagents selected to identify the desired epitope AAA can be equally well or almost equally well combined with all peptides comprising the sequence AAA. In some cases, affinity reagents of the present invention can be the desired epitope with different affinities according to the sequence environment of the epitope. In some cases, affinity reagents of the present invention can bind to several different epitopes, regardless of the sequence environment. In some cases, affinity reagents of the present invention can bind to several different epitopes with different affinities according to the sequence environment. Identification of such affinity agents can be accomplished through a three-step screening process: 1) an initial screen for binding to a target comprising an epitope, 2) a qualifying screen at the peptide level to characterize the binding of the affinity agent, and 3) a screen at the protein level to confirm the binding characterization seen in step 2. In some cases, step 3 can be performed prior to or in part prior to step 2. In some cases, step 2 can be omitted, and step 3 may be sufficient to characterize the binding of the affinity agent.

[0035] In some cases, the desired epitope can be a peptide. In some cases, several different epitopes may be required, in which case an affinity reagent that binds to the desired epitope can be selected. In some cases, one or more desired epitopes may be referred to as X. In some cases, the epitope is a discontinuous epitope. For example, the epitope may include every second amino acid residue. In another example, the epitope may be included in several amino acid residues that are adjacent to each other in the secondary or tertiary structure of the protein, even if these residues are not adjacent in the protein sequence. In some cases, the epitope is a continuous epitope. In some embodiments, the desired epitope X is a short amino acid sequence of 2, 3, 4, 5, 6 or 7 amino acids. In some cases, X includes several different short amino acid sequences. In some embodiments, the desired epitope X is a triamino acid sequence X1X2X3. Affinity reagents that bind to the desired epitope in a variety of sequence environments can be identified by screening affinity reagents that bind to a target that includes the desired epitope.

[0036] The target may include a peptide comprising a desired sequence X. In some cases, the target is a pool of peptides that are all sequence X. In some embodiments, the target may include a pool of peptides of sequence αXβ, where X is a desired epitope, and α and β may be any sequence of zero, one or more than one amino acids. For example, if the desired epitope X is AAA, examples of sequences that may be found in the target peptide may include: AAAAA, AAAAC, CAAAA, CAAAC, and CAAAD. In some cases, α and β may each be any single amino acid. In some cases, at least one of α and β may be 2, 3, 4, 5, 6, 7, 8, 9, 10 or more than 10 amino acids. In some cases, at least one of α and β may include a connector or spacer. The connector or spacer may be any connector or spacer known in the art. In some cases, the connector is an amino acid connector. In some cases, the connector is a polyethylene glycol (PEG) or a PEG polymer chain. The PEG chain can be composed of 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30, 34, 36, 38, 40, 42, 44, 46, 48, 50 or more than 50 PEG parts. In some cases, the linker can be a carbon chain. The peptide may also include an N-terminal or C-terminal modification, such as a cap. In some cases, the peptide may be modified to remove the charge, for example, terminal amidation (C-terminal) or acetylation (N-terminal). In some cases, the αXβ peptide may include non-naturally occurring amino acids. In some cases, the αXβ peptide can be modified with a linker and a functional group. For example, the molecule may have a structure FL-αXβ, wherein F is a functional group and L is a linker. In other cases, the molecule may have a structure αXβ-LF, wherein F is a functional group and L is a linker. In some cases, α and β can each be glycine, or can each be one or more glycine residues. In some embodiments, the residue can be modified to change its aptamerogenicity. For example, the residue can be changed by adding a positive charge; adding a negative charge; adding a hydrophobic group; modifying to add a sugar; or other modifications to add chemical diversity.

[0037] Any method known in the art can be used to synthesize peptides. There are several commercial platforms for peptide synthesis, such as MultiPep Rsi synthesizer (Intavis, Germany). Liquid or solid phase methods can be used to synthesize peptides. Synthesized peptides can be verified using any known peptide analysis method. For example, the following methods can be used to verify peptides: mass spectrometry, matrix-assisted laser desorption / ionization time-of-flight mass spectrometry (MALDI-TOF), matrix-assisted laser desorption / ionization, AMS (accelerator mass spectrometry), gas chromatography-MS, liquid chromatography-MS, inductively coupled plasma mass spectrometry (ICP-MS), isotope ratio mass spectrometry (IRMS), ion mobility spectrometry-MS, tandem MS, thermal ionization-mass spectrometry (TIMS) or spark source mass spectrometry (SSMS). The concentration of synthetic peptides can also be evaluated by spectroscopy. Peptide synthesis reactions and examples of verification are provided in Example 8.

[0038] Figure 1 An exemplary list of immobilized targets for selecting affinity reagents and peptides comprising the targets is shown, according to some embodiments. Figure 1 In the example of , the desired epitope is AAA, and the peptide of the target comprises the sequence αAAAβ, where α and β are each a single amino acid. In this example, the target comprises 400 different peptides, representing every possible sequence of αAAAβ, where α and β are each a single amino acid.

[0039] In this way, for any given 3-mer epitope, a target comprising a 5-mer pool can contain 400 different sequences (20 possibilities for α and 20 possibilities for β, where α and β are each a single amino acid). In some cases, a target may contain a pool of peptides longer than 5 amino acids, where each or both of α and β may contain two or more amino acids. In some cases, one of α and β may contain zero amino acids, while the other of α or β may contain one or more amino acids. In some cases, a target may contain a peptide of sequence X without additional amino acids.

[0040] In some cases, the target sequence X can be embedded in a longer sequence. For example, the target sequence X can be embedded in a 15-mer. The target sequence X can be embedded in any position within the 15-mer, for example, in the case of a triamino acid target sequence X, the target sequence X can start at the 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, 8th, 9th, 10th, 11th, 12th or 13th position of the 15-mer. The peptide containing the embedded target sequence can be synthesized in solution, or can be synthesized on a chip such as a PEPperPRINT chip or other peptide array. In some embodiments, the peptide containing the embedded target sequence can be combined or synthesized on a single molecule protein array. Longer sequences can be selected to form secondary structures, or lack secondary structures. Examples of such secondary structures include alpha helices, beta folds, proline bends, turns, loops and cysteine ​​bridges. In some cases, longer sequences can include non-naturally occurring amino acids or other groups.

[0041] The initial selection step may include screening an affinity reagent library for a target comprising a desired epitope. The affinity reagent library may include DNA, RNA or peptide aptamers having a random sequence or a sequence similar to a known protein binding aptamer sequence. In some cases, the aptamer library may be a commercial library. In some cases, the aptamer library may be obtained from a research institute, a university, an academic center or a research center. In some cases, the library may include a bead library. In some cases, the aptamer library may be generated from a library of known sequences or from a random sequence. In some cases, the aptamer library may include aptamers with a specific structure, such as a stem-loop library. In some cases, the aptamer library may include switchable aptamers-aptamers that can switch between two conformations. For example, an aptamer may require a metal ion cofactor to form a first conformation, and a chelating agent such as EDTA or EGTA is added to chelate the metal ions and cause the aptamer to change to a different conformation. Other factors that can be used to induce aptamer switching include light, pH, temperature, magnetic field and electric current.

[0042] Screening of aptamer libraries for targets can be performed by any method known in the art. On the one hand, the target can be fixed on a solid support, and aptamers can be added under conditions that allow low specific binding to aptamers. Unbound aptamers can be washed out of the target by a series of washings with gradually increasing stringency. The aptamers that remain bound to the target through the washing steps can be sequenced and amplified for further selection rounds, or for designing other aptamers with high sequence similarity. Several rounds of target binding, washing, sequencing and amplification can be repeated, or new aptamers can be designed until aptamers with desired specificity and binding affinity are generated. It is also possible to use beads that each contain multiple copies of aptamers, and use a bead-based method to screen aptamer libraries. It is also possible to use an array-based method to screen aptamer libraries, such as by spotting multiple copies of each aptamer in the library onto an array, and then evaluating the spots bound by the target. The aptamer library can also be screened using a particle display method. In some embodiments, the aptamer library can be screened using a single molecule protein array.

[0043] In some cases, the percentage of the target pool to which the identified affinity reagent is bound can be measured, for example, by comparing the number of bound copies of the affinity reagent with the number of target peptides available for binding. In some embodiments, the affinity reagent can bind to 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80% or more than 80% of the peptides comprising the target. In addition, once a specific affinity reagent has been identified and selected, the affinity reagent can be verified. In some embodiments, the selected affinity reagent can be verified for multiple sequences containing epitopes (to which the affinity reagent is characterized as binding). In some embodiments, the selected affinity reagent can be verified by evaluating the selected affinity reagent for multiple protein sequences on a single molecule protein array.

[0044] Figure 2 A flow chart of an affinity reagent selection process according to some embodiments is shown. First, a target is obtained and fixed on a solid support. The target is then exposed to an affinity reagent library. In this example, the affinity reagent is an aptamer. Unbound aptamers are washed off from the target, and the remaining aptamers are eluted. The eluted aptamers can be sequenced in a manner that retains the aptamers, or aliquots of the eluted aptamers can be sequenced. Based on the sequencing results, it is determined whether a single aptamer or a small group of aptamers is highly enriched, indicating strong binding, and should be selected for further screening, or whether there are many different aptamers showing mild enrichment, in which case the eluted aptamers can be amplified and reapplied to the target. These steps can be repeated until an aptamer with a desired binding affinity is produced. The stringency of the washing step can be increased in subsequent washing steps. The length of the washing step can be selected to obtain an affinity reagent with a low dissociation rate (off rate).

[0045] In other aspects, the initial selection step may include fixing the aptamers of the library on a solid support and adding a labeled target. The solid support may be a surface within a glass slide, a bead, a magnetic bead, or a flow cell. The aptamers of the library may be fixed as a single copy, or may be fixed in the form of a pool. For example, multiple copies of a single aptamer may be fixed on an area of ​​the solid support, while multiple copies of other aptamers may be fixed on other areas of the solid support. In some cases, the aptamers of the library may be modified with an adaptor and hybridized with an oligonucleotide-coated solid support. Cluster amplification may then be used to locally amplify each aptamer on the solid support. In some cases, the solid support is a glass slide. In some cases, the solid support is a flow cell. In some cases, the solid support is a flow cell suitable for fluorescence imaging. In some cases, the solid support is a magnetic bead or multiple magnetic beads. Each of the multiple magnetic beads may be coated in multiple copies of a single different aptamer so that each magnetic bead is coated with a different aptamer from each other.

[0046] In other aspects, the initial selection step may include injecting the target into a host animal capable of producing antibodies against the target. The host animal may be any animal capable of producing antibodies, such as rabbits, goats, mice, rats, horses, cattle, sheep, camels or donkeys. Antibodies may be recovered from the serum of the host animal. The serum from the host animal may be used as is, or the antibodies may be purified from the serum. Methods for purifying antibodies include physicochemical fractionation, class-specific affinity or antigen-specific affinity. Class-specific affinity may involve binding of antibodies to immobilized biological ligands that have specific affinity for immunoglobulins. Antigen-specific affinity may involve using an immobilized target to pull down antibodies that bind to the target. Purified antibodies may be sequenced to identify antibodies from serum, and the identified antibodies may then be synthesized in vivo or in vitro.

[0047] In some cases, instead of extracting antibodies from the serum of an immunized host animal, the spleen can be extracted and the splenocytes can be immortalized. One method of immortalizing splenocytes can be to fuse the cells with myeloma cells to form hybridomas. Single clones can be isolated from hybridomas, and each clone will produce a monoclonal antibody. Methods similar to those described above can be used to screen different monoclonal antibodies for targets. In short, the target can be fixed on a solid support, and antibodies can be added under increasingly stringent conditions to determine antibodies that bind with desired affinity. The sequence of the selected monoclonal antibody can be obtained by sequencing the protein or sequencing the coding sequence from the cell line that produces it.

[0048] In some cases, phage display can be used to select antibodies. A phage library can be obtained, wherein each phage expresses antibodies on its surface and encodes the sequence for generating the antibody. The phage library can be applied to a peptide target, and the phage expressing and encoding the antibody that is not combined with the target can be washed away. The phage expressing and encoding the antibody in conjunction with the target can be washed out from the target. The phage washed out can be amplified by infecting cells with phage, and the selection step can be repeated with the amplified phage with the required number of repetitions. Subsequent repetition can utilize the washing buffer solution that stringency increases gradually or the washing time increases gradually. Once the required number of repetitions has been completed, the selected phage can be analyzed, for example, by cracking the phage to sequence the DNA of the antibody expressed by the encoding. The sequence can be used to construct a cell line expressing the selected antibody.

[0049] In some cases, antibodies or other affinity reagents can be selected using bacteria, mammalian cells, insect cells or yeast display. These methods are similar to the phage display described above, but bacteria, mammalian cells, insect cells or yeast libraries are obtained, wherein each bacteria, mammalian cells, insect cells or yeast cells express antibodies on their surfaces and encode sequences for generating the antibodies. The selection method is identical to the selection method described for phage display.

[0050] In some embodiments, affinity reagents can be selected in parallel for many different targets. For example, many different targets can be fixed to different positions on the array and used with the SELEX selection method. In another example, many different targets can be attached to magnetic beads and used with the SELEX selection method. In some embodiments, affinity reagents can be selected for targets on a single molecule protein array.

[0051] In one embodiment, an Illumina-type next-generation sequencing platform can be adapted for aptamer selection. An aptamer library can be labeled with an adapter at each end and hybridized with an oligonucleotide in a flow cell. Solid-state amplification is used to create aptamer clusters from each starting aptamer in the aptamer library. The surface of the flow cell can now be covered with multiple nucleic acid clusters, each of which contains many copies of a single aptamer from the aptamer library. A sequencing reaction can be run on the flow cell using a nucleic acid polymerase, primers, and four different fluorescently labeled reversible terminators (A, T, C, and G). This can provide the sequence of the aptamer at each position in the flow cell. Once the identity of each aptamer cluster is known, one or more detectably labeled peptide targets can be added to the flow cell and incubated with the aptamer clusters. After a period of incubation, unbound targets are washed away and the flow cell is imaged to determine the position to which the target is bound. When multiple stages of washing steps are performed, K can be generated. DIn particular, after an initial wash step to remove unbound targets, an initial image can be taken to determine which bound targets are still present. After taking the initial image, a second wash step can be performed to remove targets that have become unbound over the elapsed time, and then a second image can be taken. This process can be repeated over multiple stages to calculate K across multiple targets based on the time they became unbound. D In some cases, the initial wash step may be followed immediately by a second wash step. K may be measured using techniques including, for example, surface plasmon resonance or biolayer interferometry (BLI) techniques. D Affinity determination.

[0052] In some cases, many different targets with different distinguishable labels can be added to the flow cell to perform multiple parallel selections. An example of this approach is provided in Example 7.

[0053] In some embodiments, the ease of affinity reagent selection may be affected by the sequence / structure of the epitope. Epitopes with high "immunogenicity", "antigenicity", "aptamerogenicity" or "affinity reagent genicity" may be easier to design affinity reagents for. For example, epitopes containing amino acid residues with very different chemical properties may be easier to select affinity reagents for. Epitopes such as GGG or AAA may be more difficult to select affinity reagents for than epitopes such as KWK or DCY. In some embodiments, epitopes can be modified, for example, using chemical bioconjugation to identify affinity reagents that interact with the modified epitope. Amino acids KRDEYWC are particularly easy to modify. In some embodiments, when identifying affinity reagents that bind to the desired epitope, it may be advantageous to modify DYW with a chemical reaction that adds a component that is easy to identify using an affinity reagent. In particular, DYW can be modified to add positive charges or nucleic acids, and then affinity reagents customized for the component can be used. In this way, the DYW epitope may be more easily highlighted. In addition, when the platform is run to identify a desired epitope that has been modified, a protein sample (such as a blood sample) can be exposed to a chemical reaction that will modify the DYW epitope in an expected and specific manner, which can then allow the protein to be evaluated for a specific affinity reagent to determine whether an unmodified DYW epitope corresponding to a modified DYW epitope with a positive charge or nucleic acid component is present in the original protein sample. In some embodiments, the chemical reaction to the DYW epitope can be reversible. In some embodiments, the chemical reaction to the DYW epitope can be irreversible. In this way, a customized assay can be run to identify a modified DYW epitope.

[0054] In some cases, the set of potential targets can be stratified into different groups based on the "antigenicity" or "affinity reagent originality" of the prediction of the target epitope. For example, in some cases, both high antigenicity targets and low antigenicity targets are combined in a single parallel selection step, which can cause the affinity reagent for the high antigenicity target to annihilate the signal from the affinity reagent for the low antigenicity target. The predicted antigenicity can be based on the chemical properties of the amino acid residues in the target, wherein charged and larger amino acids are expected to have higher antigenicity. The antigenicity of a target containing two or more identical or chemically similar amino acid residues may be lower than the antigenicity of a target containing amino acid residues with very different chemical properties. In some cases, the antigenicity of different targets can be determined experimentally according to the result of the initial affinity reagent selection screening.

[0055] Characterization and screening of affinity reagents

[0056] Any affinity agent identified as binding to a target can be subjected to peptide characterization or qualification screening. Affinity agents can be identified in screening as described above, or can be obtained by any method known in the art. In some embodiments, a second characterization screen can be used to determine the specificity of an affinity agent to a desired epitope relative to other possible sequences. For a second screen, a library of possible n-mer sequences can be used to determine whether the affinity agent recognizes other sequences. In some cases, the library contains at least 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, 99.9% or more than 99.9% of all possible n-mer sequences. In some cases, the library contains all possible n-mer sequences. If it is believed that the affinity agent binds to 3-mers, other sequences can be selected to represent all possible 3-mers (20 3 =8000). In some embodiments, the sequence selected to represent the possible epitope of interest can be embedded in other sequences such as flanking sequences. In some embodiments, the sequence selected to represent the desired epitope can have flanking sequences or chemical linkers added to the side of the desired epitope that is attached to the solid substrate so that the desired epitope is located far enough from the solid substrate to allow the affinity reagent to approach the desired epitope. If the affinity reagent is believed to bind to 4-mers, other sequences can be selected to represent all possible 4-mers (20 4=160,000). In one embodiment, pools of different n-polymers can be fixed in different wells of a multi-well plate, or fixed on different areas of a solid support. The affinity reagent can then be incubated with the library, and the unbound affinity reagent can be gently washed away. The area where the affinity reagent has been bound to the library can be detected by any method suitable for visualization of the affinity reagent. In some instances, the affinity reagent can be visualized by a dye or reagent that binds to a protein and / or nucleic acid, or the affinity reagent can be labeled with a detectable portion before being added to the library. In some cases, the binding of the affinity reagent to the target can be detected by fluorescence resonance energy transfer (FRET) microscopy, surface plasmon resonance (SPR), bioluminescence resonance energy transfer (BRET), NanoBRET, biolayer interferometry (BLI) or Octet.

