Protein redesign method

Abstract

The invention provides a method for redesigning hydrophobic residues on the surface of protein, which comprises the following steps of: firstly, determining the hydrophobic residues on the surface in a specified area, writing the hydrophobic residues into a readable file, and then redesigning the hydrophobic residues on the surface according to a certain rule, so as to obtain the protein with reduced surface hydrophobicity. The problem of reducing protein aggregation is effectively solved.

Description

Technical Field

[0001] The present application relates to the field of biotechnology, specifically to the field of synthetic biotechnology, and more specifically to a method for redesigning hydrophobic residues on the surface of a protein. Background Art

[0002] Protein is the main functional molecule of cells and the main component of cells. Protein can participate in the metabolism of the body and the transport of substances in the body. It can catalyze chemical reactions in the form of enzymes, can serve as a nutrient storage structure to play a storage role, can provide structural elements in cells and extracellular matrices, act as messengers, and so on. It can be said that proteins are the basis of life, and they are involved in almost all cell functions. Each protein will spontaneously fold into a unique natural structure. However, when proteins are physiologically active, they often aggregate due to incorrect folding or environmental conditions during the process of unfolding or refolding, thereby affecting their physical and chemical properties and biological functions. This is also one of the main reasons for the poor performance of proteins.

[0003] Studies have confirmed that the deposition of amyloid protein aggregates in the brain is the main pathological feature of Alzheimer's disease. The main component is a peptide called βA4, and researchers have reduced the amyloidosis of the βA4 peptide in Alzheimer's disease by replacing hydrophobic amino acids. Other diseases have also been increasingly confirmed to be related to protein aggregation. Therefore, studying the mechanism of protein aggregation and finding ways to prevent protein aggregation have become hot topics for researchers in the pharmaceutical industry.

[0004] Unlike finding and solving aggregation problems in natural proteins with known conformations, artificial protein synthesis requires considering multiple factors such as protein functionality, biological activity, and three-dimensional structure while taking into account the solution of aggregation problems. This is a complex project. As more proteins are designed or modified to be used as tools for studying biological systems or as more wild-type or modified proteins are used as therapeutic agents, these proteins need to be modified into more stable and / or aggregation-resistant systems.

[0005] US Patent No. US10407474B2 provides a system for modifying proteins to make them more stable, and points out that modifying the hydrophobic regions on the surface of proteins can improve the super-thermodynamic properties of proteins. It also points out that the larger the protein, the more residues need to be modified, and the more hydrophobic or the easier it is to aggregate or precipitate, the more residues need to be modified. However, this substitution has limitations, and it is easy for those skilled in the art to think that when the number of residues that need to be redesigned reaches a certain number, it is impossible to achieve it through artificial mutation, and it can only be designed by artificial intelligence.

[0006] Jue WANG et al. provided a method for designing scaffolds for protein functional domains, namely the RFDesign algorithm, which connects two or more protein functional regions by redesigning one or more scaffolds, keeping the structure and function of the functional regions basically unchanged. Although this method can use the proportion of surface hydrophobic residues as one of the loss functions to reduce the overall aggregation of the protein, it is affected by other necessary loss functions and it takes a long time to design a sequence, so it is impossible to complete the redesign of the surface hydrophobic residues in the specified area within a limited time; moreover, the protein obtained by this method only ensures that the structure and basic function of the functional region remain unchanged, and the overall structure of the protein has undergone significant changes, which may have a significant impact on the overall physical and chemical properties and biological activity.

[0007] J.DAUPARAS et al. also provided a method for redesigning specified regions of proteins - ProteinMPNN. This algorithm can specify the use of only certain given residues. However, through multiple tests, it was found that the structure predicted by AlphaFold after the redesigned sequence was quite different from the original structure of the protein. In the case of large structural changes, it is difficult to ensure the physiological function of the protein. Therefore, this method still cannot meet the design needs.

[0008] It can be seen that the methods in the existing technology are still not sufficient to solve the problem of protein aggregation. It is urgent to develop a design method that can effectively reduce protein aggregation while maintaining good physiological activity of the protein. Summary of the invention

[0009] The purpose of the present invention is to overcome the problems existing in the above-mentioned prior art protein design and provide a method for redesigning proteins, specifically a method for redesigning the surface hydrophobic residues of proteins. Through the method of the present application, the modification of the surface hydrophobic residues of proteins can be achieved, thereby reducing the aggregation of proteins.

[0010] The first aspect of the present invention provides a method for redesigning the surface hydrophobic residues of a specified region of a protein while keeping the structure substantially unchanged, comprising the following steps:

[0011] S1: Determine the surface hydrophobic residue sites within the specified region and write them into a readable file;

[0012] S2: These sites are considered as scaffolds that need to be redesigned and redesigned using scaffold design software.

[0013] The designated region refers to a region designated by the user according to his / her needs. It can be a continuous region on the same chain, multiple discontinuous regions on the same chain, or a region distributed on two chains. If there are continuous residues without structural information at both ends of the designated region, these residues will be automatically subtracted from the designated region.

[0014] In another preferred embodiment, when the designated region requiring redesign of the surface hydrophobic residues is distributed on two or more chains, the process is performed twice or more times.

[0015] In another preferred embodiment, the protein has two or less peptide chains, more preferably one peptide chain.

[0016] In another preferred embodiment, the step S1 comprises the following steps:

[0017] S11: Determination of surface hydrophobic residues within a specified region;

[0018] S12: Generate a character string based on the information of the surface hydrophobic residues in the specified region, which can be read by the scaffold design software for redesign.

[0019] In another preferred embodiment, the step S11 includes the following steps:

[0020] S111: Parse protein structure files, such as pdb files, read residue sequences, atomic coordinates, and lists of chain names + residue numbers and save them to a dictionary for later use;

[0021] S112: One-hot encoding for all residues;

[0022] S113: Calculate a neighbor number for each residue using the main chain atomic coordinates;

[0023] S114: setting a neighbor number threshold, preferably a value less than or equal to 3.5, more preferably 3, and more preferably 2.5;

[0024] S115: determine the types of hydrophobic residues, such as A, F, I, L, M, V and W, representing alanine, phenylalanine, isoleucine, leucine, methionine, valine and tryptophan, respectively;

[0025] S116: When a residue in the designated region is a hydrophobic residue and the number of its neighbors is less than the neighbor number threshold, the residue is a surface hydrophobic residue.

[0026] In another preferred embodiment, the above step S115 may be arranged after step S111, that is, the step S11 includes the following steps:

[0027] S111: Parse the structure file of the protein to be redesigned, read the residue sequence, atomic coordinates, and the list of chain name + residue number and save them into the dictionary for later use;

[0028] S115: determining whether it is a hydrophobic residue according to the read residue sequence, and marking the name;

[0029] S112: One-hot encoding for all residues;

[0030] S113: Calculate a neighbor number for each residue using main chain atomic coordinates;

[0031] S114: setting a neighbor number threshold, such as 2.5 or 2.0;

[0032] S116: When a residue in the designated region is a hydrophobic residue and the number of its neighbors is less than the neighbor number threshold, the residue is a surface hydrophobic residue.

[0033] In another preferred embodiment, the number of neighbors is calculated by using Layerselector in ResidueSelectors program.

[0034] In another preferred embodiment, the specific algorithm for calculating the number of neighbors in step S113 is:

[0035] The number of adjacent residues within a cone is obtained from the CA-CB vector of each residue, where CA refers to the alpha carbon atom of the residue and CB refers to the beta carbon atom of the residue, as follows:

[0036] The neighboring residues obtained in the aforementioned vector range are counted, that is, the number of neighboring residues. Each of the aforementioned neighboring residues is weighted by multiplying the distance coefficient by the angle coefficient to obtain the weight of this level. The weights of all the neighboring residues of this residue are added together to obtain the value obtained by the neighbor number of this residue.

[0037] The distance coefficient and the angle coefficient are calculated as follows:

[0038] Distance coefficient = 1 / (1+exp(n*(dm))),

[0039] Where d is the distance between the residue CA and the adjacent residue CA;

[0040] m is the midpoint of distance attenuation;

[0041] n is the decay exponential factor, which determines the sharpness of the distance decay (the larger the value, the sharper the decay near the midpoint distance).

[0042] Angle coefficient = ((cos(theta)+a) / (1+a))^b

[0043] Where theta is the angle between the CA-CB vector of the residue and the CA vectors of the residue and the neighboring residues;

[0044] a is an offset factor used to widen the cone to a certain extent;

[0045] b is an index that determines the sharpness of the angle drop (the smaller the value, the wider the cone and the sharper the edge drop).

[0046] In another preferred embodiment, the step S12 includes the following steps:

[0047] S121: Generate a list according to the chain where the specified region is located and all its residue numbers. If the residue with the number is not a surface hydrophobic residue in the specified region, the element corresponding to the number is the chain name + residue number. If the residue with the number is a surface hydrophobic residue in the specified region, the element corresponding to the number is 1.

[0048] S122: Write the above list into a readable file, such as a text file.

[0049] In another preferred embodiment, the step S2 comprises the following steps:

[0050] S21: Open the virtual environment required for the operation of the stent design software;

[0051] S22: Using the scaffold design software, the readable file obtained in the above step is read as a parameter, and the remaining parameters (such as the parameter num for the number of design entries, the parameter steps for the optimization step, the exclude_aa parameter, etc.) are set to perform sequence design to obtain the redesigned protein sequence.

[0052] In another preferred embodiment, after step S2, a step of verifying the redesigned protein sequence is added, preferably by methods such as structure prediction, aggregation prediction, protein synthesis and functional experiments, for example, using the AlphaFold program to predict the structure of the designed protein sequence, or using the Aggrescan3D server to predict the aggregation of the designed protein.

[0053] In another preferred embodiment, if the protein to be redesigned is two peptide chains, the method specifically comprises the following steps:

[0054] The atomic coordinate data of the two peptide chains were extracted from the pdb and saved as separate files, and stored in the corresponding locations for the scaffold design software to read and design;

[0055] When the designated region is distributed in one peptide chain, the chain is designated as a template chain for redesign, and another chain is used as a target chain to verify and adjust the redesign result of the template chain to obtain a new chain;

[0056] When the designated region is distributed in two peptide chains, one of the chains is first designated as the template chain for redesign, and the other chain is designated as the target chain to verify and adjust the result of the redesign of the template chain to obtain a first new chain; then the other chain is designated as the template chain for redesign, and the first new chain is designated as the target chain to verify and adjust the result of the redesign of the other chain to obtain a second new chain. Specifically, there are two methods:

[0057] The first method is to use the first chain as a template to be designed and the second chain as a binding target, perform the aforementioned steps S1-S2 on the first chain, redesign the surface hydrophobic residues in the specified area on the chain, and obtain a new single-chain sequence as the new first chain; then use the second chain as a template to be designed and the new first chain as a binding target, perform the aforementioned steps S1-S2 on the second chain, redesign the surface hydrophobic residues in the specified area on the chain, and obtain a new single-chain sequence as the new second chain sequence. Finally, use the new first chain and the new second chain to form a new double-chain protein, verify the structure and aggregation, and compare and verify it with the original protein structure. Then adjust and optimize according to the actual situation.

[0058] The second method is to first use the second chain as the template chain to be designed, the first chain as the binding target, perform the aforementioned steps S1-S2 on the second chain, redesign the surface hydrophobic residues in the specified area on the chain, obtain a new single-chain sequence as the new second chain; then use the first chain as the template to be designed, the new second chain as the binding target, perform the aforementioned steps S1-S2 on the first chain, redesign the surface hydrophobic residues in the specified area on the chain, obtain a new single-chain sequence as the new first chain, and finally, use the new first chain and the new second chain to form a new double-chain protein, perform structural and aggregation verification, and compare and verify with the original protein structure. Then adjust and optimize according to actual conditions.

[0059] The second aspect of the present invention provides a protein with reduced aggregation, which is designed using the design method of the first aspect. The surface hydrophobic residues of the protein are reduced, thereby reducing aggregation, but the basic structure and function remain unchanged. The protein includes but is not limited to insulin, growth hormone, myoglobin, pancreatic ribonuclease, hemoglobin, etc.

[0060] The third aspect of the present invention provides a computer program stored in a computer-readable medium and loadable into the internal memory of a computer, comprising a software code portion, which is arranged to execute the design method described in any one of the first aspects of the present invention when the program is run on a computer.

[0061] The fourth aspect of the present invention provides an application of the design method described in the first aspect, specifically, the application of the design method in the design of proteins, especially functional proteins. The functional protein is any protein required in the field of medicine and biology, and can be a naturally occurring protein or an artificially designed protein sequence. When the protein is a natural protein, its source includes prokaryotes, eukaryotes or humans.

[0062] Compared with the prior art, the present invention has the following beneficial effects or advantages:

[0063] (1) The present invention changes the surface hydrophobicity of a protein by redesigning the surface amino acid residue sequence of a specified region of the protein, thereby reducing the aggregation of the protein and improving the stability of the protein, which is beneficial for the protein to exert its biological function and play a more effective role in health care or treatment, thereby achieving improved therapeutic effects.

[0064] (2) Although the design method of the present invention changes the surface hydrophobicity of the protein, the overall structure of the protein remains basically unchanged. The redesigned protein is close to its original natural structure. Therefore, except for the reduction of aggregation, the other physical and chemical properties and biological activities are less affected, which is more conducive to the exertion of its biological activity.

[0065] (3) The protein design method of the present invention can perform large-scale calculations by designing a predetermined algorithm, and can quickly design a method for modifying the surface hydrophobic residues of proteins within two chains. It has the advantages of being faster and more efficient than other design methods in the prior art, and the obtained protein structure has the advantage of being closer to the natural structure than other design methods, and has broad application prospects.