[0057] In some cases, the presence or absence of an affinity agent can be determined under several different conditions - e.g., different wash stringencies, which can be achieved, for example, by using different wash buffers, different wash times, different wash temperatures, or different agitation levels. In some cases, binding of an affinity agent can be measured repeatedly over time to generate a time course of binding. In some cases, determining binding of an affinity agent under different conditions and / or at different times can enable estimation of the K of the affinity agent for the target. D , K ON or K OFF .

[0058] In some embodiments, different washing and detection conditions can be used to identify affinity reagents with desired properties, such as desired on- and off-rates, or association constants (K ON ) and dissociation constant (K OFF ). In some cases, the desired affinity reagent may have a slow dissociation rate. The affinity reagent epitope binding reaction reaches equilibrium when:

[0059] [Affinity reagent]·[Epitope]·K ON =[affinity reagent-epitope]·K OFF .

[0060] In some cases, the K of the preferred affinity reagent is OFF The value can be lower than about 1s -1 , 10 -1 s -1 , 10 -2 s -1 , 10 -3 s -1 , 10 -4 s -1 , 10 -5 s-1 , 10 -6 s -1 , 10 -7 s -1 , 10 -8 s -1 , 10 -9 s -1 or 10 -10 s -1 . In some cases, the dissociation rate may be more critical than the association rate because a poor association rate can be compensated for by increasing the amount of affinity reagent added to the reaction or by extending the incubation time before washing. In some cases, the affinity reagent is selected so that the dissociation rate is sufficient to enable the affinity reagent to remain bound after one or more washing steps until after the imaging step is completed. In some cases, the affinity reagent can remain bound to the epitope for at least about 1 min, 2 min, 3 min, 4 min, 5 min, 6 min, 7 min, 8 min, 9 min, 10 min, 11 min, 12 min, 13 min, 14 min, 15 min, 16 min, 17 min, 18 min, 19 min, 20 min, 25 min, 30 min, 35 min, 40 min, 45 min, 50 min, 55 min, 60 min, or more than about 60 min.

[0061] In some cases, the preferred affinity reagent is only combined with one of the 3-polymers in the library. In other cases, the preferred affinity reagent can be combined with 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 40, 50, 60, 70, 80, 90, 100 or more than 100 different 3-polymers. In some embodiments, the step of washing out the affinity reagent can be repeated with increasing stringency, and the affinity reagent that remains bound can be detected after each washing step.

[0062] In another method, affinity reagent can be fixed on solid support, and the pool of all possible n-aggregates can be added to affinity reagent, unbound n-aggregates can be removed by washing, and remaining n-aggregates can be analyzed to find the sequence bound by affinity reagent. The screening step repeated can be carried out with the washing step of different stringency. Any suitable method can be used, for example mass spectrometry or high performance liquid chromatography mass spectrometry to identify the n-aggregates of binding. In some cases, random n-aggregates pools can be used, rather than the pool of all possible n-aggregates. When random n-aggregates pools are used to characterize screening, combined and unbound n-aggregates can all be identified. It can be pointed out that sequences not represented in combined or unbound n-aggregates pools are for tracking in subsequent characterization screening. In some cases, the random n-mer pool can be characterized prior to application to an affinity reagent to ensure that the pool contains at least 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or more than 99% of the possible n-mer sequences. In some cases, a preliminary characterization screen can be performed with random n-mers, and affinity reagents that bind less than about 30, 25, 20, 15, 10, or 5 n-mers can be selected for further characterization using a pool of all possible n-mers.

[0063] Can carry out further peptide qualification screening, to determine whether affinity reagent is specifically bound to desired epitope, or whether affinity reagent is bound to the sub-part of epitope.For example, it is believed that the affinity reagent that is specific to 3-aggregate sequence AKD may actually be specific to dimer sequence AK.Therefore, for every kind of n-aggregate affinity reagent, can carry out screening step for the set of all (n-1)-aggregates.In some cases, when the set of all n-aggregates that have been characterized and screened and bound by affinity reagent is known, screening can be limited to be contained in the (n-1)-aggregate in the n-aggregate bound by affinity reagent.For example, if the only 4-aggregate bound by affinity reagent is AAKD, then it can be screened for AAK and AKD instead of all possible 3-aggregates.If affinity reagent is really bound to (n-1)-aggregate, then it can be screened for (n-2)-aggregate, to further limit the specificity of the affinity reagent.In some cases, it is possible not to select the affinity reagent bound to (n-1)-aggregate for further screening. In other cases, affinity agents that bind to no more than one (n-1)-mer may be selected. In other cases, affinity agents that bind to more than one (n-1)-mer may be retained for further screening.

[0064] Additional peptide qualification screening can be performed to determine the effect of flanking sequences on the affinity of affinity reagents to one or more epitopes thereof. For this step, a peptide library of sequence αXβ can be created (wherein X is an affinity reagent epitope, or multiple epitopes, and α and β can be any one or more amino acids). In the example of an affinity reagent that binds to a single epitope, if α and β are each a single amino acid, assuming that the sequence of the epitope to be bound is set, then regardless of the length of the desired epitope, there are 400 possible αXβ sequences. For affinity reagents that bind to two or more different epitopes, or where α and β may each be zero, one or more than one amino acid, there may be many possible αXβ sequences. In some cases, a group of αXβ peptides used in this screening step may be the same group of αXβ peptides that comprise the target for selecting an affinity reagent. Affinity reagents can be screened for this group of αXβ sequences using standard screening methods. Viewing a list of αXβ sequences bound by an affinity reagent and relative binding affinity can allow determination of the effect of flanking sequences on the binding affinity of an affinity reagent. If a strong effect is observed, further screening can be performed with longer flanking sequences, for example, if the first sequence library consists of peptides αXβ, where α and β are each a single amino acid, a subsequent screening can be performed with a peptide library consisting of γXδ, where γ and δ each contain two amino acids.

[0065] Figure 3A Shown is an exemplary flow chart of the screening step in the affinity reagent characterization process according to some embodiments. The affinity reagent is labeled with a fluorescent marker to facilitate the detection of the affinity reagent. The affinity reagent is then applied to a peptide pool containing all possible 3-mer sequences. The peptide pool containing all possible 3-mer sequences can be fixed on a solid support as an array. In some embodiments, the peptide pool containing all possible 3-mer sequences can be fixed on one or more single-molecule protein arrays. In some embodiments, multiple 3-mer sequences can be fixed on a solid substrate. In some embodiments, multiple 3-mer sequences can be fixed on a single-molecule protein array. In another step that can be performed in parallel, the affinity reagent is then applied to a peptide pool containing all possible 2-mer sequences. All 3-mers and 2-mer peptides bound by the affinity reagent are identified. A 5-mer peptide pool is obtained, which contains all possible 5-mer peptides, and the 5-mer peptides contain the sequences of the identified 3-mers and 2-mers of the binding. For example, if the affinity reagent is known to bind only to epitope AAA, then the pool of all possible 5-mer peptides consists of peptides αAAAβ, such as Figure 3BAs shown. The affinity reagent is then applied to the 5-mer peptide pool, and the data from all three screening steps are compiled to provide the binding characterization of the affinity reagent. In other examples, the steps can be omitted or performed in different orders. Protein qualification screening can be performed to confirm that the affinity reagent binds to the protein rather than one or more desired epitopes in the peptide environment. Proteins of known identity and sequence can be fixed on a solid support and then exposed to the affinity reagent. Proteins can be applied to the solid support from purified protein reserves, or can be synthesized on a solid support by processes such as nucleic acid programmable protein arrays (NAPPA). In some cases, the protein used for the screening can be composed of proteins including one or more predicted or known epitopes of the affinity reagent. In some cases, the protein used for the screening can be composed of proteins that do not include one or more predicted or known epitopes of the affinity reagent. In some cases, the protein used for the screening can be composed of proteins that include one or more predicted or known epitopes of the affinity reagent and do not include these epitopes. The binding of affinity reagents to different proteins can be evaluated by any method known in the art. For example, the bound affinity agent can be identified by an affinity agent-specific antibody, by a dye that binds to a protein or nucleic acid, or by labeling the affinity agent with a detectable portion prior to screening. Once the bound and unbound proteins are determined, the sequences can be compared to determine how often the affinity agent binds to one or more of its epitopes and whether the binding is affected by the surrounding protein sequence. The binding results and protein sequence can also be input into a machine learning algorithm to verify the most likely binding site. Off-target effects can also be determined by this method. In some cases, if the affinity agent for the epitope binds at least about 1%, 2%, 3%, 5%, 7%, 10%, 15%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95% or more than 95% of the proteins containing the epitope, the affinity agent can be selected. In some cases, an affinity reagent for an epitope can be selected if it binds less than about 50%, 40%, 30%, 20%, 10%, 5%, or less than 5% of proteins that do not contain the epitope. In some cases, multiple affinity reagents can be selected for the same epitope. The multiple affinity reagents can be combined together and used as a combined affinity reagent.

[0066] Characterize affinity reagents using binding data from interactions with known proteins

[0067] It is useful to find methods for characterizing the binding affinity of a specific affinity agent to proteins that constitute a proteome (e.g., the human proteome, the yeast proteome, the E. coli proteome). Even when looking at the approximately 70,000 canonical protein sequences defined in the reference proteome database on Uniprot (https: / / www.uniprot.org / proteomes / UP000005640), it takes a lot of effort to characterize the binding affinity of a set of specific affinity agents to a whole set of proteins in the human proteome. In addition, when considering protein variants (proteoform), the number of different protein variants in the human proteome that can be identified can be hundreds of thousands or millions. In this way, it is advantageous to effectively characterize the binding affinity of affinity agents in a way that can be applied to a variety of unknown proteins. In some embodiments, the binding affinity of an affinity agent is generated by evaluating the interaction of an affinity agent with a known protein. In some embodiments, the binding affinity of an affinity agent is generated by evaluating the interaction of an affinity agent with a protein having a sequence derived from a sequence database (such as a reference proteome from NCBI or Uniprot). In some embodiments, the protein sequences used in the methods provided herein may not have a known natural source, such as proteins with random sequences. In some embodiments, printing proteins with non-natural sequences can be used to provide additional input to models such as those discussed herein. In some embodiments, the binding affinity of an affinity reagent to a specific target within a known protein can be evaluated based on the presence of the target within the known protein and based on the number of copies of the target within the known protein.

[0068] Therefore, in some cases, affinity reagents, such as affinity reagents disclosed herein, can be characterized by screening affinity reagents for known protein arrays. The binding data generated by screening affinity reagents for known proteins can be used to generate binding affinity information, which can be used to assess whether there is an epitope or multiple epitopes bound by an affinity reagent in an unknown protein. In particular, each affinity reagent can be bound to one or more epitopes. In addition, each protein can contain multiple epitopes, each of which can be bound to a specific affinity reagent. For proteins containing multiple epitopes bound to a specific affinity reagent, the protein can have multiple copies of a specific epitope and / or can have multiple copies of different epitopes. In this way, in some cases, a specific protein can be bound to multiple affinity reagent molecules, each of which may be attached to one of the multiple epitopes present in the protein. Therefore, information that helps to characterize affinity reagents and predict binding to proteins can be applied to methods for identifying unknown proteins. In some embodiments, the generated binding data can be used to identify unknown proteins whose sequences are represented in a protein database. In some embodiments, the binding data can be used to generate a specific overview, which can be used to characterize an unknown protein until a sequence associated with the unknown protein is identified. In particular, the binding data generated from the protein interactions of an affinity reagent with a known protein array can be used to determine the binding affinity of the affinity reagent to one or more epi-positions within a known protein. Subsequently, this can be used to assess the binding affinity of the selected affinity reagent to proteins having one or more copies of the one or more epi-positions. Additionally or alternatively, the binding affinity of the affinity reagent to one or more epi-positions can be used to assess the binding affinity of the selected affinity reagent to proteins having other epi-positions similar or identical to the characterized one or more epi-positions. In some embodiments, the binding data from the interaction of affinity reagents screened for proteins in a known protein array can be used to determine the affinity properties of a specific type of epitope in one or more proteins within a known protein array, such as each possible epitope of a specific length, an epitope within a specific subset of a complete protein sequence, an epitope at a specific position of a folded protein structure, each epitope predicted or observed to be highly accessible in a folded protein structure, or each epitope identified from an empirical or computer-simulated binding screen. In some embodiments, the binding data from the interaction of affinity reagents screened for proteins in a known protein array can be used to determine the binding affinity of an affinity reagent to multiple epitopes in one or more proteins within a known protein array. In some embodiments, the binding data from the interaction of affinity reagents screened for proteins in a known protein array can be used to determine the binding affinity of an affinity reagent to a protein with multiple copies of a specific epitope.

[0069] The method for affinity reagent screening for known protein arrays is described below. As described below, the method can provide that the affinity reagent tested is in conjunction with at least one triamino acid epitope (trimer) or a trimeric combination. However, the embodiment of the method can include one or more epitopes of different lengths except the trimer, such as dimer, 4-aggregate, 5-aggregate, 6-aggregate, 7-aggregate, longer epitopes and discontinuous epitopes. In some embodiments, the method can provide that the affinity reagent tested is in conjunction with at least one dimer or a dimer combination. In some embodiments, the method can provide that the affinity reagent tested is in conjunction with at least one dimer or a trimer, or a dimer or a trimer combination. In some embodiments, the method can provide that the affinity reagent tested is in conjunction with two or more n-aggregates, and the n-aggregate is selected from dimer, trimer, 4-aggregate, 5-aggregate, 6-aggregate, 7-aggregate, n-aggregate longer than 7-aggregate and a discontinuous epitope. In some embodiments, the method can provide that the affinity reagent tested binds three or more n-polymers selected from dimers, trimers, 4-polymers, 5-polymers, 6-polymers, 7-polymers, n-polymers longer than 7-polymers, and discontinuous epitopes. In some embodiments, the method can provide that the affinity reagent tested binds more than three n-polymers selected from dimers, trimers, 4-polymers, 5-polymers, 6-polymers, 7-polymers, n-polymers longer than 7-polymers, and discontinuous epitopes.

[0070] Using the methods provided herein, protein inferences can be calculated. In some embodiments, protein inferences can be calculated based on one or more of the following considerations, such as the binding measurements of a particular affinity agent to a protein; information about the protein sequence of a possible candidate protein for the protein; and information from which predictions with a specific confidence level of the expected degree of binding of each affinity agent to each candidate protein can be derived, such as binding affinity information. Some methods described herein are provided for generating a statistical model of an affinity agent, from which the probability of binding of an affinity agent to the protein can be calculated, given the primary sequence of the protein.

[0071] In some embodiments, a series of binding measurements of each of a plurality of specific affinity reagents and a plurality of known proteins can be used to train the statistical model. When no direct empirical measurement is available, the method can be used to predict affinity reagent-protein binding. In the case of the method being applied to the protein in the human proteome, the minimum list of candidate human proteins can include about 20,000 proteins. In addition, allowing splice variants, polymorphisms, partial degradation and other non-human genomes can rapidly expand the candidate proteins to hundreds of thousands or millions of numbers.

[0072] method

[0073] In some embodiments, an array of known protein sequences is obtained. In some embodiments, this known protein array can be generated by spotting a pure protein sample onto a chip. In some embodiments, this known protein array can be generated by directly translating the protein onto a chip, for example, by using a nucleic acid programmable protein array (NAPPA). The number of known proteins that can be analyzed across one or more specific arrays can include less than 50 different protein sequences, can include 50 different protein sequences, can include more than 50 protein sequences, can include about 100, 150, 200, 250, 300, 350, 400, 500, 600, 700, 800, 900, 1,000, 2,000, 3,000, 5,000, 10,000, 15,000, 20,000 or more than 20,000 protein sequences. As a broad concept, the accuracy of the developed model can be improved when more proteins are analyzed. In some embodiments, the number of known proteins that can be analyzed can be about 300 to about 3000 different protein sequences; about 500 to about 2000 different protein sequences; about 600 to about 1500 different protein sequences; about 600 to about 1000 different protein sequences; or about 1000 to 3000 different protein sequences, etc. In some embodiments, each known protein can be present in multiple copies at a single position on the array. In some embodiments, each protein can be present in multiple copies at many positions across the array. In some embodiments, each protein can be present in two positions, three positions, four positions, five positions, six positions, seven positions, eight positions, more than eight positions, or a combination of different numbers of positions across the array. In some embodiments, each protein may be present in greater than about 1000 copies, 2000 copies, 3000 copies, 5000 copies, 10,000 copies, 20,000 copies, 50,000 copies, 100,000 copies, 500,000 copies, 1,000,000 copies, 2,000,000 copies, 5,000,000 copies, 10,000,000 copies, 100,000,000 copies, 1 billion copies, 10 billion copies, 100 billion copies, or more than 100 billion copies at each position in one, two, three, four, five, six, seven, eight, or more than eight positions across the array. Additionally, in some embodiments, affinity reagents can be characterized by evaluating small amounts of protein at specific locations, such as 1 copy per location, 2 copies per location, 3 copies per location, 4 copies per location, 5 copies per location, 10 copies per location, or more than 10 copies per location.In some embodiments, locations may be distinguished from one another when each particular location is resolvably different from another location based on optical detection or other detection sensors.

[0074] In some embodiments, the binding of one or more affinity reagents to known proteins within the array can be assessed by hybridizing a fluorescently labeled affinity reagent to the array and measuring the fluorescence observed at each spot on the array. In some embodiments, the identity of each protein at each spot can be known, so that the fluorescence (representing binding) can correspond to the protein identity.