[0066] (4) Through the design method of the present invention, not only can the aggregation of proteins be reduced, but also the aggregation of proteins can be increased according to needs. That is, the protein sequence can be modified according to actual needs to obtain reduced or increased aggregation, thereby improving the physical and chemical properties of proteins. Therefore, the instability of proteins in the application of the prior art is overcome, and the application scope of proteins is expanded, which is of great significance for pharmaceutical and clinical research.

[0067] (5) The present invention also provides a computer program corresponding to the protein design method, which is stored in a computer-readable medium and can be loaded into the internal memory of a computer, so that users can use and calculate conveniently. There is no need to construct algorithms one by one. It is only necessary to import corresponding files into the computer program and adjust some parameters to realize the intelligent design of protein hydrophobic residues.

[0068] It should be understood that within the scope of the present invention, the above-mentioned technical features of the present invention and the technical features specifically described below (such as embodiments) can be combined with each other to form a new or preferred technical solution. Due to space limitations, they will not be described one by one here. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 The structure predicted by Alphfold2 (cyan) of sequence A3, which was redesigned directly using the scaffold design software RFDEsign for the designated region A110-582 of the A chain of 1AO6 (i.e., residues 110 to 582 of the A chain), is shown, as well as its alignment with the original structure of the A chain of 1AO6 (green).

[0070] Figure 2 The structure predicted by AlphaFold (cyan) of sequence A4, which was directly designed using ProteinMPNN to redesign the surface hydrophobic residues in the specified region of 1AO6, is shown, as well as its alignment with the original structure of chain A of 1AO6 (green).

[0071] Figure 3 The structures predicted by AlphaFold for two sequences designed using the inventive method are shown (A1 in cyan and A2 in magenta), as well as their structural alignment with the original structure of chain A of 1AO6 (green).

[0072] Figure 4 The structures predicted by AlphaFold (cyan and magenta) of two sequences designed by the method of the present invention, with the residues missing at both ends of the A chain and the complete B chain added, are shown, and compared with the 1AO6 structure with both A and B chains complete (green).

[0073] Figure 5 The Aggrescan3D score of the A chain of 1AO6 is shown.

[0074] Figure 6 The Aggrescan3D score of sequence A1 designed using the method of the present invention is shown.

[0075] Figure 7 The Aggrescan3D score of sequence A2 designed using the method of the present invention is shown. DETAILED DESCRIPTION

[0076] After extensive and in-depth research, a large number of screenings and explorations, the present invention proposes for the first time a method for redesigning hydrophobic residues on the surface of proteins. This method changes the surface hydrophobicity of proteins by redesigning the surface amino acid residue sequence of a specified region of the protein, reduces the overall aggregation of the protein, and thus improves the stability of the protein, while maintaining the basic structure of the protein unchanged and maintaining the integrity of its biological activity.

[0077] The present invention is further illustrated below in conjunction with specific implementation methods and examples. For the specific methods or materials used in the examples, those skilled in the art can make conventional replacement selections based on the technical ideas of the present invention and according to existing technologies, and are not limited to the specific description of the embodiments of the present invention.

[0078] Terminology

[0079] The "hydrophobic residue" described in the present invention is also called "amino acid hydrophobic residue" or "hydrophobic amino acid residue", which refers to an amino acid residue whose side chain does not contain a hydrophilic group. Hydrophobic residues can be tightly combined with other hydrophobic amino acid residues in an aqueous medium through hydrophobic interactions, and are the main residue types in the inner core of globular proteins. Natural proteins all contain certain hydrophobic amino acid residues, which can form aggregates through the interaction of hydrophobic regions on their outer surface regions, which is also the main reason for protein aggregation. Common hydrophobic amino acids include alanine, phenylalanine, isoleucine, leucine, methionine, valine and tryptophan, which are represented by A, F, I, L, M, V and W, respectively.

[0080] The "peptide chain" of the present invention, sometimes referred to as "chain" in the present invention, refers to a chain structure formed by the dehydration condensation of multiple amino acids to form peptide bonds (chemical bonds), which is also the basic structure of proteins. The design method of the present invention can be used to redesign the peptide chain of a protein, taking into account the complexity of the tertiary structure of the protein, and the fewer the number of peptide chains, the easier it is to ensure the stability of the redesigned tertiary structure. Therefore, the method of the present invention is preferably redesigned for proteins composed of two or less peptide chains. The "pdb file" of the present invention refers to the PDB archive file (Protein Data Bank) of a protein, which is a repository of atomic coordinates and other information describing proteins and other important biological macromolecules. Structural biologists use methods such as X-ray crystallography, nuclear magnetic resonance spectroscopy and cryo-electron microscopy to determine the position of each atom in a molecule relative to each other, and then store this information in an archive, annotate it and publish it publicly in the PDB archive. The "dictionary" of the present invention is a data structure of Python, consisting of a collection of key-value pairs, each of which maps a key to its associated value.

[0081] The "one-hot encoding" described in the present invention is One-Hot encoding, also known as one-bit effective encoding, and its method is to use an N-bit state register to encode N states, each state has its own independent register bit, and at any time, only one bit is valid.

[0082] The "AlphaFold" described in the present invention is an artificial intelligence program from DeepMind that has ultra-high accuracy in predicting protein structures. The predicted structure of most proteins is only one atom away from the real structure, reaching the level of human observation and prediction using complex instruments such as cryo-electron microscopes.

[0083] The "Aggrescan3D" described in the present invention, referred to as "A3D", is a server for predicting aggregation tendency in protein structure and rationally designing protein solubility. In dynamic mode, A3D takes into account the flexibility of natural and designed peptides. Aggrescan3D can reflect the predicted aggregation of proteins, and the lower the score, the less likely the surface protein is to aggregate.

[0084] The "neighbor number" of the present invention is a parameter set to reflect the spatial relationship between each residue and other residues or atomic positions. By calculating the number of neighbors, the spatial relationship between the amino acid residue and the surrounding amino acid residues can be displayed, reflecting the possibility of its being located on the surface of the protein. In short, the larger the number of neighbors of an amino acid residue, the more other amino acid residues are surrounding it, and it is usually not on the surface of the protein. When the number of neighbors of an amino acid residue is small, it means that there are few other amino acid residues surrounding it, and it is usually closer to the protein surface. The number of neighbors can be calculated using the Layerselector in the ResidueSelectors program, and the specific algorithm is:

[0085] The number of adjacent residues within a cone is obtained from the CA-CB vector of each residue, where CA refers to the alpha carbon atom of the residue and CB refers to the beta carbon atom of the residue, as follows:

[0086] The neighboring residues obtained in the aforementioned vector range are counted, that is, the number of neighboring residues. Each of the aforementioned neighboring residues is weighted by multiplying the distance coefficient by the angle coefficient to obtain the weight of this level. The weights of all the neighboring residues of this residue are added together to obtain the value obtained by the neighbor number of this residue.

[0087] The distance coefficient and angle coefficient are calculated as follows:

[0088] Distance coefficient = 1 / (1+exp(n*(dm))),

[0089] Where d is the distance between the residue CA and the adjacent residue CA;

[0090] m is the midpoint of distance attenuation;

[0091] n is the decay exponential factor, which determines the sharpness of the distance decay (the larger the value, the sharper the decay near the midpoint distance).

[0092] Angle coefficient = ((cos(theta)+a) / (1+a))^b

[0093] Where theta is the angle between the CA-CB vector of the residue and the CA vectors of the residue and the neighboring residues;

[0094] a is an offset factor used to widen the cone to a certain extent;

[0095] b is an index that determines the sharpness of the angle drop (the smaller the value, the wider the cone and the sharper the edge drop).

[0096] The "neighbor number threshold" described in the present invention is a setting parameter for distinguishing whether an amino acid in a protein is a surface hydrophobic residue, and is an empirical value summarized after a large amount of data collection and collation. Usually, the value is set below 3.5, preferably 3, and more preferably 2.5.

[0097] The "readable file" described in the present invention refers to a file format that can be read by the protein scaffold design software described in the present invention, and can be selected according to the requirements of the protein scaffold design software. In the present invention, there is no strict requirement. Common readable files include text files, such as txt, jsonl, json, etc.

[0098] The method of the present invention for redesigning the surface hydrophobic residues of a specified region of a protein while keeping the structure substantially unchanged comprises the following steps:

[0099] S1: Determine the surface hydrophobic residue sites within the specified region and write them into a readable file;

[0100] S2: These sites are considered as scaffolds that need to be redesigned and redesigned using scaffold design software.

[0101] The designated region is a region designated by the user according to his / her needs. It can be a continuous region on the same chain, multiple discontinuous regions on the same chain, or a region distributed on two chains. If there are continuous residues without structural information at both ends of the designated region, these residues will be automatically subtracted from the designated region.

[0102] The specific hydrophilicity and hydrophobicity of the amino acid residues in the specified region of the peptide chain and their positions in the three-dimensional structure of the protein are obtained through step S1 and written into a file that can be read by subsequent protein scaffold design software to facilitate redesign.

[0103] The step S2 can be designed using known protein scaffold design software, such as RFDesign and AFDesign. RFDesign is preferably used in the embodiment of the present invention. Based on the overall situation of the amino acid residues on the peptide chain obtained in step S1 and the readable file obtained, the protein scaffold design software can design a new replacement peptide chain according to the needs, thereby reducing the hydrophobicity of the peptide chain.

[0104] In one example, the specific operation method of step S1 includes the following steps:

[0105] S11: Determination of surface hydrophobic residues within a specified region;

[0106] S12: Generate information that can be read by the scaffold design software for redesign based on the information of the surface hydrophobic residues in the designated area and form the readable file, which is preferably a text file.

[0107] In order to redesign the surface hydrophobic residues, it is necessary to first determine which amino acid residues on the peptide chain are surface hydrophobic residues, that is, the surface hydrophobic residues must simultaneously meet the two conditions of being hydrophobic amino acid residues and being located on the surface of the three-dimensional structure of the protein. Therefore, it is necessary to parse the peptide chain structure of the protein and calculate the spatial position relationship of the amino acid residues to determine the surface hydrophobic residues. The parsed file must be able to express the hydrophobic amino acid residues and their spatial position characteristics at the same time, and be written in a readable file format to facilitate the subsequent reading and use of protein scaffold design software.

[0108] In one example, the step S2 specifically includes: using scaffold design software, reading the readable file obtained in the step S1 as a parameter, and setting the remaining parameters to perform sequence design on the designated region to obtain a redesigned protein sequence.

[0109] After obtaining a readable file that can describe the hydrophobic amino acids and spatial positions of the peptide chain in step S1, the readable file can be imported into the protein scaffold design software, and the parameters can be set according to the parameter setting requirements of the design software to start redesigning the hydrophobic amino acid residues in the specified area.

[0110] In one example, the specific operation method of step S11 includes the following steps:

[0111] S111: Parse the file of protein structure to be redesigned, read the residue sequence, atomic coordinates, and the list of chain name + residue number and save them into the dictionary for later use;

[0112] S112: One-hot encoding for all residues;

[0113] S113: Calculate a neighbor number for each residue using main chain atomic coordinates;

[0114] S114: Setting a neighbor number threshold;

[0115] S115: determining whether it is a hydrophobic residue according to the read residue sequence, and marking the name;

[0116] S116: When a residue in the designated region is a hydrophobic residue and the number of its neighbors is less than the neighbor number threshold, the residue is a surface hydrophobic residue.

[0117] As mentioned above, the determination of surface hydrophobic residues in a specified region requires the analysis of the amino acid structure of the peptide chain and the spatial position of the hydrophobic amino acid. Therefore, first, a file recording the protein structure to be redesigned must be obtained. This file is often a document that has been recorded in the prior art and is scientifically reliable, such as a file (referred to as a pdb file) published on the Protein Data Bank website. After obtaining the pdb file of the protein, it is necessary to parse it, read the residue sequence, atomic coordinates, chain name + residue number, that is, to know which positions of the amino acid residues are hydrophobic amino acids, and what kind of hydrophobic amino acid residues they are, and save it in a dictionary. Afterwards, a unique hot encoding is used to represent all amino acid residues in the peptide chain according to the content recorded in the dictionary. Afterwards, the number of neighbors is calculated for each amino acid residue, and the number of neighbors can reflect the spatial positional relationship between an amino acid and other amino acids. When the number of neighbors of a certain amino acid residue is less than a threshold value, the surface position of its position protein structure, i.e., the surface amino acid, can be determined, and when the amino acid residue is just a hydrophobic amino acid, it can be determined that the amino acid residue is a surface hydrophobic residue of the protein, i.e., an amino acid that needs to be redesigned.

[0118] In one example, in the specific operation method of step S11, considering the relationship between the steps, step S115 may be set before step S112, which does not affect the final determination of the surface hydrophobic residues, that is, step S11 includes the following steps:

[0119] S111: Parse the structure file of the protein to be redesigned, read the residue sequence, atomic coordinates, and the list of chain name + residue number and save them into the dictionary for later use;

[0120] S115: determining whether it is a hydrophobic residue according to the read residue sequence, and marking the name;

[0121] S112: One-hot encoding for all residues;

[0122] S113: Calculate a neighbor number for each residue using main chain atomic coordinates;

[0123] S114: Setting a neighbor number threshold;

[0124] S116: When a residue in the designated region is a hydrophobic residue and the number of its neighbors is less than the neighbor number threshold, the residue is a surface hydrophobic residue.

[0125] In one example, the number of neighbors is calculated using LayerSelector in the ResidueSelectors program, which can evaluate the spatial position of the amino acid residue by the number of side chain neighbors along the cone of the CA-CB vector to determine whether it is located on the surface.