[0075] In some embodiments, the trimer epitope can be used to evaluate the binding of affinity reagents to specific epitopes within proteins. However, in some embodiments, the methods discussed herein can be extended to epitopes of different lengths other than trimers. In some embodiments, fluorescence measurements at the protein level can be used to derive a fractional contribution to fluorescence measurements from a single epitope, such as from each single epitope. In some embodiments, protein binding can be simulated as a linear combination of the counts of each of the 8000 possible trimers in the protein sequence and the fractional fluorescence from these trimers. This can be expressed as:

[0076]

[0077] in:

[0078] F pr =Measurement result of protein pr (e.g. fluorescence)

[0079] c t,pr = the count of trimer t in protein pr

[0080] β t = fractional fluorescence from binding of affinity reagent to trimer t

[0081] In some embodiments, the measurement results of affinity reagents for multiple proteins can form a linear system of equations:

[0082]

[0083] in:

[0084] is a length N column vector containing the measurements (e.g. fluorescence) observed for each protein C is an N x 8000 matrix of trimer counts, where each column is the count of a specific trimer in each measured protein

[0085] is the length 8000 column vector of fractional fluorescence from reagent binding to each possible trimer ∈ is a scalar constant corrected for background binding or background noise

[0086] In some embodiments, the methods provided herein may include non-standard amino acids and / or simulate n-mers of different lengths. In the case of known NAPPA or similar binding measurements, and C are known variables, and The values ​​of and ∈ can be derived by linear regression or related methods. In particular, non-negative least squares and non-negative least absolute shrinkage and selection operator (LASSO) regression may be well suited to the considerations of the method described in this paper. Non-negative least squares will The solution of is restricted to non-negative values, and non-negative LASSO regression further imposes sparse constraints. When the system of equations is underdetermined, that is, when the number of unique proteins measured is less than the number of unique trimers (8,000 in this example), LASSO regression is particularly effective. In some embodiments, the fractional fluorescence derived from each epi-position can be used to estimate the binding properties of a specific affinity reagent with each epi-position, such as binding kinetics. In some embodiments, it can be considered that the relative fractional fluorescence is proportional to the relative binding affinity of the affinity reagent with each epi-position. In some embodiments, the fractional fluorescence can be converted to the calculated value of the fraction of the combined site by dividing by the number of fluorescence counts of each fluorophore, and then divided by the expected number of protein molecules of each spot on the array. A simulation of the method is provided in Example 5, wherein a set of binding affinities of theoretical affinity reagents is used to predict the combination of affinity reagents with 720 kinds of human proteins, and then the predicted binding data are solved to determine the affinity of the affinity reagent to each possible trimer epi-position.

[0087] The data collected from the embodiment in which the protein is combined with the affinity reagent for simulation can be used to help train other models for predicting protein binding affinity from the primary sequence of the protein. In some embodiments, the data collected from the embodiment in which the protein is combined with the affinity reagent for simulation can be used to help train other models for predicting protein binding affinity from the derivatives of the primary sequence (e.g., amino acid composition, trimer count, predicted three-dimensional structure). In some embodiments, the model may include parameters similar to those discussed for epitopes of different lengths (e.g., 1 polymer, 2 polymers, 4 polymers, etc.) or mixtures with epitopes of different lengths. In some embodiments, the model can use a nonlinear model, such as an exponential function that associates trimer affinity with observed fluorescence, or a multiplication model in which trimer level affinity is multiplied to generate protein level affinity. In some embodiments, a neural network can be used to predict protein binding affinity from a protein sequence or trimer composition. In addition, a support vector regression model can be used to predict protein binding affinity from a protein sequence or trimer composition.

[0088] Assessing the impact of secondary structure on affinity reagent protein binding

[0089] The binding of affinity reagents to proteins may be affected by the presence of secondary structural elements such as helices, turns, loops, and folds. In some embodiments, methods utilizing peptide arrays and protein structure databases can be used to assess how the presence of secondary structures changes the binding of affinity reagents to epitopes. In some embodiments, information associated with known protein structures can be used to identify protein regions that form different secondary structures such as alpha helices and beta folds and also contain target epitopes. In some embodiments, the target epitope can be within the identified secondary structure. In some embodiments, the target epitope can be near the identified secondary structure. In some embodiments, protein regions with known secondary structures can be synthesized and compared with different protein regions that have not been observed to have secondary structures to assess the binding of affinity reagents. In some embodiments, protein regions with known secondary structures can be synthesized and compared with scrambled sequences containing the target epitope and the same amino acid composition as the structured region but with different amino acid sequences to assess the binding of affinity reagents.

[0090] In some embodiments, the method for measuring the binding affinity of an affinity agent to different secondary structures may include using hundreds or thousands of peptides on a printed peptide array (e.g., those available from PEPperPRINT). For example, an array containing 11,000 peptides, each having a length of 10-30 residues, may be printed and then hybridized with a fluorescently labeled affinity agent. The amount of fluorescence measured at each spot on the array may be considered proportional to the binding affinity of the affinity agent to the corresponding peptide located at the spot. In some embodiments, a method for testing structural effects may be provided. For example, an array may be printed that contains "structural" peptides--for example, peptides such as 15-mers that may be expected to have a specific secondary structure that also includes an epitope of interest. In addition, the array may be printed alternatively or additionally to include "unstructured" peptides that have not been observed to have a secondary structure and also include the same epitope of interest. "Unstructured" peptides can be constructed as follows: Modify a portion of a specific sequence to modify one or more sequence portions associated with a secondary structure, thereby producing a modified sequence that is observed to have less secondary structure than the original peptide, or is observed to have no secondary structure. By measuring different indicators of binding affinity (e.g., the amount of fluorescence, or another indicator of attachment between the affinity agent and the peptide), it can be determined how the presence of secondary structure components can affect the specific binding affinity of a particular affinity agent for one or more epitopes of interest.

[0091] Discovery of structured peptides

[0092] In some embodiments, the sequence for structured peptides can be derived from the region of known structured endogenous proteins. In some embodiments, these peptides can be found by initially searching a reference protein group (e.g., Uniprot human reference protein group) to find all occurrences of epitopes in the annotated protein sequence of the target species. Once the occurrence of the epitope in the annotated protein sequence of the target species is determined, the sequence position comprising the epitope can be cross-referenced with the structural data from a protein database such as Protein Data Bank (PDB). Additionally or alternatively, software for predicting secondary structure can be used to determine whether the sequence position comprising the epitope is related to the target structural motif (e.g., secondary structure). Software programs that can be used to predict secondary structure include but are not limited to Rosetta (https: / / www.rosettacommons.org / software), I-TASSER (https: / / zhanglab.ccmb.med.umich.edu / I-TASSER / ) and PEP-FOLD (http: / / bioserv.rpbs.univ-paris-diderot.fr / services / PEP-FOLD / ), etc.

[0093] Once the sequence position / sequence portion associated with the structural data has been determined, epitopes that are not within the sequence portion associated with the target structural motif (e.g., beta fold, alpha helix, turn) can be removed. In addition, for each case of an epitope within a structural element, the epitope and structured flanking sequence can be extracted from either side of the epitope. Several different extractions can be performed for each epitope, for example, a case where the epitope is at the beginning of the peptide, a case where the epitope is near the middle of the peptide, and a case where the epitope is at the end of the peptide. In some cases, the length of the peptide can be selected, for example, the length of the peptide can be less than 10, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 or more than 40 residues. In some embodiments, possible fragments of epitopes and flanking sequences can be selected. For example, when the epitope length is 3 and the peptide length is 15, 13 different sequences can be selected according to whether the epitope starts at the 1st-13th position of the peptide. In some cases, multiple peptide lengths can be selected. In addition, for each extracted peptide sequence (epitope+structured flanking sequence), "unstructured" peptides, wherein the sequence part related to the secondary structure is modified, to change or remove the secondary structure feature. In some embodiments, "unstructured" peptides can be generated by reorganizing one or more flanking sequences but retaining the epitope in the same position. The reorganization of the sequence can include changing the order of residues in the sequence while keeping the composition. In addition, the composition can be regarded as the count of each residue in the sequence. In some embodiments, the reorganization of the sequence can include randomly changing the order of residues in the sequence while keeping the composition. In some embodiments, "unstructured" peptides can be generated by non-randomly changing the order of residues in the sequence while keeping the sequence composition. In some embodiments, "unstructured" peptides can be generated by mixing non-natural or modified amino acid residues into the sequence. In some embodiments, "unstructured" peptides can be generated by applying one or more such methods in combination in the sequence or to a subset of one or more flanking sequences associated with the epitope. In some embodiments, the secondary structure of the "unstructured" peptide can be assessed by synthesizing the peptide and using methods such as circular dichroism spectroscopy.

[0094] Evaluation of structured peptides

[0095] Once the peptides comprising structured and unstructured epitopes have been identified as described above, an array with many repeats of each structured and unstructured peptide can be printed. The array can then be hybridized with a fluorescently labeled affinity reagent. The amount of fluorescence measured at each spot on the array can be measured, and a paired t-test of the fluorescence measurement results between the structured peptide and the unstructured peptide can be compared to determine whether there is a significant increase or decrease in binding caused by, for example, a secondary structure. If the fluorescence of the structured peptide is reduced relative to the unstructured peptide, this may indicate a certain degree of destruction of the affinity reagent binding caused by the secondary structure element. Due to the challenge of protein folding, when printed on the array, not all "structured" peptides may fold into a predicted "native" conformation. Using multiple different structured peptide sequences and redundant fragments for each epitope may help solve this challenge. Any peptide that shows a significant change in binding affinity compared to other peptides with the same epitope can be selected for further study. For example, the peptide can be evaluated by circular dichroism (CD spectroscopy) to determine whether it has taken an expected secondary structure. In some cases, all peptides can be evaluated by CD spectroscopy to determine secondary structure prior to hybridization with the affinity reagent.

[0096] Another method for assessing the influence of secondary structure on affinity reagent binding can be to use full-length protein from cell-free expression or from cell lysate.In some cases, highly structured proteins, such as prion protein, amyloid plaque or tau fibril can be isolated from tissue samples.For example, samples of different recombinant proteins with known secondary structure can be obtained, and compared with unstructured protein, disordered protein or denatured protein, affinity reagent is assessed for binding to structured protein.Methods as above can be used for any epi-position in the structured region of selected structured protein.Described protein can be combined with solid support, to promote the assessment to binding, or can be maintained in solution, and can use the method such as affinity pull-down to assess binding.

[0097] In some cases, a peptide qualification screen and a protein qualification screen can be performed to select affinity reagents with desired properties. In other cases, an affinity reagent can be selected before performing any one screen, and a screen can be performed to collect binding data about the affinity reagent.

[0098] Can be for all possible epitopes of given n-aggregates, for example for all 8000 kinds of possible 3-aggregates, carry out screening and selection steps.In some cases, can only carry out screening steps to a subset of possible n-aggregates.For example, instead of screening affinity reagents for all 8000 kinds of 3-aggregates, can select 800 kinds of 3-aggregates as required epitopes, and can not be screened for other 7200 kinds of 3-aggregates.

[0099] In some cases, an affinity reagent selected using the methods described herein can bind to about 50% of all possible proteins in a proteome. In some cases, multiple affinity reagents with different binding properties can be selected and combined together so that the affinity reagent pool binds to about half of all proteins in a proteome. In some cases, an affinity reagent pool can be used instead of a single affinity reagent.

[0100] Affinity reagents can be any reagents that specifically bind proteins or peptides in a reproducible manner. For example, affinity reagents can be antibodies, antibody fragments, aptamers or peptides. In some instances, monoclonal antibodies may be preferred. In some instances, antibody fragments such as Fab fragments may be preferred. In some instances, aptamers may be preferred. In some cases, affinity reagents can be commercially available affinity reagents, such as commercially available antibodies. In some cases, the desired affinity reagents can be selected by screening commercially available affinity reagents to identify affinity reagents with useful properties. In some cases, affinity reagents can be screened for their ability to bind to single proteins. In some cases, affinity reagents can be screened for their ability to bind to epitopes or amino acid sequences. In some cases, affinity reagent groups can be screened for their ability to jointly identify similar proteins (e.g., proteins with highly similar sequences) by differential binding. In some cases, affinity reagents can be screened for overlapping binding properties to increase binding specificity to specific proteins. The screening of affinity reagents can be carried out in a variety of different ways. An example is to screen affinity reagents for NAPPA or epitope tiling arrays. In some cases, protein-specific affinity reagents (such as commercially available antibodies or aptamers) designed to bind to protein targets can be used. In some cases, multiple protein-specific or epitope-specific affinity reagents can be mixed before performing binding measurements. For example, for each binding measurement run (pass), a new mixture of protein-specific affinity reagents can be selected, which contains a subset of available affinity reagents randomly selected from a complete group. For example, each subsequent mixture can be generated in the same random manner, and it is expected that many affinity reagents will be present in more than one mixture. In some cases, protein identification can be generated more quickly using a mixture of protein-specific affinity reagents. In some cases, such mixtures of protein-specific affinity reagents can increase the percentage of unknown proteins bound by affinity reagents in any separate operation. The mixture of affinity reagents can be composed of 1%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or more of all available affinity reagents.

[0101] Affinity reagents can have high, medium or low specificity. In some instances, affinity reagents can recognize several different epitopes. In some instances, affinity reagents can recognize epitopes present in two or more different proteins. In some instances, affinity reagents can recognize epitopes present in multiple different proteins. In some cases, the affinity reagents used in the methods of the present disclosure can have a high degree of specificity for a single epitope. In some cases, the affinity reagents used in the methods of the present disclosure can have a high degree of specificity for a single epitope containing a post-translational modification.

[0102] In some embodiments, the affinity reagent used to identify the target amino acid sequence may actually include a group of different components, which are indistinguishable from each other when used in the methods described herein. In particular, different components that can be used to identify the same target amino acid sequence can use the same detection part to identify the same target amino acid sequence. For example, an affinity reagent that binds to a trimer amino acid sequence (AAA) regardless of the flanking sequence may include a single probe that binds to the trimer AAA sequence without any influence of the flanking sequence, or a group of 400 probes, each of which binds to a different 5 amino acid epitope in the form of αAAAβ, wherein α and β can be any amino acid. In some cases of the second case, the 400 probes can be combined so that the amount of each probe is equal. In some cases of the second case, the 400 probes can be combined so that the amount of each probe can be weighted by the characteristic binding affinity of each probe, so that the probability of any given 5 amino acid epitope being bound is equal.

[0103] New affinity reagents can be generated by any method known in the art. Methods for developing affinity reagents include SELEX, phage display, and inoculation. In some instances, affinity reagents can be designed using structure-based drug design methods. Structure-based drug design (or direct drug design) utilizes knowledge of the three-dimensional structure of the epitope of interest and the binding site of the affinity reagent.

[0104] In some cases, the affinity reagent can be labeled with a nucleic acid barcode. In some instances, the nucleic acid barcode can be used to purify the affinity reagent after use. In some instances, the nucleic acid barcode can be used to sort the affinity reagent for reuse. In some cases, the affinity reagent can be labeled with a fluorophore, which can be used to sort the affinity reagent after use.

[0105] In some cases, multiple affinity reagents labeled with nucleic acid barcodes can be multiplexed and then detected using complementary nucleic acid probes. A set of multiplexed affinity reagents can be detected in a single cycle using multiple complementary nucleic acids with different detection moieties. In some cases, a set of multiplexed affinity reagents can be detected in multiple cycles using a single complementary nucleic acid conjugated to a detection moiety. In some cases, a set of multiplexed affinity reagents can be detected in multiple cycles using multiple complementary nucleic acids each conjugated to a different detection moiety. In some cases, a set of multiplexed affinity reagents can be detected in multiple cycles using multiple complementary nucleic acids each conjugated to a different set of detection moieties.

[0106] In some cases, one or more affinity reagents labeled with nucleic acid bar codes can be cross-linked to bound proteins. Once the one or more affinity reagents are cross-linked to proteins, the bar codes can be sequenced to determine the identity of the cross-linked affinity reagents. In some cases, a plurality of bound proteins can be exposed to the one or more affinity reagents. In some cases, when a plurality of bound proteins are cross-linked to the one or more affinity reagents, the bar codes associated with the bound affinity reagents can be sequenced to determine the identity of the cross-linked affinity reagents associated with each of the plurality of bound proteins.

[0107] Affinity reagent families may include one or more types of affinity reagents. For example, the methods of the present disclosure may use affinity reagent families that include one or more of antibodies, antibody fragments, Fab fragments, aptamers, peptides, and proteins.

[0108] The affinity reagent may be modified. Modifications include, but are not limited to, attachment of the detection moiety. The detection moiety may be attached directly or indirectly. For example, the detection moiety may be covalently attached directly to the affinity reagent, or may be attached via a linker, or may be attached via an affinity reaction, such as a complementary nucleic acid tag or a biotin-streptavidin pair. Attachment methods that can withstand gentle washing and elution of the affinity reagent may be preferred.

[0109] The detection moiety includes, but is not limited to, a fluorophore, a bioluminescent protein, a nucleic acid segment comprising a constant region and a barcode region, or a chemical tether for attachment to a nanoparticle such as a magnetic particle. The detection moiety may include several different fluorophores with different excitation or emission modes.

[0110] The detection moiety may be cleavable from the affinity reagent. This may allow for a step of removing the detection moiety from an affinity reagent that is no longer of interest to reduce signal contamination.

[0111] In some cases, the affinity reagent is unmodified. For example, if the affinity reagent is an antibody, the presence of the antibody can be detected by atomic force microscopy. The affinity reagent can be unmodified and can be detected, for example, by an antibody specific for one or more affinity reagents. For example, if the affinity reagent is a mouse antibody, the mouse antibody can be detected by using an anti-mouse second antibody. Alternatively, the affinity reagent can be an aptamer, which is detected by an antibody specific for the aptamer. The second antibody can be modified with a detection moiety as described above. In some cases, the presence of the second antibody can be detected by atomic force microscopy.

[0112] In some examples, the affinity reagents may contain the same modification, such as conjugated green fluorescent protein, or may contain two or more different types of modifications. For example, each affinity reagent may be conjugated to one of several different fluorescent moieties, each having a different excitation or emission wavelength. This can allow multiplexing of affinity reagents, because several different affinity reagents can be combined and / or distinguished. In one example, the first affinity reagent may be conjugated to green fluorescent protein, the second affinity reagent may be conjugated to yellow fluorescent protein, and the third affinity reagent may be conjugated to red fluorescent protein, so that the three affinity reagents can be multiplexed and identified by their fluorescence. In another example, the first, fourth and seventh affinity reagents may be conjugated to green fluorescent protein, the second, fifth and eighth affinity reagents may be conjugated to yellow fluorescent protein, and the third, sixth and ninth affinity reagents may be conjugated to red fluorescent protein; in this case, the first, second and third affinity reagents may be multiplexed together, and the second, fourth and seventh, and the third, sixth and ninth affinity reagents form two further multiplexed reactions. The number of affinity reagents that can be multiplexed together may depend on the detection moieties used to distinguish them. For example, multiplexing of affinity reagents labeled with fluorophores may be limited by the number of unique fluorophores available. For further example, multiplexing of affinity reagents labeled with nucleic acid tags may be determined by the length of the nucleic acid barcode.

[0113] The specificity of each affinity reagent can be determined before being used for determination. The binding specificity of the affinity reagent can be determined in a control experiment using a known protein. Any suitable experimental method can be used to determine the specificity of the affinity reagent. In one example, a solid support can load a known protein standard at a known position and be used to evaluate the specificity of a variety of affinity reagents. In another example, the solid support can include an experimental sample and a set of controls and standards, so that the specificity of each affinity reagent can be calculated from the combination with the controls and standards, and then used to identify the experimental sample. In some cases, an affinity reagent with unknown specificity and an affinity reagent with known specificity can be included, and the data from the affinity reagent with known specificity can be used to identify the protein, and the binding pattern of the affinity reagent with unknown specificity and the identified protein can be used to determine its binding specificity. The specificity of any single affinity reagent can also be reconfirmed by using the known binding data of other affinity reagents to evaluate which proteins the single affinity reagent binds. Therefore, by the repeated use of the affinity reagent group, the specificity of the affinity reagent can be gradually improved with each repetition. Although affinity reagents with unique specificity for specific proteins can be used, the methods described herein may not require them. Additionally, the methods may be effective for a range of specificities. In some instances, the methods described herein may be particularly effective when the affinity reagent is not specific for any particular protein, but is specific for an amino acid motif (eg, the tripeptide AAA).

[0114] In some examples, one or more affinity agents can be selected to bind to an amino acid motif of a given length, such as 2, 3, 4, 5, 6, 7, 8, 9, 10 or more than 10 amino acids. In some examples, one or more affinity agents can be selected to bind to an amino acid motif of a range of different lengths, from 2 amino acids to 40 amino acids.

[0115] In some examples, affinity reagents with high, medium, or low binding affinities can be selected. In some cases, affinity reagents with low or medium binding affinities may be preferred. In some cases, affinity reagents may have about 10 - 3 M, 10 -4 M, 10 -5 M, 10 -6 M, 10 -7 M, 10 -8 M, 10 -9 M, 10 -10 M or less. In some cases, the affinity reagent may have a dissociation constant greater than about 10 -10 M, 10 -9 M, 10 -8 M, 10 -7 M, 10-6 M, 10 -5 M, 10 -4 M, 10 -3 M, 10 -2 M or higher dissociation constant.

[0116] Some affinity reagents can be selected to bind to modified amino acid sequences, such as phosphorylated or ubiquitinated amino acid sequences. In some instances, one or more affinity reagents with broad specificity can be selected to the epitope family that can be contained by one or more proteins. In some instances, one or more affinity reagents can be combined with two or more different proteins. In some instances, one or more affinity reagents can be weakly bound to one or more targets. For example, an affinity reagent can be combined with one or more targets with less than 10%, less than 10%, less than 15%, less than 20%, less than 25%, less than 30%, less than 35% or less than 35%. In some instances, one or more affinity reagents can be combined with one or more targets moderately or strongly. For example, an affinity agent may bind to its one or more targets with greater than 35%, greater than 40%, greater than 45%, greater than 60%, greater than 65%, greater than 70%, greater than 75%, greater than 80%, greater than 85%, greater than 90%, greater than 91%, greater than 92%, greater than 93%, greater than 94%, greater than 95%, greater than 96%, greater than 97%, greater than 98%, or greater than 99%.

[0117] To compensate for weak binding, an excess of affinity reagent can be applied to the solid support. Affinity reagent can be applied in an excess of about 1:1, 2:1, 3:1, 4:1, 5:1, 6:1, 7:1, 8:1, 9:1 or 10:1 relative to the sample protein. Affinity reagent can be applied in an excess of about 1:1, 2:1, 3:1, 4:1, 5:1, 6:1, 7:1, 8:1, 9:1 or 10:1 relative to the expected occurrence of epitopes in the sample protein.

[0118] The affinity reagent may also include a magnetic component. The magnetic component may be used to manipulate some or all of the bound affinity reagents into the same imaging plane or z-stack. Manipulating some or all of the affinity reagents into the same imaging plane may improve the quality of the imaging data and reduce noise in the system.

[0119] Substrate / Solid Support

[0120] In some embodiments, the peptides can be applied to a functionalized substrate, such as a solid support, to facilitate affinity agent selection or screening procedures. Figure 3BShown is an example of a solid support with an array peptide conjugated to a solid support, which is used for affinity reagent characterization screening to determine the influence of flanking sequences on the binding of affinity reagents to epitope AAA. In some cases, the target peptide can be directly applied to the solid support. In some cases, the target peptide can be synthesized or grown on a solid support. In some cases, peptides can be synthesized on a solid support such as an array or a bead. For example, a peptide array can be made to include multiple copies of a single peptide sequence at each position of the array. Customized peptide microarrays can also be purchased commercially, such as PEPperPRINT.

[0121] Substrate can be any substrate that can form a solid support.Substrate as used herein, such as solid support, can refer to any solid surface on which peptide can be covalently or non-covalently attached.Non-limiting examples of solid support include particles, beads, slides, surfaces of device elements, films, flow cells, holes, chambers, macro fluid chambers, microfluid chambers, channels, microfluid channels or other any surfaces.Solid support surface can be flat or curved, or can have other shapes, and can be smooth or textured.Solid support surface can contain micropores.In some embodiments, solid support can be composed of glass, carbohydrates such as dextran, plastics such as polystyrene or polypropylene, polyacrylamide, latex, silicon, metals such as gold, or cellulose, and can be further modified to allow or enhance the covalent or non-covalent attachment of oligonucleotides.For example, solid support surface can be functionalized by modifying with specific functional groups such as maleic acid or succinic acid moieties, or by modifying (such as by silanization) with chemically reactive groups such as amino, sulfhydryl or acrylic acid groups to carry out derivatization. The solid support can be used for the oligonucleotides of the present invention.Suitable silane reagents include aminopropyl trimethoxysilane, aminopropyl triethoxysilane and 4-aminobutyl triethoxysilane.Solid support can be functionalized with N-hydroxysuccinimide (NHS) functional group.Glass surface can also use epoxy silane, acrylic silane or acrylamide silane, by other reactive groups such as acrylic acid or epoxy groups, derivatization.Solid support and method for oligonucleotide attachment are preferably stable for repeated combination, washing, imaging and elution steps.In some instances, solid support can be a slide glass or a flow cell.

[0122] The functional groups on the solid support can be spaced apart in an orderly or random manner. The ordered array of functional groups can be created by photolithography. The functional groups in the ordered array can be positioned so that each functional group is at least about 200nm, about 250nm, about 300nm, about 350nm, about 400nm, about 450nm, about 500nm, about 550nm, about 600nm, about 650nm, about 700nm, about 750nm, about 800nm, about 850nm, about 900nm, about 950nm or about 1000nm away from any other functional group. The randomly spaced functional groups can be provided in a concentration such that a functional group is, on average, at least about 200 nm, about 250 nm, about 300 nm, about 350 nm, about 400 nm, about 450 nm, about 500 nm, about 550 nm, about 600 nm, about 650 nm, about 700 nm, about 750 nm, about 800 nm, about 850 nm, about 900 nm, about 950 nm, or about 1000 nm away from any other functional group.

[0123] A substrate, such as a solid support, can be functionalized indirectly. For example, a solid support can be PEGylated and functional groups can be applied to all or a portion of the PEG molecules.

[0124] The substrate may comprise any material, including metal, glass, plastic, ceramic or a combination thereof. In some preferred embodiments, the solid support may be a flow cell. The flow cell may be composed of a single layer or multiple layers. For example, the flow cell may comprise a base layer (e.g., a borosilicate glass layer), a channel layer (e.g., an etched silicon layer) covering the base layer, and a cover layer or top layer. When these layers are assembled together, a closed channel may be formed, with an inlet / outlet through the cover layer at either end. The thickness of each layer may vary, but is preferably less than about 1700 μm. These layers may be composed of any suitable material known in the art, including but not limited to photosensitive glass, borosilicate glass, molten silicate, PDMS or silicon. Different layers may be composed of the same material or different materials.

[0125] In some embodiments, the flow cell may include a channel opening at the bottom of the flow cell. The flow cell may include millions of attached target conjugation sites at a position that can be visualized discretely. In some embodiments, the various flow cells used together with embodiments of the present invention may include different numbers of channels (e.g., 1 channel, 2 or more channels, 3 or more channels, 4 or more channels, 6 or more channels, 8 or more channels, 10 or more channels, 12 or more channels, 16 or more channels, or more than 16 channels). Various flow cells may include channels of different depths or widths, and the depth or width may be different between the channels in a single flow cell, or different between the channels in different flow cells. The depth and / or width of a single channel may also vary. For example, at one or more points in a channel, a channel may be less than about 50 μm deep, about 50 μm deep, less than about 100 μm deep, about 100 μm deep, about 100 μm to about 500 μm deep, about 500 μm deep or more than about 500 μm deep. The channel may have any cross-sectional shape, including but not limited to circular, semi-circular, rectangular, trapezoidal, triangular, or oval cross-sections.

[0126] The peptide can be spotted, dripped, pipetted, flowed, washed or otherwise applied to a solid support. In the case where the solid support has been functionalized with a portion such as an NHS ester, the peptide does not need to be modified. In the case where the solid support has been functionalized with an alternative portion (e.g., sulfhydryl, amine, or linker DNA), a cross-linking reagent (e.g., disuccinimidyl suberate, NHS, sulfonamide) can be used. In the case where the solid support has been functionalized with a linker DNA, the peptide of the target can be modified with a complementary DNA tag.

[0127] Photoactivatable cross-linkers can be used to direct cross-linking of samples to specific areas on a solid support. Photoactivatable cross-linkers can be used to allow multiplexing of peptide samples by attaching each sample to a known area of ​​a solid support. Photoactivatable cross-linkers can allow specific attachment of already successfully labeled peptides, for example, by detecting fluorescent tags prior to peptide cross-linking. Examples of photoactivatable cross-linkers include, but are not limited to, N-5-azido-2-nitrobenzoyloxysuccinimide, 6-(4'-azido-2'-nitrophenylamino)hexanoic acid sulfosuccinimide ester, 4,4'-diazoxide valerate succinimide ester, 4,4'-diazoxide valerate sulfosuccinimide ester, 6-(4,4'-azapentanamido)hexanoic acid succinimide ester, 6-(4,4'-azapentanamido)hexanoic acid sulfosuccinimide ester, 2-((4,4'-azapentanamido)ethyl)-1,3'-dithiopropionic acid succinimide ester, and 2-((4,4'-azapentanamido)ethyl)-1,3'-dithiopropionic acid sulfosuccinimide ester.

[0128] The peptide may be attached to the substrate via one or more residues. In some examples, the peptide may be attached via the N-terminus, the C-terminus, both termini, or via internal residues.

[0129] In some embodiments, each peptide sequence can be associated with a unique spatial address. For example, once the peptide is attached to a solid support at a spatially separated position, each peptide sequence can be assigned an indexed address, for example, by coordinates. In some instances, the grid of the pre-assigned unique spatial address can be predetermined. In some embodiments, the solid support may contain a fixed mark that is easily recognizable so that the placement of each peptide can be determined relative to the fixed mark of the solid support. In some instances, the solid support may have grid lines and / or "origin" or other reference points permanently marked on the surface. In some instances, the surface of the solid support can be permanently or semi-permanently marked to provide a reference for locating the cross-linked peptide. The patterned shape itself, such as the outer boundary of the conjugated peptide, can also be used as a reference point for determining the unique position of each spot.

[0130] Use of affinity reagents to identify proteins

[0131] Affinity reagents disclosed herein can be used to identify and quantify proteins in samples. In some instances, the method may include three aspects: 1) an addressable substrate to which proteins and / or protein fragments can be conjugated; 2) a group of affinity reagents, such as where each affinity reagent can bind to a peptide with different specificities; and 3) software, which can use a priori knowledge of the binding properties of affinity reagents, a specific binding mode of affinity reagents at each address in a substrate, and / or a combination of a database of possible sequences of proteins in a mixture (e.g., a human proteome) to infer the identity of proteins at precise spatial addresses in the substrate. In some instances, precise spatial addresses can be unique spatial addresses. In some embodiments, affinity reagents with reproducible affinity for one or more epitopes that may be present in a sample can be used together with the method. Affinity reagents that bind to different epitopes or have different affinities for epitopes in each assay may not be suitable for the method.

[0132] sample

[0133] The sample can be any biological sample containing protein. The sample can be taken from a tissue or cell, or from the environment of a tissue or cell. In some instances, the sample can be a tissue biopsy, blood, plasma, extracellular fluid, cultured cells, culture medium, discarded tissue, plant matter, synthetic protein, archaeal sample, bacterial sample and / or viral sample, fungal tissue, archaea or protozoa. In some instances, the protein is separated from its main source (cell, tissue, body fluid such as blood, environmental sample, etc.) during sample preparation. The protein may or may not be purified from its main source. In some cases, the main source is homogenized before further processing. In some cases, cells are lysed using a buffer such as RIPA buffer. Denaturing buffers may also be used at this stage. The sample can be filtered or centrifuged to remove lipids and particulate matter. The sample can also be purified to remove nucleic acids, or can be treated with RNase and DNase. The sample may contain intact protein, denatured protein, protein fragments or partially degraded proteins.

[0134] The sample can be taken from a subject with a disease or condition. The disease or condition can be an infectious disease, an immune disorder or disease, cancer, a genetic disease, a degenerative disease, a lifestyle disease, an injury, a rare disease or an age-related disease. The infectious disease can be caused by bacteria, viruses, fungi and / or parasites. Non-limiting examples of cancer include bladder cancer, lung cancer, brain cancer, melanoma, breast cancer, non-Hodgkin's lymphoma, cervical cancer, ovarian cancer, colorectal cancer, pancreatic cancer, esophageal cancer, prostate cancer, kidney cancer, skin cancer, leukemia, thyroid cancer, liver cancer and uterine cancer. Some examples of genetic diseases or conditions include but are not limited to cystic fibrosis, Charcot-Marie-Tooth disease, Huntington's disease, Peutz-Jeghers syndrome, Down syndrome, rheumatoid arthritis and Tay-Sachs disease. Non-limiting examples of lifestyle diseases include obesity, diabetes, atherosclerosis, heart disease, stroke, hypertension, cirrhosis, nephritis, cancer, chronic obstructive pulmonary disease (COPD), hearing problems and chronic back pain. Some examples of damage include but are not limited to abrasions, brain damage, bruises, burns, concussions, congestive heart failure, building damage, dislocations, flail chest, fractures, hemothorax, herniated discs, hip protuberance contusions, hypothermia, tearing, nerve pinching, pneumothorax, rib fractures, sciatica, spinal cord injuries, tendon ligament fascia injuries, traumatic brain injury and whiplash. Samples can be obtained before and / or after treating a subject with a disease or condition. Samples can be obtained before and / or after treatment. Samples can be obtained during treatment or a treatment regimen. Multiple samples can be obtained from a subject to monitor the effect of treatment over time. Samples can be obtained from a subject known or suspected of having an infectious disease without available diagnostic antibodies.

[0135] Samples may be taken from subjects suspected of having a disease or condition. Samples may be taken from subjects experiencing unexplained symptoms such as fatigue, nausea, weight loss, aches and pains, weakness, or memory loss. Samples may be taken from subjects with symptoms of a clear cause. Samples may be taken from subjects at risk of developing a disease or condition due to factors such as family history, age, environmental exposure, lifestyle risk factors, or the presence of other known risk factors.

[0136] The sample may be taken from an embryo, fetus, or pregnant woman. In some instances, the sample may include proteins isolated from maternal plasma. In some instances, the sample may include proteins isolated from circulating fetal cells in maternal blood. In some cases, the sample may include proteins isolated from amniotic fluid.

[0137] Protein can be treated to remove modifications that may interfere with epitope binding. For example, protein can be treated with glycosidase to remove post-translational glycosylation. Protein can be treated with a reducing agent to reduce the disulfide bonds in the protein. Protein can be treated with phosphatase to remove phosphate groups. Other non-limiting examples of post-translational modifications that can be removed include acetic acid groups, amide groups, methyl groups, lipids, ubiquitin, myristoylation, palmitoylation, isoprenylation or isoprenylation (e.g., farnesol and geranylgeraniol), farnesylation, geranylgeranylation, glycosylphosphatidylinositolization, lipidation, flavin moiety attachment, phosphopantetheinization and retinyl Schiff base formation. Samples can also be treated to retain post-translational protein modifications. In some instances, phosphatase inhibitors can be added to the sample. In some instances, an oxidant can be added to protect disulfide bonds.

[0138] Next, the protein can be denatured completely or partially. In some embodiments, the protein can be denatured completely. The protein can be denatured by applying external stress such as detergents, strong acids or bases, concentrated inorganic salts, organic solvents (e.g., alcohol or chloroform), radiation or heat. The protein can be denatured by adding a denaturing buffer. The protein can also be precipitated, lyophilized and suspended in a denaturing buffer. The protein can be denatured by heating. Denaturation methods that are unlikely to cause chemical modification to the protein may be preferred.