[0126] In one example, the S2 step specifically re-mutates the sites identified as surface hydrophobic residues in the specified region. Only by re-designing the surface hydrophobic residues can the surface hydrophobicity be most effectively and directly reduced to prevent protein aggregation. However, this does not mean that the surface hydrophobic residues can be simply replaced with non-hydrophobic amino acid residues, because simple replacement may lead to changes in the spatial structure of the protein, affect the overall activity and function of the protein, or produce more serious changes. Therefore, the re-design needs to consider various factors such as surface hydrophobicity, the mutual influence of each amino acid residue in the spatial position, and the biological activity of the protein.

[0127] In one example, the step S12 includes the following steps:

[0128] S121: Generate a list according to the chain where the specified region is located and all its residue numbers. If the residue with the number is not a surface hydrophobic residue in the specified region, the element corresponding to the number is the chain name + residue number. If the residue with the number is a surface hydrophobic residue in the specified region, the element corresponding to the number is 1.

[0129] S122: writing the above list into a readable file;

[0130] Furthermore, the step S2 performs a re-mutation design on the surface hydrophobic residues corresponding to the corresponding element 1 in the list.

[0131] By using symbols to record the hydrophobicity and surface position of amino acid residues, readable data is provided for subsequent computer program design, and those skilled in the art can also intuitively see the hydrophobic amino acids in the peptide chain and their specific positions on the peptide chain.

[0132] In one example, the protein has two or less peptide chains, more preferably one peptide chain. As mentioned above, the design method of the present invention can be used to redesign the peptide chains of a protein, taking into account the complexity of the tertiary structure of the protein, and the fewer the number of peptide chains, the easier it is to ensure the stability of the redesigned tertiary structure, so the method of the present invention is preferably used to redesign proteins composed of two or less peptide chains.

[0133] In one example, when the protein is a double chain and the designated region is distributed in one chain, the chain is designated as the template chain for redesign, and the other chain is used as the target chain to verify and adjust the result of the redesign of the template chain to obtain a new chain; when the protein is a double chain and the designated region is distributed in two chains, one chain is first designated as the template chain for redesign, and the other chain is used as the target chain to verify and adjust the result of the redesign of the template chain to obtain a first new chain; then the other chain is used as the template chain for redesign, and the first new chain is used as the target chain to verify and adjust the result of the redesign of the other chain to obtain a second new chain. The above method can avoid the adverse effects on the structure and function when designing two peptide chains at a time, and is necessary for maintaining the stability of protein structure and biological activity.

[0134] In one example, after step S2, a step of verifying the redesigned protein sequence is added, preferably structure prediction, aggregation prediction, protein synthesis, protein functionality prediction, etc. Through verification of structure, aggregation, functionality, etc., it is ensured that the redesigned protein has the same or very similar structure and biological activity as the original protein. Because biological activity is the basis for the utilization of protein, the verification step is a verification and supplement to the protein design method described in the present invention.

[0135] In one example, a protein with reduced aggregation obtained by redesigning the above method is also provided. The redesigned protein is very close to the original protein in terms of three-dimensional structure and physiological activity, and may even improve certain aspects of its performance, such as increased activity, extended half-life, etc., due to reduced aggregation.

[0136] In one example, a computer program stored in a computer-readable medium and loadable into an internal memory of a computer is also provided, comprising a software code portion, wherein the software code portion is arranged to execute any of the aforementioned design methods when the program is run on a computer. A person skilled in the art only needs to provide a structure file of a protein to be redesigned and input it into the computer program to obtain a redesigned protein.

[0137] In one example, the application of the above-mentioned method for redesigning hydrophobic residues on the surface of a protein in protein design is also provided, and the protein is a natural protein or an artificial protein. The method of the present invention has a very wide range of applications, and there is no special limitation on the source of the designed protein. As long as the protein needs to be designed and its structure is known, it can be redesigned in a targeted manner.

[0138] Example 1 Redesign of the human serum albumin crystal structure 1AO6

[0139] In this example, the human serum albumin crystal structure 1AO6 (hereinafter referred to as 1AO6) is used as the target protein, and its surface hydrophobic residues are modified by the method described in the present invention to obtain 1AO6 with low aggregation. Specifically, the surface hydrophobic sites (alanine, phenylalanine, isoleucine, leucine, methionine, valine and tryptophan) in the specified region (A110-585) of 1AO6 are all used as the scaffolds to be designed and redesigned using RFDesign, and the above hydrophobic residues are specified not to be used during the design. The specific design method steps are as follows:

[0140] S1: Determine the surface hydrophobic residue sites in the specified region and write them into a readable file, including the following specific steps (S11-S12):

[0141] S11: Determination of the surface hydrophobic residue sites in the designated region, including the following specific steps (S111-S116):

[0142] S111: Obtain the pdb file for parsing 1AO6 from the PDB archive server, read the amino acid residue sequence, atomic coordinates, and the list of chain name + residue number and save them to the dictionary for later use. The download address of the protein template pdb file is: https: / / files.rcsb.org / download / 1AO6.pdb. This example uses 1AO6_A.pdb, which is a pdb file consisting of all atomic coordinates of chain A extracted from 1AO6.pdb.

[0143] The template pdb consists of two chains. We hope that after redesigning the surface hydrophobic residues in the specified region A110-585, the structure will remain basically unchanged when performing structure prediction together with the B chain. Therefore, we need to specify the B chain as the binding target and extract the atomic coordinates of the B chain from the template pdb and save it as a separate pdb file, namely 1AO6_B.pdb, as the binding target.

[0144] It should be noted that in many cases, you may find that the coordinates presented in the ATOM record in the PDB file may not completely match the sequence in the SEQRES record. In particular, when there are missing residues at the front end of the chain, the residue numbers of the designed sequence may be inconsistent with the corresponding residue numbers of the template, such as 1AO6.pdb, where the first four and last three residues of each chain are unstructured.

[0145] When specifying a region, you must specify whether the count of the start and end sites should start from the beginning or from the sites with structures in the pdb. All the following will be set to start from the beginning.

[0146] Assuming the specified area is A110-585, the sequence of the specified area is as follows:

[0147] PNLPRLVRPEVDVMCTAFHDNEETFLKKYLYEIARRHPYFYAPELLFFAKRYKAAFTECCQAADKAACLLPKLDELRDEGKASSAKQRLKCASLQKFGERAFKAWAVARLSQRFPKAEF AEVSKLVTDLTKVHTECCHGDLLECADDRADLAKYICENQDSISSKLKECCEKPLLEKSHCIAEVENDEMPADLPSLAADFVESKDVCKNYAEAKDVFLGMFLYEYARRHPDYSVVLLL RLAKTYETTLEKCCAAADPHECYAKVFDEFKPLVEEPQNLIKQNCELFEQLGEYKFQNALLVRYTKKVPQVSTPTLVEVSRNLGKVGSKCCKHPEAKRMPCAEDYLSVVLNQLCVLHEK TPVSDRVTKCCTESLVNRRPCFSALEVDETYVPKEFNAETFTFHADICTLSEKERQIKKQTALVELVKHKPKATKEQLKAVMDDFAAFVEKCCKADDKETCFAEEGKKLVAASQAALGL

[0148] However, since there is no structure of the last three residues LGL in the PDB, it is actually A110-582.