[0139] Before or after conjugation, the protein of the sample can be processed to produce shorter polypeptides. The remaining protein can be partially digested with enzymes such as proteinase K to generate fragments, or it can be kept intact. In a further example, the protein can be exposed to proteases such as trypsin. Other examples of proteases can include serine proteases, cysteine ​​proteases, threonine proteases, aspartic proteases, glutamic proteases, metalloproteases and asparagine peptide lyases.

[0140] In some cases, it may be useful to remove very large and small proteins (e.g., titin), which may be removed by filtration or other appropriate methods. In some instances, very large proteins may include proteins of more than 400 kD, 450 kD, 500 kD, 600 kD, 650 kD, 700 kD, 750 kD, 800 kD, or 850 kD. In some instances, very large proteins may include proteins of more than about 8,000 amino acids, about 8,500 amino acids, about 9,000 amino acids, about 9,500 amino acids, about 10,000 amino acids, about 10,500 amino acids, about 11,000 amino acids, or about 15,000 amino acids. In some instances, small proteins may include proteins of less than about 10 kD, 9 kD, 8 kD, 7 kD, 6 kD, 5 kD, 4 kD, 3 kD, 2 kD, or 1 kD. In some examples, small proteins can include proteins of less than about 50 amino acids, 45 amino acids, 40 amino acids, 35 amino acids, or about 30 amino acids. Very large or small proteins can be removed by size exclusion chromatography. Very large proteins can be separated by size exclusion chromatography, treated with a protease to produce medium-sized polypeptides, and recombined with the medium-sized proteins of the sample.

[0141] In some cases, proteins can be sorted by size. In some cases, proteins can be sorted by sorting them into micropores. In some cases, proteins can be sorted by sorting them into nanopores. In some cases, proteins can be sorted by passing them through gels such as SDS-PAGE gels. In some cases, proteins can be sorted by other size-dependent fractionation methods. In some cases, proteins can be separated based on charge. In some cases, proteins can be separated based on hydrophobicity. In some cases, proteins can be separated based on other physical properties. In some cases, proteins can be separated under denaturing conditions. In some cases, proteins can be separated under non-denaturing conditions. In some cases, different fractions of fractionated proteins can be placed on different regions of a substrate. In some cases, different parts of the separated proteins can be placed on different regions of a substrate. In some cases, a protein sample can be separated in an SDS-PAGE gel and transferred from the SDS-PAGE gel to a substrate so that the protein is sorted continuously by size. In some cases, a protein sample can be sorted into three fractions based on size, and these three fractions can be applied to the first, second, and third regions of the substrate, respectively. In some cases, the proteins used in the systems and methods described herein can be sorted. In some cases, the proteins used in the systems and methods described herein can not be sorted.

[0142] For example, proteins can be labeled with identifiable tags to allow multiplexing of samples. Some non-limiting examples of identifiable tags include: fluorophores, or base linkers of nucleic acid barcoding. The fluorophores used can include fluorescent proteins such as GFP, YFP, RFP, eGFP, mCherry, tdtomato, FITC, Alexa Fluor 350, Alexa Fluor 405, Alexa Fluor 488, Alexa Fluor 532, Alexa Fluor 546, Alexa Fluor 555, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 647, Alexa Fluor 680, Alexa Fluor 750, Pacific Blue, coumarin, BODIPY FL, Pacific Green, Oregon Green, Cy3, Cy5, Pacific Orange, TRITC, Texas Red, R-phycoerythrin, allophycocyanin, or other fluorophores known in the art.

[0143] Any number of protein samples can be multiplexed. For example, a multiplexed reaction can contain proteins from 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, about 20, about 25, about 30, about 35, about 40, about 45, about 50, about 55, about 60, about 65, about 70, about 75, about 80, about 85, about 90, about 95, about 100 or more than 100 original samples. Identifiable tags can provide a means of interrogating the sample of origin of each protein, or can direct proteins from different samples to be segregated to different regions on a solid support.

[0144] Base

[0145] In some embodiments, the protein is then applied to a functionalized substrate, thereby chemically attaching the protein to the substrate. In some cases, the protein can be attached to the substrate by biotin attachment. In some cases, the protein can be attached to the substrate by nucleic acid attachment. In some embodiments, the protein can be applied to an intermediate substance, which is then attached to the substrate. In some cases, the protein can be conjugated to beads (e.g., gold beads), and the beads can then be captured on a surface (e.g., a thiolated surface). In some cases, one protein can be conjugated to each bead. In some cases, the protein can be conjugated to beads (e.g., one protein per bead), and the beads can be captured on a surface (e.g., in a micropore and / or nanopore).

[0146] Substrate can be any substrate capable of forming a solid support. Substrate or solid substrate as used herein can refer to any solid surface to which protein can be covalently or non-covalently attached. Non-limiting examples of solid substrates include particles, beads, slides, surfaces of device elements, films, flow cells, holes, chambers, macro fluid chambers, which are flat or curved, or can have other shapes, and can be smooth or textured. In some cases, the substrate surface may include micropores. In some cases, the substrate surface may include nanopores. In some cases, the substrate surface may include one or more micropores combined with one or more nanopores. In some embodiments, the substrate can be composed of glass, carbohydrates such as dextran, plastics such as polystyrene or polypropylene, polyacrylamide, latex, silicon, metals such as gold, or cellulose, and can be further modified to allow or enhance the covalent or non-covalent attachment of oligonucleotides. For example, the substrate surface can be functionalized by modifying with specific functional groups such as maleic acid or succinic acid moieties, or by modifying with chemically reactive groups such as amino, sulfhydryl or acrylic acid groups (e.g., by silanization) to carry out derivatization. The substrate can be functionalized with N-hydroxysuccinimide (NHS) functional group. The glass surface can also be derivatized with epoxy silane, acrylic silane or acrylamide silane, such as acrylic acid or epoxy groups. The substrate and method for oligonucleotide attachment are preferably stable for repeated combination, washing, imaging and elution steps. In some instances, the substrate can be a slide or a flow cell.

[0147] The ordered array of functional groups can be created by, for example, photolithography, dip-pen nanolithography, nanoimprint etching, nanosphere lithography, nanoball lithography, nanopillar arrays, nanowire etching, scanning probe etching, thermochemical etching, thermal scanning probe etching, local oxidation nanolithography, molecular self-assembly, template etching, or electron beam etching. The functional groups in the ordered array can be positioned so that each functional group is less than 200 nanometers (nm) away from any other functional group, or about 200nm, about 225nm, about 250nm, about 275nm, about 300nm, about 325nm, about 350nm, about 375nm, about 400nm, about 425nm, about 450nm, about 475nm, about 500nm, about 525nm, about 550nm, about 5 75nm, about 600nm, about 625nm, about 650nm, about 675nm, about 700nm, about 725nm, about 750nm, about 775nm, about 800nm, about 825nm, about 850nm, about 875nm, about 900nm, about 925nm, about 950nm, about 975nm, about 1000nm, about 1025nm, about 1050nm, about 1075nm, about 1 100nm, about 1125nm, about 1150nm, about 1175nm, about 1200nm, about 1225nm, about 1250nm, about 1275nm, about 1300nm, about 1325nm, about 1350nm, about 1375nm, about 1400nm, about 1425nm, about 1450nm, about 1475nm, about 1500nm, about 1525nm, about 1550nm, about 1575 nm, about 1600 nm, about 1625 nm, about 1650 nm, about 1675 nm, about 1700 nm, about 1725 nm, about 1750 nm, about 1775 nm, about 1800 nm, about 1825 nm, about 1850 nm, about 1875 nm, about 1900 nm, about 1925 nm, about 1950 nm, about 1975 nm, about 2000 nm or more than 2000 nm. The randomly spaced functional groups can be provided in a concentration such that a functional group is, on average, at least about 50 nm, about 100 nm, about 150 nm, about 200 nm, about 250 nm, about 300 nm, about 350 nm, about 400 nm, about 450 nm, about 500 nm, about 550 nm, about 600 nm, about 650 nm, about 700 nm, about 750 nm, about 800 nm, about 850 nm, about 900 nm, about 950 nm, about 1000 nm, or more than 100 nm away from any other functional group.

[0148] The substrate can be functionalized indirectly. For example, the substrate can be PEGylated and the functional group can be applied to all or a group of PEG molecules. In addition, as described above, in some cases, beads (e.g., gold beads) can be conjugated and then the beads can be captured on a surface (e.g., a thiolated surface). In some cases, each bead can be conjugated with a protein. In some cases, proteins can be conjugated with beads (e.g., one protein per bead) and the beads can be captured on a surface (e.g., in a micropore and / or nanopore).

[0149] The substrate can be functionalized using a technique suitable for microscale or nanoscale structures (e.g., ordered structures such as micropores, nanopores, micropillars, single molecule arrays, nanospheres, nanopillars, or nanowires). In some cases, the substrate can have micropores of different sizes. In some cases, the micropores can be 1 micron (μm), can be about 2 μm, about 3 μm, about 4 μm, about 5 μm, about 6 μm, about 7 μm, about 8 μm, about 9 μm, about 10 μm, about 15 μm, about 20 μm, about 25 μm, about 30 μm, about 35 μm, about 40 μm, about 45 μm, about 50 μm, about 55 μm, about 60 μm, about 65 μm, about 70 μm, about 75 μm, about 80 μm, about 85 μm, about 90 μm, about 95 μm, about 100 μm, about 105 μm, about 110 μm, about 120 μm, about 130 μm, about 140 μm, about 150 μm, about 160 μm, about 170 μm, about 180 μm, about 190 μm, about 200 μm, about 210 μm, about 220 μm, about 230 μm, about 240 μm, about 250 μm, about 260 μm, about 270 μm, about 280 μm, about 290 μm, about 300 μm, about 310 μm, about 310 μm, about 320 μm, about 330 μm, about 340 μm, about 350 μm, about 110μm, about 115μm, about 120μm, about 125μm, about 130μm, about 135μm, about 140μm, about 145μm, about 150μm, about 155μm, about 160μm, about 165μm, about 170μm, about 175μm, about 180μm, about 185μm, about 190μm, about 195μm, about 200μm, about 205μm, about 210μm, about 215μm, about 220μm, about 225μm, about 230μm, about 235μm, about 240μm , about 245μm, about 250μm, about 255μm, about 260μm, about 265μm, about 270μm, about 275μm, about 280μm, about 285μm, about 290μm, about 295μm, about 300μm, about 305μm, about 310μm, about 315μm, about 320μm, about 325μm, about 330μm, about 335μm, about 340μm, about 345μm, about 350μm, about 355μm, about 360μm, about 365μm, about 370μm, about 375μm In some cases, the substrate may have micropores ranging in size from 5 μm to 500 μm. In some cases, the substrate may have micropores ranging in size from about 5 μm to about 500 μm. In some cases, the substrate can have micropores ranging in size from 10 μm to 100 μm. In some cases, the substrate can have micropores ranging in size from about 10 μm to about 100 μm.In some cases, the substrate may have a series of micropores of different sizes so that proteins of different sizes can be sorted into micropores of different sizes. In some cases, the micropores in the substrate can be distributed by size (e.g., larger micropores are distributed in the first region, and smaller micropores are distributed in the second region). In some cases, the substrate may have about ten different sized micropores. In some cases, the substrate may have about 20 different sizes, about 25 different sizes, about 30 different sizes, about 35 different sizes, about 40 different sizes, about 45 different sizes, about 50 different sizes, about 55 different sizes, about 60 different sizes, about 65 different sizes, about 70 different sizes, about 75 different sizes, about 80 different sizes, about 85 different sizes, about 90 different sizes, about 95 different sizes, about 100 different sizes or more than 100 different sizes of micropores.

[0150] In some cases, the substrate may have nanopores of different sizes. In some cases, the nanopore may be about 100 nanometers (nm), about 150nm, about 200nm, about 250nm, about 300nm, about 350nm, about 400nm, about 450nm, about 500nm, about 550nm, about 600nm, about 650nm, about 700nm, about 750nm, about 800nm, about 850nm, about 900nm, about 950nm, or 950nm to 1 micron. In some cases, the substrate may have nanopores with a size range of 100nm to 1 micron. In some cases, the substrate may have nanopores with a size range of 100nm to 500nm. In some cases, the substrate may have a series of nanopores of different sizes so that proteins of different sizes can be sorted into nanopores of different sizes. In some cases, the nanopores in the substrate can be distributed by size (e.g., larger nanopores are distributed in the first region, and smaller nanopores are distributed in the second region). In some cases, the substrate can have nanopores of about ten different sizes. In some cases, the substrate can have nanopores of about 20 different sizes or more than 30 different sizes.

[0151] In some cases, the substrate may have a series of nanopores and / or micropores of different sizes so that proteins of different sizes can be sorted into nanopores and / or micropores of different sizes. In some cases, the nanopores and / or micropores in the substrate can be distributed by size (e.g., larger micropores are distributed in the first region, and smaller nanopores are distributed in the second region). In some cases, the substrate may have about ten nanopores and / or micropores of different sizes. In some cases, the substrate may have about 20 different sizes, about 25 different sizes, about 30 different sizes, about 35 different sizes, about 40 different sizes, about 45 different sizes, about 50 different sizes, about 55 different sizes, about 60 different sizes, about 65 different sizes, about 70 different sizes, about 75 different sizes, about 80 different sizes, about 85 different sizes, about 90 different sizes, about 95 different sizes, about 100 different sizes or more than 100 different sizes of nanopores and / or micropores.

[0152] The substrate may comprise any material, including metal, glass, plastic, ceramic or a combination thereof. In some preferred embodiments, the solid substrate may be a flow cell. The flow cell may be composed of a single layer or multiple layers. For example, the flow cell may comprise a base layer (e.g., a borosilicate glass layer), a channel layer (e.g., an etched silicon layer) covering the base layer, and a cover layer or top layer. When these layers are assembled together, a closed channel may be formed, with an inlet / outlet through the cover layer at either end. The thickness of each layer may vary, but is preferably less than about 1700 μm. These layers may be composed of any suitable material known in the art, including but not limited to photosensitive glass, borosilicate glass, molten silicate, PDMS or silicon. Different layers may be composed of the same material or different materials.

[0153] In some embodiments, the flow cell may include a channel opening on the bottom of the flow cell. The flow cell may include millions of attached target conjugation sites at a position that can be discretely visualized. In some embodiments, the various flow cells used together with the embodiments of the present invention may include different numbers of channels (e.g., 1 channel, 2 or more channels, 3 or more channels, 4 or more channels, 6 or more channels, 8 or more channels, 10 or more channels, 12 or more channels, 16 or more channels, or more than 16 channels). Various flow cells may include channels of different depths or widths, and the depth or width may be different between the channels in a single flow cell, or different between the channels in different flow cells. The depth and / or width of a single channel may also vary. For example, at one or more points in a channel, a channel may be less than about 50 μm deep, about 50 μm deep, less than about 100 μm deep, about 100 μm deep, about 100 μm to about 500 μm deep, about 500 μm deep or more than about 500 μm deep. The channel may have any cross-sectional shape, including but not limited to circular, semi-circular, rectangular, trapezoidal, triangular, or oval cross-sections.

[0154] The protein can be spotted, dripped, pipetted, flowed, washed or otherwise applied to the substrate. In the case where the substrate has been functionalized with a portion such as an NHS ester, it is not necessary to modify the protein. In the case where the substrate has been functionalized with an alternative portion (e.g., sulfhydryl, amine, or linker nucleic acid), a cross-linking reagent (e.g., disuccinimidyl suberate, NHS, sulfonamide) can be used. In the case where the substrate has been functionalized with a linker nucleic acid, the protein of the sample can be modified with a complementary nucleic acid tag.

[0155] In some cases, protein can be conjugated with nucleic acid. Using nucleic acid, nucleic acid nanoballs can be formed so that protein is connected to the nucleic acid nanoballs. When the nucleic acid nanoballs are attached to a substrate, the protein attached to the nucleic acid is attached to the substrate through the nucleic acid nanoballs. DNA nanoballs can be attached (e.g., by adsorption or by conjugation) to a substrate. The substrate can have an amine-functionalized surface to which the nucleic acid nanoballs can be attached.

[0156] In some cases, a nucleic acid nanoball with a functionally active end (e.g., maleimide, NHS-ester, etc.) can be formed. Then, a protein can be conjugated to the nanoball, thereby connecting the protein to the nucleic acid nanoball. When the nucleic acid nanoball is attached to a substrate, the protein attached to the nucleic acid is attached to the substrate through the nucleic acid nanoball. The DNA nanoball can be attached (e.g., by adsorption or by conjugation) to the substrate. The substrate can have an amine-functionalized surface to which the nucleic acid nanoball can be attached.

[0157] Photoactivatable crosslinkers can be used to direct crosslinking of samples to specific areas on a substrate. Photoactivatable crosslinkers can be used to allow multiplexing of protein samples by attaching each sample to a known area of ​​a substrate. Photoactivatable crosslinkers can allow specific attachment of already successfully labeled proteins, for example, by detecting fluorescent tags prior to protein crosslinking. Examples of photoactivatable cross-linkers include, but are not limited to, N-5-azido-2-nitrobenzoyloxysuccinimide, 6-(4'-azido-2'-nitrophenylamino)hexanoic acid sulfosuccinimide ester, 4,4'-diazoxide valerate succinimide ester, 4,4'-diazoxide valerate sulfosuccinimide ester, 6-(4,4'-azapentanamido)hexanoic acid succinimide ester, 6-(4,4'-azapentanamido)hexanoic acid sulfosuccinimide ester, 2-((4,4'-azapentanamido)ethyl)-1,3'-dithiopropionic acid succinimide ester, and 2-((4,4'-azapentanamido)ethyl)-1,3'-dithiopropionic acid sulfosuccinimide ester.

[0158] Samples can also be multiplexed by limiting the binding of each sample to discrete areas on the substrate. For example, the substrate can be organized into lanes. Another approach for multiplexing is to repeatedly apply the sample to the entire substrate, performing a protein detection step after each sample application using a nonspecific protein binding reagent or dye. In some cases, examples of dyes can include fluorescent protein gel stains, such as Ruby, Orange, Red, Tangerine and Coomassie TM Fluor Orange.

[0159] By tracking the positions of all proteins after each addition of a sample, it is possible to determine the stage at which each position on the substrate initially contained a protein, and thus determine from which sample the protein originated. This method also allows the saturation of the substrate to be determined after each application of a sample, and allows for maximization of protein binding on the substrate. For example, if only 30% of the functionalized positions are occupied by protein after the first application of a sample, a second application of the same sample or an application of a different sample can be performed.

[0160] The polypeptide may be attached to the substrate via one or more residues. In some examples, the polypeptide may be attached via the N-terminus, the C-terminus, both termini or via internal residues.

[0161] In addition to permanent crosslinkers, the use of photocleavable linkers may also be appropriate for some applications, and doing so enables selective extraction of proteins from the substrate after analysis. In some cases, a photocleavable crosslinker can be used for several different multiplexed samples. In some cases, a photocleavable crosslinker can be used for one or more samples in a multiplexed reaction. In some cases, a multiplexed reaction can include a control sample crosslinked to the substrate via a permanent crosslinker and an experimental sample crosslinked to the substrate via a photocleavable crosslinker.

[0162] Each conjugated protein can be separated from each other in space, so that each conjugated protein is optically resolvable. Therefore, proteins can be individually labeled with unique spatial addresses. In some embodiments, this can be achieved by conjugating using low concentrations of protein and low density attachment sites on the substrate so that each protein molecule is separated from each other in space. In the example of using a photoactivatable cross-linking agent, a light pattern can be used to attach the protein to a predetermined position.

[0163] In some methods, purified bulk proteins can be conjugated to substrate and processed using the methods described herein to identify purified proteins. Bulk proteins can include purified proteins collected together. In some instances, bulk proteins can be conjugated spatially at a position separated from each other conjugated protein or bulk proteins so that each conjugated protein or bulk proteins is optically resolvable. Therefore, protein or bulk proteins can be individually labeled with a unique spatial address. In some embodiments, this can be conjugated using low-density attachment sites on low-concentration proteins and substrates so that each protein molecule is spatially separated from each other to achieve. In the example of using a photoactivated cross-linking agent, light patterns can be used so that one or more proteins are attached to a predetermined position.

[0164] In some embodiments, each protein may be associated with a unique spatial address. For example, once the proteins are attached to the substrate at spatially separated locations, each protein may be assigned an indexed address, for example, by coordinates. In some instances, a grid of pre-assigned unique spatial addresses may be predetermined. In some embodiments, the substrate may contain easily recognizable fixed marks so that the placement of each protein can be determined relative to the fixed marks of the substrate. In some instances, the substrate may have grid lines and / or "origins" or other reference points permanently marked on the surface. In some instances, the surface of the substrate may be permanently or semi-permanently marked to provide a reference for locating the cross-linked proteins. The patterned shape itself, such as the outer boundary of a conjugated polypeptide, may also be used as a reference point for determining the unique position of each spot.

[0165] The substrate may also contain conjugated protein standards and controls. The conjugated protein standards and controls may be peptides or proteins of known sequence conjugated to known positions. In some instances, the conjugated protein standards and controls may serve as internal controls in the assay. Proteins may be applied to the substrate from purified protein stocks, or may be synthesized on the substrate by processes such as nucleic acid programmable protein arrays (NAPPA).

[0166] In some examples, the substrate may include fluorescent standards. These fluorescent standards can be used to calibrate the fluorescence signal intensity between assays. These fluorescent standards can also be used to correlate the fluorescence signal intensity with the number of fluorophores present in a region. The fluorescent standards can include some or all of the different types of fluorophores used in the assay.

[0167] Affinity reagents

[0168] Once the substrate is conjugated to the proteins from the sample, a multi-affinity reagent measurement can be performed. The measurement procedures described herein can employ a variety of affinity reagents described herein.

[0169] Combined measurement

[0170] For a set of modified affinity reagents and conjugated substrates, the affinity reagents can be repeatedly applied to the substrate. Each measurement cycle consists of several stages. In the first stage, the affinity reagents are applied to the substrate where they can adsorb to the conjugated protein.

[0171] Next, the substrate can be gently washed to remove non-specific binding. This washing step can be performed under conditions that will not elute the affinity reagent bound to the immobilized protein. Some examples of buffers that can be used for this step include phosphate buffered saline, Tris buffered saline, phosphate buffered saline containing Tween 20, and Tris buffered saline containing Tween 20.

[0172] After adsorption, the binding address of each modified affinity reagent is determined, for example by measuring a fluorophore that has been conjugated directly to the affinity reagent, or by measuring a fluorophore that has been conjugated to a complementary nucleic acid to a nucleic acid strand conjugated to the affinity reagent. The detection method depends on the choice of detection moiety. Fluorophores and bioluminescent moieties can be detected optically, and in some cases a second detection reagent is required. The unique address of each immobilized protein on the substrate can be determined prior to the binding measurement, or a list of addresses containing immobilized proteins can be generated by the binding measurement.

[0173] Next, the affinity reagent can be desorbed by a more stringent wash. This wash step can remove some or all of the affinity reagent from the immobilized substrate. In some cases, the affinity reagent can be selected to have a low to medium binding affinity for easy removal. The used affinity reagent can be recaptured for reuse or discarded. In instances where an affinity reagent with a cleavable detection portion is used, the detection portion can be cut and removed at this stage. After the stringent wash, in some instances, any remaining fluorescence can be quenched and an even more stringent wash can be applied to remove the residual affinity reagent. Residue / contamination can be detected by re-imaging the substrate before applying the next affinity reagent. Contamination can also be detected by continuously monitoring the reproducing signal in the image. This ends an analysis cycle.

[0174] In some embodiments, a fluorescently labeled affinity reagent can be quenched by exposure to prolonged intense light at an activation wavelength. Quenching of the fluorescent tag can replace a wash step to remove the affinity reagent. In some embodiments, it may be necessary to cycle n fluorophores to distinguish which signals came from the previous n-1 cycles.

[0175] The cycle continues for each affinity reagent or multiplex thereof. The result of the measurement phase is a very large table listing the binding coordinates of each affinity reagent, or the affinity reagent bound at each coordinate position, see e.g. Fig.10 .

[0176] analyze

[0177] The final step of protein identification may include a software tool to determine the most likely identity of each protein at each coordinate of the substrate from information about which affinity reagent binds to the coordinate. The software may utilize information about the binding properties of each affinity reagent. For example, whether a given affinity reagent preferentially binds to a protein containing the tripeptide epitope AAA. Taking into account the binding properties of each affinity reagent, the database of proteins in the sample and the list of binding coordinates, and the information about the binding mode, the software tool assigns a possible identity to each coordinate and a confidence level for the identity. In the extreme case of an exact 1-1 correspondence between an affinity reagent and a protein, this can be accomplished by a simple table lookup. However, in more complex binding situations, this can be accomplished by solving appropriate satisfaction problems. In highly complex binding situations, an expectation maximization approach can be employed.

[0178] The software can also utilize a list of some or all positions that each affinity reagent does not bind here, and use this information about the absence of epitopes to determine the proteins that exist. The software can also utilize information about which affinity reagents bind and do not bind to each address. Therefore, the software will use information about which epitopes exist and which epitopes do not exist. The software can include a database. The database can include the sequences of some or all known proteins in the species from which the sample is obtained. For example, if the known sample is derived from humans, a database with the sequences of some or all human proteins can be used. If the species of the sample is unknown, a database of some or all protein sequences can be used. The database can also include the sequences of some or all known protein variants and mutant proteins, and the sequences of some or all possible proteins that can be produced by DNA frameshift mutations. The database can also include the sequences of possible truncated proteins that may be produced by premature termination codons or by degradation.

[0179] The software may include one or more algorithms, such as machine learning, deep learning, statistical learning, supervised learning, unsupervised learning, clustering, expectation maximization, maximum likelihood estimation, Bayesian reasoning, linear regression, logistic regression, binary classification, multinomial classification, or other pattern recognition algorithms. For example, the software may execute one or more algorithms to analyze the following information (e.g., as input to one or more algorithms): (i) the binding properties of each affinity agent, (ii) a database of proteins in a sample, (iii) a list of binding coordinates, and / or (iv) the binding pattern of the affinity agent to the protein, so as to generate or assign (e.g., as output of one or more algorithms) (a) a possible identity of each coordinate, and / or (b) a confidence in the identity (e.g., a confidence level and / or confidence interval). Examples of machine learning algorithms may include support vector machines (SVMs), neural networks, convolutional neural networks (CNNs), deep neural networks, cascaded neural networks, k-nearest neighbor (k-NN) classification, random forests (RFs), and other types of classification and regression trees (CARTs).

[0180] The software can be trained by performing the method of the present disclosure on a substrate where the identity of the protein at each address on the substrate is predetermined. For example, the software can be trained using a nucleic acid programmable protein array or an epitope tiling array as a training data set.

[0181] The raw data from the binding assay may include a series of images from which the binding measurements of the labeled affinity reagent and the unknown protein are derived. These binding measurements can then be used to infer the identity of the determined protein. In some embodiments, the identity of each determined protein can be inferred using these binding measurements. In some embodiments, these binding measurements can be used to provide characteristics related to the identity of the determined protein. In some embodiments, protein inferences can be determined based on the following three types of information: 1) the binding measurement results of the affinity reagent and a specific protein; 2) a list of candidate protein sequences; and 3) information from which a semi-accurate prediction of the expected degree of binding of each affinity reagent to each candidate protein can be derived. The method discussed herein provides a statistical model of the affinity reagent, from which the probability of the affinity reagent binding to any protein can be calculated given the primary sequence of the protein. In some embodiments, a series of binding measurements of each of the affinity reagents and the protein set can be used to train the statistical model. When there is no direct empirical measurement available, these types of methods can be used to predict the measurement results and / or characteristics of affinity reagent-protein binding. These methods may be used, for example, when the number of proteins in a protein collection is such that determining empirical measurements would be prohibitive due to time or cost.

[0182] First, binding measurements of affinity reagents for a variety of polypeptides or proteins can be obtained. These binding measurements can be used to train a statistical model from which the binding of an affinity reagent to any arbitrary protein can be predicted, provided its primary sequence is provided. Methods for obtaining such measurements include, but are not limited to, ELISA assays, flow cytometry assays, binding to protein arrays, binding to peptide arrays, and surface plasmon resonance assays. Protein binding can be modeled as a linear combination of the counts of each of the J epitopes in the protein sequence and the fractional contribution of the binding of the affinity reagent to the epitope to the protein level binding measurements. This can be expressed as:

[0183]

[0184] in:

[0185] M pr = Binding measurement results of affinity reagent and protein pr

[0186] c t,pr = count of epitope t in protein pr

[0187] β t = fractional binding contribution of affinity reagent to epitope t

[0188] The results of the measurements of affinity reagents against various proteins form a linear system of equations:

[0189]

[0190] in:

[0191] is a length N column vector containing the observed binding measurements for each protein C is an N x J matrix of epitope counts, where each column is the count of a specific epitope in each measured protein

[0192] is a column vector of length J of the fractional binding of the reagent to each possible epitope

[0193] ∈ is a scalar constant corrected for background binding or noise floor

[0194] It is not difficult to extend the model to include non-standard amino acids or to simulate different types of epitopes (e.g., single residues, dimers, trimers, 4-mers, 5-mers). In the case of binding measurements to known proteins, and C are known variables, and The values ​​of and ∈ can be obtained by linear regression or related methods. In particular, non-negative least squares and non-negative least absolute shrinkage and selection operator (LASSO) regression may be well suited for this problem. Non-negative least squares will The solution of is restricted to non-negative values, and non-negative LASSO regression further imposes a sparsity constraint. LASSO regression is particularly effective when the system of equations is underdetermined, that is, when the number of unique proteins measured is less than the number of unique epitopes. Once the The value of c can be generated by counting each epitope in the protein sequence and evaluating the sum To generate binding affinity measurement predictions for any protein.

[0195] This is just one example of how a simulated protein binds to an affinity agent. These data can be used to train any model that predicts protein binding affinity from the primary sequence of a protein or a derivative of the primary sequence (e.g., amino acid composition, trimer count, predicted three-dimensional structure). Another example can include the use of a nonlinear model, such as an exponential function that associates trimer affinity with observed protein binding measurements, or a multiplication model in which trimer level affinities are multiplied to generate protein level affinities. As another example, a neural network can be used to predict protein binding affinity from a protein sequence or trimer composition. In addition, a support vector regression model can be used to predict protein binding affinity from a protein sequence or trimer composition.

[0196] Determine the characteristics of your sample

[0197] Once the decoding is complete, the possible identities of the proteins conjugated to each address are defined. Therefore, their abundance in the mixture can be estimated by counting the observations. Therefore, a list of each protein present in the mixture can be compiled, along with the number of observations of that protein.

[0198] In addition, if a photocleavable linker or other form of specifically cleavable linker is used to attach the protein to the substrate, the specific target protein can be released from the substrate and collected for further study. For example, a specific protein can be identified and eluted for further study. The method disclosed herein can also be used as a way to purify and / or separate the desired protein from a mixture. In some cases, the method may be able to purify and / or separate specific isoforms or post-translationally modified proteins. In samples where a complete list of possible proteins and related sequences is not available, the method may be able to distinguish different proteins in different groups of proteins, which can then be eluted for further study. For example, for highly complex samples containing many unknown proteins, such as intestinal microbiome samples, the methods described herein can be used to fractionate the samples before mass spectrometry. In some cases, once the identity of the protein can be determined, the proteins can be eluted from the substrate. Removing the protein from the substrate when the protein is identified allows subsequent rounds of affinity reagent binding to continue for proteins whose identities have not yet been determined, and can reduce the background noise and off-target signals of the remaining rounds. In some instances, one or more affinity reagents specific for a particular protein can be used as a first round to identify high abundance proteins in a blood sample, such as serum albumin or immunoglobulins, and these high abundance proteins can then be removed early in the process. In some cases, a subset of proteins on the substrate can be removed after each round of affinity reagent binding, or after each second, third, fourth, fifth, sixth, seventh, eighth, ninth, tenth, fifteenth, twentieth, or more than twentieth round of affinity reagent binding. The signal-to-noise ratio may increase after each round of protein elution.

[0199] In some cases, unidentified proteins can be grouped or clustered based on their binding patterns. For example, in some cases, the protein present in the sample may not be represented in the sequence database. Unidentified proteins can be clustered into groups based on their binding patterns with affinity probes, with the goal that each group contains a group of unknown proteins with the same sequence in the sample. The amount of protein in each group can be estimated and included in a quantitative analysis, which includes but is not limited to quantitative analysis of the difference between a healthy state and a disease state, longitudinal analysis, or biomarker discovery. In some cases, unidentified groups can be selectively removed from substrates for identification by mass spectrometry. In other cases, unidentified groups can be identified by conducting further binding affinity measurement experiments specifically designed to generate confident identifications.

[0200] In some cases, additional sample can be added to the substrate after a protein or group of proteins has been removed. For example, serum albumin is a high abundance protein in serum that can account for about half of the protein in a sample, and removing serum albumin after a first round of affinity reagent binding can allow for the addition of additional blood sample to the substrate. In some embodiments, it may be preferred to remove high abundance proteins prior to immobilizing the sample on a substrate, such as by immunoprecipitation or affinity column purification.

[0201] Protein modifications can be identified using the methods of the present disclosure. For example, post-translational modifications can be identified by repeated cycles of detection using modification-specific detection reagents and interspersed with enzyme treatments (e.g., phosphatase treatments). Affinity reagents specific for different modifications can be used to determine whether such modifications are present on the immobilized protein. The method also allows for quantification of the number of instances of each protein with and without a given modification.

[0202] Mutations in proteins can be detected by matching the inconsistency between the binding pattern of a sample protein and the predicted protein identity. For example, an immobilized protein or polypeptide on a substrate that matches the affinity reagent binding profile of a known protein may have amino acid substitutions in addition to the binding of an affinity reagent. Since affinity reagents may have overlapping epitopes, an immobilized protein may have several mismatches with the predicted affinity binding pattern despite having a single amino acid substitution. DNA mutations that cause premature stop codon frameshifts can also be detected.

[0203] The number of affinity reagents required may be less than the total number of epitopes present in the sample. For example, if the affinity reagent is selected so that each affinity reagent recognizes a unique tripeptide epitope, the total set of affinity reagents for all possible epitopes in the recognition sample is 20x 20x 20=8000. However, the method of the present disclosure may only require about 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1500, 2000, 2500, 3000, 3500, 4000, 4500, 5000, 5500 or 6000 of these affinity reagents. In some cases, the method may only require less than about 500, 1000, 2500, 3000, 3500, 4000, 4500, 5000, 5500 or 6000 kinds of affinity reagents. The results of the simulation show the percentage of known human proteins that can be identified given a set of x affinity reagents specific for unique amino acid 3-mers as a function of the binding efficiency of each affinity reagent. The simulation results show that 98% of human proteins can be uniquely identified with 8000 3-mer affinity reagents with a binding probability of 10%.

[0204] The methods of the present disclosure may be highly accurate. The methods of the present disclosure may be able to identify each protein with an accuracy of about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 98.5%, 99%, 99.5%, 99.9% or greater than 99.9%.

[0205] The methods disclosed herein may be able to predict the identity of each protein with a confidence of about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 98.5%, 99%, 99.5%, 99.9% or greater than 99.9%. The confidence may be different for different proteins in a sample. For example, the confidence in identifying a protein with a very unique sequence may be higher than the confidence in identifying a protein that is highly similar to other proteins. In some cases, a protein may be identified as part of a protein family with a high confidence, yet the exact identity of the protein may be predicted with a lower confidence. In some cases, the confidence in predicting extremely large or extremely small proteins may be lower than the confidence in predicting more medium-sized proteins.

[0206] In some cases, a protein may be identified with high confidence as being part of a protein family, however the exact identity of the protein may be predicted with lower confidence. For example, a protein containing a single amino acid variation may be difficult to distinguish with high confidence from the canonical form of the protein. In this case, neither the canonical sequence nor the form containing the single amino acid variation may be identified with high confidence, but a high confidence assessment may be made that the unknown protein is part of a group of proteins containing both sequences. Similar situations may occur where a protein may have multiple related isoforms with similar sequences.