[0149] S112: One-hot encoding for all amino acid residues;

[0150] S113: Calculate a neighbor number for each residue using the main chain atomic coordinates, using Layerselector in the ResidueSelectors program;

[0151] S114: Setting a neighbor number threshold, preferably 2.5;

[0152] S115: determine the types of hydrophobic residues, such as A, F, I, L, M, V and W, representing alanine, phenylalanine, isoleucine, leucine, methionine, valine and tryptophan, respectively;

[0153] S116: When a residue in the designated region is a hydrophobic residue and the number of its neighbors is less than the neighbor number threshold, the residue is a surface hydrophobic residue.

[0154] S12: Generate a file that can be read by RFDesign for redesign based on the information of the surface hydrophobic residues in the specified area, including the following specific steps (S121-S122):

[0155] S121: Generate a list according to the chain where the specified region is located and all its residue numbers. If the residue with the number is not a surface hydrophobic residue in the specified region, the element corresponding to the number is the chain name + residue number. If the residue with the number is a surface hydrophobic residue in the specified region, the element corresponding to the number is 1.

[0156] S122: Write the above list into a text file 1.

[0157] The content of text file 1 is as follows:

[0158] <h2 style=";text-align:left;direction:ltr">A5, A6, A7, A8, A9, A10, A11, A12, A13, A14, A15, A16, A17, A18, A19, A20, A21, A22, A23, A24, A25, A26, A27, A28, A29, A30, A31, A32, A33, A34, A35, A36, A 37, A38, A39, A40, A41, A42, A43, A44, A45, A46, A47, A48, A49, A50, A51, A52, A53, A54, A55, A56, A57, A58, A59, A60, A61, A62, A63, A64, A65, A66, A67, A6 8, A69, A70, A71, A72, A73, A74, A75, A76, A77, A78, A79, A80, A81, A82, A83, A84, A85, A86, A87, A88, A89, A90, A91, A92, A93, A94, A95, A96, A97, A98, A99 ,A100,A101,A102,A103,A104,A105,A106,A107,A108,A109,A110,A111,A 112,A113,A114,A115,1,A117,A118,A119,A120,A121,1,A123,A124,A125, 1,A127,A128,A129,A130,A131,A132,A133,A134,1,A136,A137,A138,A13 9,A140,A141,A142,A143,A144,A145,A146,A147,A148,A149,A150,A151, A152,A153,A154,A155,A156,A157,A158,A159,A160,A161,A162,1,A164, A165,A166,A167,A168,A169,A170,A171,1,A173,A174,A175,A176,A177,A 178, A179, A180, A181, A182, A183, A184, A185, A186, A187, A188, A189, A190, A192, A193, A194, A195, A196, A197, A198, A199, A200, A201, A202, A203 A204, A205, A206, A207, A208, A209, A210, A211, A212, A213, A214, A215, A216, A217, A218, A219, A220, A221, A222, A223, A224, A225, A226, A227, 1, 1,<h2 style=";text-align:left;direction:ltr">A230, A231, A232, A233, A234, A235, A236, A237, A238, A239, A240, A241, A242, A243, A244, A245, A246, A247, A248, A249, A250, A251, A252, A253, A254, A255, A256, A257, A258, A259, A260, A261, A262, A263, A264, A265, A266, A267, A268, A269, A270, A271, A272, A273, A274, A275, A276, A277, A278, A279, A280, A281, A282, 1, A284, A285, A286, A287, A288, A289, A290, A291, A292, A293, A294, A295, A296, A297, 1, A299, 1, A301, A302, A303, A304, A305, 1, 1, A308, A309, A310, A311, A312, A313, A314, A315, A316, A317, A318, A319, 1, A321, 1, A323, A324, 1, A326, A327, A328, 1, A330, A331, A332, A333, A334, A3 35, A336, A337, A338, A339, A340, A341, A342, A343, A344, A345, A346, A347, A348, A349, A350, A351, A352, A353, A354, A355, A356, A357, A358, A359, A3 60,A361,1,A363,1,A365,A366,A367,A368,A369,A370,1,A372,A373,1,A375,A376,A377,A378,A379,A380,1,A382,A383,A384,A385,A386,A387,A3 88,A389,A390,A391,A392,A393,1,A395,A396,A397,A398,A399,A400,A401,A402,A403,A404,A405,A406,A407,A408,A409,A410,A411,A412,A413, A414,A415,A416,A417,A418,A419,A420,A421,A422,A423,A424,A425,A4 26,A427,A428,A429,A430,A431,A432,A433,A434,A435,A436,A437,A438,A439,A440,A441,A442,1,A444,A445,A446,A447,A448,A449,A450,A451,A452,A453,A454,A455,A4 56,A457,A458,A459,A460,A461,A462,A463,A464,A465,A466,A467,A468,A469,A470,A471,A472,A 473,A474,A475,A476,A477,A478,A479,A480,A481,A482,A483,A484,A485,A486,A487,A488,A489, 1,A491,A492,1,A494,A495,A496,A497,1,A499,A500,A501,1,A503,1,A505,A506,A507,A508,A509 ,A510,1,A512,A513,A514,A515,A516,A517,A518,A519,A520,A521,A522,1,A524,A525,A526,A527 ,A528,A529,A530,A531,A532,A533,A534,A535,A536,A537,A538,1,A540,A541,A542,A543,A544,A 545,1,A547,A548,A549,A550,A551,1,1,A554,A555,A556,A557,A558,A559,A560,A561,A562,A563 ,A564,A565,A566,A567,A568,1,A570,A571,A572,A573,A574,A575,A576,1,1,A579,A580,1,A582,

[0159] After all the specific operation methods of step S1 are completed, the above text file 1 is obtained, and then step S2 is performed as shown below:

[0160] S2: The process of redesigning these sites as scaffolds that need to be redesigned using RFDesign is as follows (S21-S22):

[0161] S21: Open the virtual environment required for RFDesign to run;

[0162] S22: Use the hallucination algorithm in RFDesign to read the above text file 1 as the mask parameter, and set the remaining parameters (such as the parameter num for the number of design items, the parameter steps for the optimization steps, the exclude_aa parameter, the force_aa parameter, etc.) to perform sequence design and obtain the sequence A1.