[0207] The methods of the present disclosure may be able to identify some or all of the proteins in a given sample. The methods of the present disclosure may be able to identify about 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 98.5%, 99%, 99.5%, 99.9% or more than 99.9% of the proteins in a sample.

[0208] The methods of the present disclosure may be capable of rapidly identifying proteins in a sample. The methods of the present disclosure may be capable of identifying more than about 100, about 1000, about 5000, about 10000, about 20,000, about 30,000, about 40,000, about 50,000, about 100,000, 1,000,000, about 10,000,000, about 100,000,000, about 1,000,000,000, about 10,000,000,000, about 100,000,000,000, about 1,000,000,000,000, about 1,000,000,000,000,000 proteins per flow cell per day. The methods of the present disclosure may be capable of identifying more than about 10 10 , 10 11 , 10 12 , 10 13 , 10 14 , 10 15 , 10 16 , 10 17 or more than about 10 17 The method disclosed herein may be able to identify approximately 10 proteins per flow cell per day. 10 -10 12 , 10 11 -10 14 , 10 12 -10 16 or 10 13 -10 17The methods of the present disclosure may be capable of identifying about 10 pg, about 20 pg, about 30 pg, about 40 pg, about 50 pg, about 60 pg, about 70 pg, about 80 pg, about 90 pg, about 100 pg, about 300 pg, about 300 pg, about 400 pg, about 500 pg, about 600 pg, about 700 pg, about 800 pg, about 900 pg, about 1 ng, about 2 ng, about 3 ng, about 4 ng, about 5 ng, about 6 ng, about 7 ng, about 8 ng, about 8 ng, about 10 ng, about 10 ng, about 20 ng, about 30 ng, about 40 ng, about 50 ng, about 60 ng, about 70 ng, about 80 ng, about 90 ng, about 100 n In some embodiments, the present invention relates to an aqueous solution of at least about 1% protein in an amount of at least about 5% protein in an amount of at least about 2% protein. In some embodiments, the aqueous solution of at least about 1% protein in an amount of at least about 5% protein in an amount of at least about 2% protein is at least about 1% protein.

[0209] The methods of the present disclosure can be used to assess the proteome after an experimental treatment. The methods of the present disclosure can be used to assess the effects of a therapeutic intervention.

[0210] The methods disclosed herein can be used for biomarker discovery. Monitoring proteome expression in subjects with and without a disease can identify biomarkers. Monitoring proteome expression in subjects before the onset of a disease or in subjects at risk of developing a disease can identify biomarkers that predict risk. Evaluating the proteome expression of a subject can indicate the health status of the subject or the risk of developing certain diseases or conditions. The methods disclosed herein can be used to evaluate therapies, or to distinguish responders from non-responders to a drug / therapy. The methods disclosed herein can be particularly useful in personalized medicine.

[0211] The methods disclosed herein can be used to diagnose disease. Different diseases or disease stages can be associated with different sets of protein expression. Different sets of protein expression can be associated with different treatment outcomes for each given treatment. The proteomic expression data of a subject can be used to diagnose the subject and / or select the most appropriate therapy.

[0212] The method of the present disclosure can be used to identify the individual or species from which the sample comes. For example, the method of the present disclosure can be used to determine whether the sample is actually from the claimed species or source. For samples with abundant protein but limited nucleic acid, the method described herein may be superior to the method based on PCR. For example, the method of the present disclosure can be used to identify the origin of a honey sample. For further examples, the method of the present disclosure can be used to assess food safety and food quality control.

[0213] The method disclosed herein can be used to identify any single protein molecule from a pool of protein molecules using an affinity reagent that is less than the number of possible proteins. For example, the method can use a group of affinity reagents to identify unidentified single protein molecules from a pool of n possible proteins with a certainty higher than a threshold amount, wherein the number of affinity reagents in the group is m, and wherein m is less than n. Unidentified proteins can be known proteins corresponding to known proteins and gene sequences, or can be unknown proteins that do not have known proteins or gene sequences. In the case of unknown proteins, the method can identify the characteristics of unknown proteins, thereby identifying the presence and amount of the unknown protein, rather than the amino acid sequence. The method disclosed herein can be used to select a group of m kinds of affinity reagents, which can identify unidentified proteins selected from a pool of n possible proteins. The method disclosed herein can also uniquely identify and quantify n kinds of proteins in a protein mixture using m kinds of binding reagents, and wherein each protein is identified by the unique binding profile of a subset of m kinds of binding reagents. In addition, m can be less than about one-half, one-third, one-quarter, one-fifth, one-sixth, one-seventh, one-tenth, one-twentieth, one-fiftieth, or one-hundredth of n. For further example, the present disclosure can be used to select a panel of less than about 100, 200, 300, 400, 500, 600, 700, 800, 900, 1000, 1500, 2000, 2500, 3000, 3500, or 4000 affinity agents such that the panel of affinity agents uniquely identifies at least about 100, 500, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000, 10,000, 12,000, 14,000, or 5000 affinity agents. , 1,000, 2,000, 3,000, 000, 4,000, 000, or 5,000,000 different proteins.

[0214] The methods of the present disclosure may be capable of identifying a majority of proteins in a proteome. The methods of the present disclosure may be capable of identifying a majority of proteins in a mammal, bird, fish, amphibian, reptile, vertebrate, invertebrate, plant, fungus, bacteria, or archaea proteome. The methods of the present disclosure may be capable of identifying greater than about 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% of the proteins in a proteome.

[0215] Example

[0216] Example 1: Protein identification using antibodies that bind unique 3-mer peptides

[0217] A computational experiment was performed to determine the relationship between the percentage coverage of the group of all epitopes in the proteome and the percentage of the proteome that can be identified using the method of the present disclosure. For this experiment, a group of all 3-mer amino acid epitopes was selected. Protein modifications were not considered. Since there are 20 naturally occurring amino acids, the total group of all 3-mer epitopes is 20x20x20=8000 possible epitopes. For this simulation, x was set to the number of epitopes screened in the experiment, and for each x value from 1 to 8000, a group of x epitopes was randomly selected, and the percentage of the proteome that could be identified was calculated.

[0218] Example 2: Protein identification using antibodies that bind unique 3-mer peptides

[0219] Further computational experiments were performed to determine the effect of the number of affinity reagents on identifiability and coverage. Data series were calculated for a range of affinity reagent pool sizes to show the percentage (y-axis) of the protein group that can be identified for each possible coverage of the protein, and the results are shown in Table 1. For example, if 20% of these 3-mer amino acid epitopes are bound (they may be sufficient or insufficient to identify the protein), a protein with 100 amino acids has 98 3-mer amino acid epitope "landing sites". For an affinity reagent pool with 250 3-mer specific affinity reagents, if 20% of the landing sites of each protein are bound, only about 7% of the protein group can be identified. For an affinity reagent pool with 8000 affinity reagents, when 20% of the landing sites are bound, about 98% of the protein group can be identified.

[0220] Table 1: Effect of the number of 3-mer decoding (d-code) probes on proteome discriminability-coverage

[0221] 8000 7000 6000 5000 4000 3000 2000 1000 500 250 1.00％ 0.1825 0.135 0.0845 0.072 0.042 0.0125 0.0035 0 0 0 2.00％ 0.492 0.41 0.3515 0.26 0.156 0.0985 0.037 0.0035 0.0005 0 3.00％ 0.677 0.614 0.55 0.455 0.344 0.2175 0.0985 0.015 0.0005 0 4.00％ 0.786 0.745 0.676 0.604 0.472 0.334 0.176 0.029 0.003 0 5.00％ 0.843 0.811 0.765 0.7005 0.61 0.4765 0.269 0.054 0.0075 0 6.00％ 0.9025 0.852 0.809 0.7645 0.6815 0.569 0.3485 0.092 0.012 0.0015 7.00％ 0.9005 0.877 0.8435 0.81 0.7285 0.626 0.4345 0.1395 0.022 0.0025 8.00％ 0.9275 0.9025 0.8875 0.835 0.782 0.678 0.491 0.192 0.034 0.002 9.00％ 0.9415 0.923 0.898 0.8725 0.814 0.728 0.5495 0.221 0.0415 0.0065 10.00％ 0.9575 0.941 0.919 0.8835 0.8535 0.751 0.601 0.261 0.0715 0.007 12.00％ 0.9635 0.957 0.946 0.913 0.8825 0.81 0.663 0.3445 0.0955 0.0145 15.00％ 0.978 0.969 0.962 0.9505 0.9185 0.8605 0.7675 0.443 0.1585 0.0295 17.00％ 0.981 0.9765 0.9645 0.9575 0.927 0.884 0.8005 0.503 0.1915 0.0435 20.00％ 0.9885 0.986 0.9725 0.9635 0.9575 0.9105 0.847 0.584 0.2525 0.0775 25.00％ 0.99 0.9865 0.9785 0.9745 0.966 0.9445 0.8915 0.6955 0.357 0.1165 30.00％ 0.9865 0.9895 0.985 0.9825 0.973 0.9625 0.9245 0.76 0.4355 0.1665 50.00％ 0.9915 0.9915 0.9935 0.9895 0.9855 0.978 0.967 0.89 0.691 0.374

[0222] Example 3: Irradiation of protein molecules conjugated to a substrate

[0223] The fluorescent protein sample phycoerythrin was directly conjugated to NHS-ester coated coverslips in an incubation chamber at 4 degrees for 4 hours. The fluorescent protein sample was then imaged using a Leica DMi8 with a Hamamatsu orca flash 4.0 camera using a 300ms exposure. The resulting image showed that the protein signal could be clearly distinguished from the background noise.

[0224] The second protein sample, green fluorescent protein, was denatured and directly conjugated to NHS-ester coated coverslips in an incubation chamber at 4 degrees for 4 hours. Initial imaging showed no baseline residual fluorescence, indicating that the green fluorescent protein was completely denatured. The protein was then incubated with an anti-peptide antibody attached with Alexa-Fluor 647. The anti-peptide antibody was then rinsed with 0.1% Tween-20. It was then imaged with TIRF using a Nikon Eclipse Ti with an Andor NEO sCMOS camera.

[0225] Example 4: Identification of proteins

[0226] A proteome of four possible proteins, green fluorescent protein, RNASE1, LTF, and GSTM1, is depicted. In this embodiment, a single molecule of an unknown protein from the proteome is conjugated to a position on a substrate. The unknown protein is interrogated sequentially by a set of nine different affinity reagents. Each of these nine different affinity reagents recognizes a different amino acid trimer [AAA, AAC, AAD, AEV, GDG, QSA, LAD, TRK, DGD] and is labeled with a fluorescent dye. It is determined that the unknown protein is bound by affinity reagents DGD, AEV, LAD, GDG, and QSA. Sequence analysis of the four proteins of the proteome shows that only GFP contains all five of these three amino acid motifs. Therefore, it is determined that the single molecule of the unknown protein is the GFP protein.

[0227] Embodiment 5: Implementation Embodiment

[0228] Considered in this example is the measurement of the binding of affinity reagents to a nucleic acid programmable protein array (NAPPA), which contains 720 unique proteins with 3-5 replicates on the array. In some embodiments, these replicates can be dispersed across the array. Binding is assessed by hybridizing a fluorescently labeled affinity reagent to the array and measuring the fluorescence observed at each spot on the array. In this example, the identity of each protein at each spot is known. Therefore, fluorescence can correspond to the protein identity.

[0229] In this example, the protein binding measurement (fluorescence) is simulated as a linear combination of the counts for each of the 8000 possible trimers in the protein sequence and the fractional contribution to the observed fluorescence attributed to the binding of the affinity agent to that trimer. That is:

[0230]

[0231] in:

[0232] F pr = Fluorescence of protein pr

[0233] c t,pr = the count of trimer t in protein pr

[0234] β t = fraction of the binding of the affinity reagent to the trimer t The measurements of the affinity reagent against various proteins form a linear system of equations:

[0235]

[0236] in:

[0237] is a length N column vector containing the fluorescence observed for each protein C is an N x 8000 matrix of trimer counts, where each column is the count of a specific trimer in each measured protein

[0238] is a length 8000 column vector of fractional fluorescence for each possible trimer ∈ is a scalar constant corrected for background binding or background noise

[0239] The models described herein can be extended to include non-standard amino acids or to simulate n-mers of varying lengths. In the case of known NAPPA or similar binding measurements, and C are known variables, and The values ​​of and ∈ can be obtained by linear regression or related methods. In particular, non-negative least squares and non-negative least absolute shrinkage and selection operator (LASSO) regression may be well suited for this problem. Non-negative least squares will The solution of is restricted to non-negative values, and non-negative LASSO regression further imposes a sparsity constraint. LASSO regression can be particularly effective when the system of equations is underdetermined, that is, when the number of unique proteins measured is less than the number of unique trimers (8,000 in this example).

[0240] Binding to the above NAPPA array was simulated for affinity reagents with the following relative affinities:

[0241] LAS: 0.75

[0242] GEL:0.05

[0243] YIC: 0.001

[0244] STK: 0.001

[0245] Binding to all other trimers was assigned an affinity of 0.00001.

[0246] Binding to each protein on the NAPPA array was simulated by calculating expected binding using the described statistical model and the above binding affinities with Gaussian noise applied (with a standard deviation of 5% of the expected binding affinity).

[0247] From these simulated data, the binding affinity of the reagents was learned from the protein binding measurements using LASSO (LASSO sparsity parameter α was set to 1e-3). The learned affinity is:

[0248] LAS 0.750034

[0249] GEL 0.040796

[0250] ASW 0.012120

[0251] SGD 0.005001

[0252] QPH 0.003975

[0253] LGC 0.003788

[0254] KGR 0.002475

[0255] VPS 0.002416

[0256] EIK 0.002305

[0257] SES 0.002149

[0258] RIS 0.001235

[0259] SSE 0.001067

[0260] STS 0.000953

[0261] APP 0.000889

[0262] HSD 0.000785

[0263] KSQ 0.000735

[0264] VQK 0.000637

[0265] PPT 0.000538

[0266] GCS 0.000417

[0267] RPR 0.000106

[0268] QPQ 0.000051

[0269] ACS 0.000026

[0270] The affinity to all other trimers is zero.

[0271] This example demonstrates that this approach is able to establish reasonable estimates of affinity reagent binding, albeit with a tolerable amount of noise.

[0272] Example 6: Selection of KW affinity reagent

[0273] Affinity agent selection screening was performed to select aptamers that bind to any of four different target epitopes - KW, LFQ, IRN and EGE. Aptamers that bind to the targets were selected from the aptamer library using array-based surface plasmon resonance imaging selection. Preliminary validation of aptamers that bind to the target KW was performed. Figure 6 Binding of the top five hits to the KW target as measured by surface plasmon resonance is shown. Figure 7 Binding of the best anti-KW aptamer compared to a negative control peptide is shown.

[0274] In order to determine the Figure 6 To investigate the epitope specificity of the top-ranked anti-KW aptamer, the aptamer was fluorescently labeled and applied to the peptide array. Based on the binding of the labeled aptamer to the peptide array, the epitopes recognized by the aptamer were determined to be KW and KR, with relatively similar affinities for both. Combining the data and preliminary motif analysis in Figure 8 Shown in.

[0275] Example 7: High-throughput sequencing of fluorescent ligand interaction profiles

[0276] High-throughput sequencing fluorescent ligand interaction profiling was performed by obtaining an aptamer library and incorporating adapter sequences at both ends of the aptamers in the library. The library of aptamers with adapters was applied to a flow cell containing immobilized oligonucleotide primers and Fig.10 Cluster amplification was performed as shown. After cluster amplification, the clusters (each cluster represents a single aptamer from the aptamer library) were sequenced using a reversible dye terminator. The clusters were then denatured to remove sequencing reagents, washed, and folded into their native aptamer conformation. Fluorescently labeled peptide targets were applied to the clusters at four different concentrations and imaged to show clusters bound to the fluorescently labeled peptide targets. Fig.11Representative images of a flow cell with a bound fluorescently labeled peptide target and two different concentrations are shown. The fluorescence data were combined with the sequencing data to provide affinity measurements for each aptamer in the aptamer library represented on the flow cell. Fig.12 The sequences and binding affinities to the fluorescently labeled peptide target (LFQ) of several different aptamers are shown.

[0277] Example 8: Peptide synthesis and verification

[0278] Material:

[0279] All reagents and solvents were of peptide synthesis grade or the highest purity available. Amino acid derivatives were obtained from Aapptech (Louisville, KY USA), while all solvents used for peptide synthesis, SPE and RP±HPLC were obtained from Acros Organics, USA.

[0280] Peptide Synthesis:

[0281] Peptides were synthesized on a MultiPep RSi synthesizer (Intavis, Germany) using standard Fmoc / tBu chemistry. Amino acid derivatives were activated with N,N,N′,N′-tetramethyl-O-(1H-benzotriazol-1-yl)uronium hexafluorophosphate (HBTU) (0.5 M). Peptides were synthesized in a 96-vessel microreaction module using standard synthesis protocols. Rink Amide Resin (Intavis, Germany) was loaded to 2 micromoles per vessel at an amine content of 0.53 mmol / g.

[0282] Add the synthesized peptide using the sequence:

[0283]

[0284]

[0285] All amino acid derivatives were used as 0.5 M solutions in DMF except Fmoc-lysine(biotin) (0.3 M).

[0286] N-methylpiperidine was used as a 2M solution in DMF and acetic anhydride was used as a 5% solution (0.529 M) in DMF. Each cycle of coupling was continued for 30 min and each building block was repeated 2 times. The amino acid was used in a 10-fold excess relative to the amine to ensure complete coupling. The peptide was cleaved with a solution of 5% water, 2.5% triisopropylsilane in trifluoroacetic acid (TFA) at room temperature for 2 h, precipitated with tert-butyl methyl ether at -20°C, and washed 3 times with cold tert-butyl methyl ether (200 ul). The dried peptide was dissolved in 1 mL of 0.1% TFA in water (500 microliters) and stored at -20°C.