[0163] The sequence of the designed sequence A1 is as follows:

[0164] SEVAHRFKDLGEENFKALVLIAFAQYLQQCPFEDHVKLVNEVTEFAKTCVADESAENCDKSLHTLFGDKLCTVAT

[0165] LRETYGEMADCCAKQEPERNECFLQHKDDNPNLPRLTRPEVDRMCTYFHDNEETFDKKYLYEIARRHPYFYAPELLFFA

[0166] KRYKDAFTECCQAPDKAACLLPKLDELRDEGKNSSAKQRLKCASLQKFGERAFKAWAVARLSQRFPKAEQGEVSKLVTD

[0167] LTKVHTECCHGDLLECADDRADLAKYICENQDSISSKLKECCEKPPLEKSHCIAEVENDEPPDDLPSLENDFVESKDVC

[0168] KNYNEYKDKFLGPFLYEYARRHPDYSVVLLLRLAKTYETTLEKCCSAYDPHECYQKVNDEFKPLNEEPQNLIKQNCEKFEQLGEYKFQNALLVRYTKKVPQVSTPTLVEVSRNLGKVGSKCCKHPEDKRMPCA EDYLSVVLNQLCVLHEKTPVSDRVTKCCTESLVNRRPCFSRLEQDETYRPKESNCETFTFHNDICTLSEKERQTKKQTALVELVKHKPKDTKEQLKTVMDDFCEFVEKCCKADDKETCFEEEGKKLVKCSQRA

[0169] Regarding the mask parameter, it is a parameter of the hallucination algorithm in RFDesign, which contains the range of residues in the template that should be constrained in the hallucination. For example, "--mask = 20-25, A119-140, 5-15, B63-82, 5" means hallucinating 20-25 residues, then constraining a region to mimic the A chain residues 119-140 of the template, followed by 5-15 hallucinated residues, then mimicking the B chain residues 63-82, and then hallucinating 5 residues. The length of each hallucination region is uniformly randomly sampled from the specified range of each design and remains fixed during the optimization process.

[0170] There is no limit on the parameter num of the number of designs, and it can be any positive integer. However, based on practical considerations, do not choose a large number, because designing a sequence may take a day, and may even occupy all the video memory. In addition, if you design too many sequences, you may design exactly the same sequence. It is recommended to design only 2 at a time. If these 2 sequences are repeated or have met the requirements, there is no need to design more.

[0171] Regarding the optimization step parameter steps, according to the optimization steps recommended in RFDesign, g600, m2000 are used here, where g represents the gradient descent method and m represents MCMC, i.e., the Markov Chain Monte Carlo method. The numerical part can be adjusted in actual use, and the design time will change accordingly.

[0172] Regarding the exclude_aa parameter, it is necessary to specify the residues that are prohibited during redesign. Here, the hydrophobic residue AVILMWF that needs to be redesigned is used. Of course, new residues can also be added.

[0173] Regarding the force_aa parameter, it refers to the residues that need to be fixed. Here, all the residues in the mask represented by chain name + residue number are taken.

[0174] Regarding the parameter w_surfnp of the surface hydrophobic residue weight, it is taken as 1 here, which means adding the surface hydrophobic residue loss surfnp_loss when calculating the loss function.

[0175] Regarding the parameter w_rog of the inertia radius weight, it is taken as 1 here, which means adding the inertia radius loss rog_loss when calculating the loss function.

[0176] Regarding the parameter rog_thresh of the inertia radius threshold, it is set to 16 here.

[0177] Regarding the parameter w_rep of the combined target weight, it is set to 2 here.

[0178] For more information about RFDesign, please refer to:

[0179] https: / / github.com / RosettaCommons / RFDesign

[0180] For a more detailed introduction to the hallucination algorithm and its parameters, please refer to:

[0181] https: / / github.com / RosettaCommons / RFDesign / blob / main / hallucination / hallucinate.py

[0182] For the Layerselector program, you can refer to:

[0183] https: / / www.rosettacommons.org / docs / latest / scripting_documentation / RosettaScripts / ResidueSelectors / ResidueSelectors#residueselectors_ conformation-dependent-residue-selectors_layerselector

[0184] Example 2 Redesign of the human serum albumin crystal structure (1AO6)

[0185] The A chain of 1AO6 was designed again using the same method as in the example, and other parameters were adjusted slightly differently to obtain the second design result, namely sequence A2.

[0186] The sequence of the designed sequence A2 is as follows:

[0187] SEVAHRFKDLGEENFKALVLIAFAQYLQQCPFEDHVKLVNEVTEFAKTCVADESAENCDKSLHTLFGDKLCTVATLRETYGEMADCCAKQEPERNECFLQHKDDNPNLPRLGRPEVDTMCTNFHDNEETFGKKYLYEIARRHPY FYAPELLFFAKRYKYAFTECCQAPDKAACLLPKLDELRDEGKPSSAKQRLKCASLQKFGERAFKAWAVARLSQRFPKAEPGEVSKLVTDLTKVHTECCHGDLLECADDRADLAKYICENQDSISSKLKECCEKPCLEKSHCIAEV ENDEHPCDLPSLNPDFVESKDVCKNYTEYKDYFLGTFLYEYARRHPDYSVVLLLRLAKTYETTLEKCCQADDPHECYHKVRDEFKPLQEEPQNLIKQNCEQFEQLGEYKFQNALLVRYTKKVPQVSTPTLVEVSRNLGKVGSKC CKHPEDKRMPCAEDYLSVVLNQLCVLHEKTPVSDRVTKCCTESLVNRRPCFSSLEGDETYKPKEQNEETFTFHKDICTLSEKERQPKKQTALVELVKHKPKPTKEQLKTVMDDFCKFVEKCCKADDKETCFNEEGKKLVCSSQCA

[0188] In control example 1, the scaffold design software RFDEsign was directly used to redesign the designated region A110-582 of the A chain of 1AO6.pdb (i.e., residues 110 to 582 of the A chain) to obtain sequence A3.

[0189] Control Example 2: The ProteinMPNN program was directly used to redesign the surface hydrophobic residues in the designated region A110-582 of the A chain of 1AO6.pdb to obtain sequence A4.

[0190] Example 3 Verification of the redesigned protein sequence structure and aggregation

[0191] S3: The two sequences designed in Examples 1 and 2 were respectively predicted for structure and aggregation to verify the specific effect of the design method of the present invention. At the same time, sequences A3 and A4 were used as controls and also predicted for structure and aggregation. When predicting the structure, the structure of the A chain of natural 1AO6 was used as a reference. The specific steps are as follows:

[0192] S31: Use AlphaFold program to predict the structure of the designed protein. The A chain of 1AO6, the designed sequence A1 and A2 chains in this example, and the control A3 and A4 chains are predicted using AlphaFold program respectively, and the structure of the A chain of natural 1AO6 is used as a reference for each group of predictions.

[0193] The prediction results of A3 are as follows Figure 1 As shown, the prediction results of A4 chain are as follows Figure 2 As shown, the predicted structures of A1 and A2 are as follows Figure 3 As shown in the figure, it can be seen that compared with sequences A3 and A4 designed by other programs, sequences A1 and A2 designed by the method of the present invention are closer to the A chain structure of natural 1AO6.