[0287] Solid Phase Extraction

[0288] Using Multi 96 vacuum manifold (Waters, USA), crude peptides were purified on Sep-PakC18 Multi 96-well plates (WAT054955) containing 40mg C18 adsorbent in each hole. The eluate was collected in a 2mL collection rack (WAT058956). In a typical scheme, the column or plate was washed with 1ml MeOH and regulated with 0.1% TFA aqueous solution before loading the peptide sample. Peptides were loaded by slowly passing the solution through the stationary phase without applying vacuum to complete near-quantitative peptide binding. The solid phase was washed with 0.1% TFA aqueous solution to remove salt and other polar impurities, and then eluted with 1mL 70% acetonitrile aqueous solution to elute the bound peptide. Solvent was evaporated overnight using Centrivap, and the precipitate was redissolved in 500 microliters of deionized water. Using Tecan Spark plate reader, water was used as blank, and absorption spectra in the range of 200 to 1000nm were collected to measure concentration. The concentration of the peptide was calculated using the absorbance at 214 nm and the extinction coefficient.The concentration of the peptide was measured using a UV-transparent 96-well plate (ThermoFisher 8404).

[0289] Mass spectrometry:

[0290] The matrix solution was purchased from Agilent Technologies (Catalog No. G2037A). The samples were prepared by co-spotting 1 μl of peptide stock solution and 1 μl of matrix onto a MALDI sample plate and allowing the sample to dry under ambient conditions. Mass spectra were recorded on a MALDI-TOF mass spectrometer AB SCIEX 5800TOF / TOF. Representative mass spectra are shown in Fig. 9 Shown in.

[0291] Computer control system

[0292] The present disclosure provides a computer control system that is programmed to implement the method of the present disclosure. A computer system 1401 is provided that is programmed or otherwise configured to characterize and identify biopolymers, such as proteins. The computer system 1401 can adjust the following aspects: evaluate and analyze the sample of the present disclosure, for example, observe the signal at the unique spatial address of the substrate; determine the presence or absence of an identifiable tag connected to the biopolymer portion at the unique spatial address based on the observed signal; evaluate the determined identifiable tag relative to the biopolymer sequence database to determine the characteristics of the biopolymer portion. The computer system 1401 can be an electronic device of the user, or a computer system located remotely relative to the electronic device. The electronic device can be a mobile electronic device.

[0293] The computer system 1401 includes a central processing unit (CPU, also referred to herein as a "processor" and "computer processor") 1405, which may be a single-core or multi-core processor, or multiple processors for parallel processing. The computer system 1401 also includes a memory or memory location 1410 (e.g., random access memory, read-only memory, flash memory), an electronic storage unit 1415 (e.g., a hard disk), a communication interface 1420 (e.g., a network adapter) for communicating with one or more other systems, and peripherals 1425, such as cache memory, other memory, data storage, and / or an electronic display adapter. The memory 1410, storage unit 1415, interface 1420, and peripherals 1425 communicate with the CPU 1405 via a communication bus (solid lines) such as a motherboard. The storage unit 1415 may be a data storage unit (or data repository) for storing data. The computer system 1401 may be operably coupled to a computer network ("network") 1430 by means of the communication interface 1420. The network 1430 may be the Internet, the Internet and / or an extranet, or an intranet and / or an extranet in communication with the Internet. The network 1430 is in some cases a telecommunications and / or data network. The network 1430 may include one or more computer servers capable of implementing distributed computing such as cloud computing. In some cases, the network 1430 may implement a peer-to-peer network with the aid of the computer system 1401, which may enable a device coupled to the computer system 1401 to function as a client or server.

[0294] The CPU 1405 may execute a series of machine-readable instructions that may be embodied in a program or software. The instructions may be stored in a memory location such as the memory 1410. The instructions may be directed to the CPU 1405, which may then program or otherwise configure the CPU 1405 to implement the methods of the present disclosure. Examples of operations performed by the CPU 1405 may include fetching, decoding, executing, and writing back.

[0295] CPU 1405 may be part of a circuit such as an integrated circuit. One or more other components in system 1401 may be included in the circuit. In some cases, the circuit is an application specific integrated circuit (ASIC).

[0296] Storage unit 1415 can store files, such as drivers, libraries, and saved programs. Storage unit 1415 can store user data, such as user preferences and user programs. In some cases, computer system 1401 may include one or more additional data storage units located external to computer system 1401, such as located on a remote server that communicates with computer system 1401 via an intranet or the Internet.

[0297] Computer system 1401 can communicate with one or more remote computer systems via network 1430. For example, computer system 1401 can communicate with a user's remote computer system. Examples of remote computer systems include personal computers (e.g., portable PCs), tablets or tablet PCs (e.g., iPad, Galaxy Tab), phones, smartphones (e.g. iPhone, Android-enabled devices, ) or a personal digital assistant. A user can access the computer system 1401 through the network 1430.

[0298] The methods described herein may be implemented by machine (e.g., computer processor) executable code stored in an electronic storage location (e.g., memory 1410 or electronic storage unit 1415) of computer system 1401. The machine executable code or machine readable code may be provided in the form of software. During use, the code may be executed by processor 1405. In some cases, the code may be retrieved from storage unit 1415 and stored in memory 1410 for access by processor 1405. In some cases, electronic storage unit 1415 may not be included, and machine executable instructions may be stored in memory 1410.

[0299] The code may be precompiled and configured for use with a machine having a processor suitable for executing the code, or it may be compiled during operation. The code may be provided in a programming language that may be selected to enable the code to be executed in a precompiled or real-time compiled manner.

[0300] Aspects of the systems and methods provided herein, such as computer system 501, can be embodied in programming. Aspects of the technology can be considered as "products" or "articles", which are generally in the form of machine (or processor) executable code and / or associated data carried or embodied in a certain type of machine-readable medium. Machine executable code can be stored in an electronic storage unit such as a memory (e.g., read-only memory, random access memory, flash memory) or a hard disk. "Storage" type media can include any or all tangible memories of a computer, processor, etc., or its related modules, such as various semiconductor memories, tape drives, disk drives, etc., which can provide non-temporary storage for software programming at any time. All or part of the software can communicate from time to time via the Internet or various other telecommunications networks. For example, such communication can enable software to be loaded from one computer or processor to another computer or processor, for example, from a management server or host to a computer platform of an application server. Therefore, another type of medium that can carry software elements includes light waves, radio waves, and electromagnetic waves, such as across physical interfaces between local devices, through wired and optical landline networks, and via various air links. Physical elements that carry such waves, such as wired or wireless links, optical links, etc., can also be considered to be media that carry the software. Unless limited to non-transitory, tangible "storage" media, terms such as computer or machine "readable media" as used herein refer to any medium that participates in providing instructions to a processor for execution.

[0301] Thus, machine-readable media, such as computer executable code, can take many forms, including but not limited to tangible storage media, carrier media, or physical transmission media. Non-volatile storage media include, for example, optical or magnetic disks, such as any storage device in any computer, etc., such as can be used to implement the databases shown in the accompanying drawings, etc. Volatile storage media include dynamic memory, such as the main memory of such a computer platform. Tangible transmission media include coaxial cables, copper wire, and optical fiber, including the wires that make up the bus within a computer system. Carrier transmission media can take the form of electrical or electromagnetic signals, or acoustic or light waves, such as those generated during radio frequency (RF) and infrared (IR) data communications. Thus, common forms of computer readable media include, for example, a floppy disk, a flexible disk, a hard disk, a magnetic tape, any other magnetic medium, a CD-ROM, a DVD or DVD-ROM, any other optical medium, a punched card stock tape, any other physical storage medium having a pattern of holes, a RAM, a ROM, a PROM and an EPROM, a FLASH-EPROM, any other memory chip or cartridge, a carrier wave that transmits data or instructions, a cable or link that transmits such a carrier wave, or any other medium from which a computer can read programming code and / or data. Many of these forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to a processor for execution.

[0302] Computer system 1401 may include or be in communication with an electronic display 1435 that includes a user interface (UI) 1440. Examples of UIs include, but are not limited to, graphical user interfaces (GUIs) and web-based user interfaces.

[0303] The methods and systems of the present disclosure may be implemented by one or more algorithms. The algorithms may be implemented by software when executed by the central processing unit 1405. The algorithms may, for example, determine the properties and / or identity of a biopolymer moiety, such as a protein moiety. For example, an algorithm may be used to determine the most likely identity of a candidate biopolymer moiety, such as a candidate protein moiety.

[0304] In some embodiments, aptamers or peptide aptamers that recognize short epitopes present in many different proteins can be referred to as digital aptamers or digital peptide aptamers. One aspect of the present invention provides a set of digital aptamers or digital peptide aptamers, wherein the set comprises at least about 15 digital aptamers or digital peptide aptamers, wherein each of the 15 digital aptamers or digital peptide aptamers has been characterized as specifically binding to a different epitope consisting of 3 or 4 or 5 consecutive amino acids, and wherein each digital aptamer or digital peptide aptamer recognizes a variety of unique and different proteins comprising the same epitope bound by the digital aptamer or digital peptide aptamer. In some embodiments, the set of digital aptamers or digital peptide aptamers comprises 100 digital aptamers or digital peptide aptamers that bind to epitopes consisting of 3 consecutive amino acids. In some embodiments, the set of digital aptamers or digital peptide aptamers further comprises 100 digital aptamers that bind to epitopes consisting of 4 consecutive amino acids. In some embodiments, the set of digital aptamers or digital peptide aptamers further comprises 100 digital aptamers or digital peptide aptamers that bind to an epitope consisting of 5 consecutive amino acids. In some cases, the digital affinity agent can be an antibody, an aptamer, a peptide aptamer, a peptide or a Fab fragment.

[0305] In some embodiments, the set of digital aptamers comprises at least about 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400, 500, 600, 700, 800, 900 or 1000 digital aptamers. In some embodiments, the set of digital aptamers comprises at least 1000 digital aptamers that bind to an epitope consisting of 4 consecutive amino acids. In some embodiments, the set of digital aptamers further comprises at least 100 digital aptamers that bind to an epitope consisting of 5 consecutive amino acids. The set of digital aptamers further comprises at least 100 digital aptamers that bind to an epitope consisting of 3 consecutive amino acids. In some embodiments, the set of digital aptamers is immobilized on a surface. In some embodiments, the surface is an array.

[0306] In another aspect, the present invention provides a method for generating a protein binding profile of a sample comprising a plurality of different proteins, the method comprising: contacting the sample with a set of digital aptamers under conditions permissive for binding, wherein the set of digital aptamers comprises at least about 15 digital aptamers, wherein each of the 15 digital aptamers has been characterized as specifically binding to a different epitope consisting of 3 or 4 or 5 consecutive amino acids, and each digital aptamer recognizes a plurality of unique and different proteins comprising the same epitope bound by the digital aptamer; optionally removing unbound proteins; and detecting binding of proteins to the digital aptamers, thereby generating a protein binding profile of the sample.

[0307] In some embodiments, the method further comprises a step of treating the sample with a protein cleaving agent prior to step (a) of contacting the sample with the set of digital aptamers under conditions that allow binding.

[0308] In another aspect, the present invention includes a library of protein binding profiles of two or more different samples, each of the samples comprising a plurality of proteins, the method comprising: contacting the sample with a set of digital aptamers under conditions permissive for binding, wherein the set of digital aptamers comprises at least about 15 digital aptamers, wherein each of the 15 digital aptamers has been characterized as specifically binding to a different epitope consisting of 3 or 4 or 5 consecutive amino acids, and each digital aptamer recognizes a plurality of unique and different proteins comprising the same epitope bound by the digital aptamer; optionally removing unbound proteins; generating a protein binding profile of the tested sample by detecting binding of proteins to the digital aptamers, thereby generating a protein binding profile; and repeating the above steps with at least two samples.

[0309] In some embodiments, the method further comprises the step of treating the sample with a protein cleaving agent prior to the step of contacting the sample with the set of digital aptamers under conditions that allow binding.

[0310] In another aspect, the present invention includes a method for characterizing a test sample, comprising: contacting the test sample with a set of digital aptamers under conditions that allow binding, wherein the set of digital aptamers comprises at least about 15 digital aptamers, wherein each of the 15 digital aptamers has been characterized as specifically binding to a different epitope consisting of 3 or 4 or 5 consecutive amino acids, and each digital aptamer recognizes multiple unique and different proteins comprising the same epitope bound by the digital aptamer; optionally removing unbound proteins; generating a protein binding profile of the test sample by detecting binding of proteins to the digital aptamers; and comparing the generated protein binding profile of the test sample with the protein binding profile of a reference sample to characterize the test sample.

[0311] In another aspect, the present invention includes a method for determining whether bacteria, viruses or cells are present in a test sample, the method comprising: contacting the test sample with a set of digital aptamers under conditions that allow binding, wherein the set of digital aptamers comprises at least about 15 digital aptamers, wherein each of the 15 digital aptamers has been characterized as specifically binding to a different epitope consisting of 3 or 4 or 5 consecutive amino acids, and each digital aptamer recognizes multiple unique and different proteins comprising the same epitope bound by the digital aptamer; optionally removing unbound proteins; generating a protein binding profile for the test sample by detecting binding of proteins to the digital aptamers, thereby generating a protein binding profile; and comparing the protein binding profile of the test sample with the protein binding profile of a reference sample, thereby determining whether bacteria, viruses or cells are present in the test sample by comparison.

[0312] In another aspect, the present invention includes a method for identifying a test protein in a sample, the method comprising: contacting a sample containing or suspected of containing a test protein with a set of digital aptamers comprising at least about 15 digital aptamers, wherein each of the 15 digital aptamers has been characterized as specifically binding to a different epitope consisting of 3 or 4 or 5 consecutive amino acids, and each digital aptamer recognizes multiple unique and different proteins containing the same epitope bound by the digital aptamer; and determining the identity of the test protein by detecting binding of the test protein to the set of digital aptamers, wherein at least about six digital aptamers bind to the test protein; and wherein the presence of binding indicates the presence of at least about six epitopes in the test protein, wherein the identities of the at least about six epitopes are used to identify the test protein.

[0313] Although preferred embodiments of the present invention have been shown and described herein, it is obvious to those skilled in the art that these embodiments are provided by way of example only. It is not intended to limit the present invention with the specific examples provided in this specification. Although the present invention has been described with reference to the above description, it is not intended that the description and illustration of the embodiments herein are interpreted in a restrictive sense. Those skilled in the art will now think of many variations, changes and substitutions without departing from the present invention. In addition, it should be understood that all aspects of the present invention are not limited to the specific description, configuration or relative proportions set forth herein, which depend on a variety of conditions and variables. It should be understood that various alternatives to the embodiments of the present invention described herein can be adopted in implementing the present invention. It is therefore contemplated that the present invention should also cover any such substitutions, changes, variations or equivalents. It is intended to define the scope of the present invention with the appended claims, and thus cover methods and structures and their equivalents within the scope of these claims.

Claims

1. A method for selecting an affinity reagent that binds a peptide epitope in a variety of sequence contexts, the method comprising: Obtaining a peptide library comprising a plurality of sequence environments of at least 400 sequence environments, each peptide having a sequence αXβ, wherein X is a peptide epitope, wherein X is a sequence of 2, 3, or 4 amino acids, wherein flanking sequences α and β each comprise at least one amino acid, and wherein flanking sequences α and β differ among the plurality of sequence environments in the peptide library; as well as A library of affinity agents is screened against the peptide library to select an affinity agent that binds to peptide epitope X in at least 10% of a plurality of the at least 400 sequence contexts.

2. The method according to claim 1, wherein α and β each consist of one amino acid. The method of claim 1 , wherein at least one of α and β further comprises a linker.

4. The method of claim 1, wherein at least one of α and β further comprises a modification selected from the group consisting of an N-terminal modification, a C-terminal modification, a positive charge, a negative charge, a hydrophobic group, and a sugar.

5. The method according to claim 1, wherein α and β each consist of one to three amino acids.

6. The method of claim 1, wherein the selected affinity agent is selected from the group consisting of an antibody, a peptide and a Fab fragment. The method according to claim 1 , wherein the selected affinity agent is an aptamer.

8. The method of claim 1, wherein the selected affinity agent is a peptide aptamer.

9. The method according to claim 1, further comprising: The binding of the selected affinity agent to a plurality of peptides having the same length as the peptide epitope X is determined by screening the selected affinity agent against the plurality of peptides.

10. The method of claim 9, wherein the plurality of peptides having the same length as peptide epitope X represents at least 90% of all possible variations of peptides having the same length as peptide epitope X.

11. The method according to claim 1, further comprising: The binding of the selected affinity agent is characterized by screening the selected affinity agent against a plurality of peptides, wherein each peptide in the plurality of peptides comprises a peptide epitope X and one or more flanking residues.

12. The method according to claim 9, further comprising: The binding of the selected affinity agent is characterized by screening the selected affinity agent against a second plurality of peptides, wherein each peptide in the second plurality of peptides comprises a second epitope, and one or more flanking amino acid residues.

13. The method according to claim 1, further comprising: Binding of the selected affinity agent to a plurality of peptides shorter than peptide epitope X is determined by screening the selected affinity agent against the plurality of peptides.

14. The method of claim 1, further comprising characterizing the selected affinity agent by screening the selected affinity agent against a panel of proteins or peptides containing peptide epitope X.

15. The method of claim 14, wherein the affinity reagent selected is retained if it binds to more than 10% of the proteins or peptides of the group.

16. The method of claim 1, further comprising characterizing the selected affinity agent by screening the selected affinity agent against a panel of proteins or peptides that do not contain the peptide epitope X.

17. The method of claim 16, wherein the affinity reagent selected is retained if it binds to less than 15% of the proteins or peptides of the group.

18. The method of claim 1, wherein the selected affinity agent is screened against a library of random peptides of 2 to 10 amino acids in length, and further comprising: identifying peptides bound by the selected affinity reagent; as well as If the selected affinity reagent 1) binds fewer than twenty different peptide epitopes, and 2) binds at least 10% of all longer peptides comprising that peptide epitope X, then the affinity reagent is retained.

19. The method of claim 18, wherein the affinity agent selected is not retained unless it binds at least 15% of all longer peptides comprising the peptide epitope X.

20. The method of claim 1, further comprising: determining binding of the affinity agent to a peptide in the peptide library by screening the affinity agent against a peptide library comprising peptides shorter than peptide epitope X, peptides of the same length as peptide epitope X, and peptides longer than peptide epitope X; and The sequence of the bound peptide is identified, thereby characterizing the binding of the affinity agent.

Images