[0194] Afterwards, the A1 chain designed in the embodiment was added with the residues missing at both ends and then with the complete B chain, and the structure predicted by AlphaFold was compared with 1AO6 with both A and B chains intact. Figure 4 As shown, it can be seen that the structure predicted by AlphaFold after adding the missing residues at both ends of the A chain to the A1 or A2 chain and the complete B chain is very different from the structure of 1AO6 with both A and B chains complete.

[0195] S32: Aggrescan3D program is used to predict the aggregation of the designed protein. Aggrescan3D program is used to predict and score the aggregation of the A chain of 1AO6.pdb and the A1 and A2 chains designed in Examples 1 and 2 of the present invention. The results are as follows: Figure 5-7 As shown. Figure 5 The Aggrescan3D scores of the A chain of 1AO6.pdb are shown. Figure 6 The Aggrescan3D score of the sequence A1 designed using the method of the present invention is shown. Figure 7 The Aggrescan3D score of the sequence A2 designed by the method of the present invention is shown. From the scoring results, it can be seen that the highest score and average score of the Aggrescan3D scores of the two sequences A1 and A2 designed by the present invention are lower than the A chain of 1AO6.pdb as the template. The lower the Aggrescan3D score, the lower the aggregation, indicating that the aggregation of the redesigned protein is lower than that of the original natural 1AO6.

[0196] For AlphaFold prediction structure, please refer to:

[0197] https: / / github.com / deepmind / alphafold /

[0198] For the presentation of the structure, please refer to:

[0199] https: / / www.rcsb.org / 3d-view

[0200] For Aggrescan3D score, please refer to:

[0201] https: / / academic.oup.com / nar / article / 47 / W1 / W300 / 5485072?login=false

[0202] It can be seen from the above experimental results that the design method of the present invention can obtain a protein sequence with a basically unchanged structure but significantly reduced aggregation, which shows the reliability and practicality of this method.

[0203] The above are only some embodiments of the present invention, and the present invention is not limited to the contents of the above embodiments.

[0204] All documents mentioned in the present invention are cited as references in this application, just as each document is cited as reference separately. In addition, it should be understood that after reading the above content of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims attached to this application.

Claims

1. A method for redesigning hydrophobic residues on the surface of a protein, characterized in that: The method comprises the following steps: S1: Determine the surface hydrophobic residue sites in a specified region of a specified template chain of a protein and write them into a readable file; S2: The residue site is taken as a scaffold that needs to be redesigned, and the designated region is redesigned using scaffold design software.

2. The method for redesigning hydrophobic residues on the protein surface according to claim 1, characterized in that: The S1 step includes the following steps: S11: Determination of surface hydrophobic residues within a specified region; S12: Generate information that can be read by the scaffold design software for redesign based on the information of the surface hydrophobic residues in the designated area and form the readable file, which is preferably a text file.

3. The method for redesigning hydrophobic residues on the protein surface according to claim 1, characterized in that: The S2 step is specifically as follows: Using the scaffold design software, the readable file obtained in the step S1 is read as a parameter, and the sequence design is performed on the designated region after setting the remaining parameters to obtain a redesigned protein sequence.

4. The method for redesigning hydrophobic residues on the protein surface according to claim 3, characterized in that: The scaffold design software is commonly used protein scaffold design software; preferably RFDesign.

5. The method for redesigning hydrophobic residues on the protein surface according to claim 2, characterized in that: The step S11 includes the following steps: S111: Parse the file of protein structure to be redesigned, read the residue sequence, atomic coordinates, and the list of chain name + residue number and save them into the dictionary for later use; S112: One-hot encoding for all residues; S113: Calculate a neighbor number for each residue using the main chain atomic coordinates S114: Set a neighbor number threshold; S115: determining whether it is a hydrophobic residue according to the read residue sequence, and marking the name; S116: When a residue in the designated region is a hydrophobic residue and the number of its neighbors is less than the neighbor number threshold, the residue is a surface hydrophobic residue.

6. The method for redesigning hydrophobic residues on the protein surface according to claim 2, characterized in that: The step S11 includes the following steps: S111: Parse the structure file of the protein to be redesigned, read the residue sequence, atomic coordinates, and the list of chain name + residue number and save them into the dictionary for later use; S115: determining whether it is a hydrophobic residue according to the read residue sequence, and marking the name; S112: One-hot encoding for all residues; S113: Calculate a neighbor number for each residue using main chain atomic coordinates; S114: Setting a neighbor number threshold; S116: When a residue in the designated region is a hydrophobic residue and the number of its neighbors is less than the neighbor number threshold, the residue is a surface hydrophobic residue.

7. The method for redesigning hydrophobic residues on the protein surface according to claim 5 or 6, characterized in that: The number of neighbors is calculated using LayerSelector in the ResidueSelectors program.

8. The method for redesigning hydrophobic residues on the surface of a protein according to any one of claims 1 to 7, characterized in that: The S2 step specifically involves re-mutation design of sites identified as surface hydrophobic residues in the designated region.

9. The method for redesigning hydrophobic residues on the protein surface according to claim 8, characterized in that: The step S12 comprises the following steps: S121: Generate a list according to the chain where the specified region is located and all its residue numbers. If the residue with the number is not a surface hydrophobic residue in the specified region, the element corresponding to the number is the chain name + residue number. If the residue with the number is a surface hydrophobic residue in the specified region, the element corresponding to the number is 1. S122: writing the above list into the readable file; Furthermore, S2 is re-mutated and designed according to the surface hydrophobic residues corresponding to the corresponding element 1 in the list.

10. The method for redesigning the hydrophobic residues on the protein surface according to claim 1, characterized in that: The protein has two or less peptide chains, more preferably one peptide chain.

11. The method for redesigning the hydrophobic residues on the protein surface according to claim 1, characterized in that: After step S2, a step of verifying the redesigned protein sequence is added, preferably structure prediction, aggregation prediction, protein synthesis, protein functionality prediction, etc.

12. The redesign method according to any one of claims 1 to 11, characterized in that: When the protein is a double chain, and the designated region is distributed in one chain, the chain is designated as the template chain for redesign, and the other chain is used as the target chain to verify and adjust the redesign result of the template chain to obtain a new chain; When the protein is a double chain and the designated region is distributed in two chains, one of the chains is first designated as the template chain for redesign, and the other chain is used as the target chain to verify and adjust the result of the redesign of the template chain to obtain a first new chain; then the other chain is used as the template chain for redesign, and the first new chain is used as the target chain to verify and adjust the result of the redesign of the other chain to obtain a second new chain.

13. A protein with reduced aggregation obtained by redesigning according to any one of claims 1 to 12.

14. A computer program stored in a computer-readable medium and loadable into an internal memory of a computer, comprising software code portions arranged to execute a design method according to any one of claims 1 to 12 when the program is run on a computer.

15. Use of the method for redesigning hydrophobic residues on the surface of a protein according to any one of claims 1 to 12 in protein design, wherein the protein is a natural protein or an artificial protein.

Images