Protein-protein binding affinity prediction method, computer and storage medium

Abstract

The present invention discloses a method for predicting protein-protein binding affinity, comprising: obtaining structural information of a protein-protein complex, and calculating surface information and interface information of atoms in the protein-protein complex according to the structural information; dividing the surface information and interface information of atoms in the protein-protein complex according to amino acid types; and integrating different contributions of the interface information and surface information to affinity through a linear model, a nonlinear model, a hybrid model, and a neural network model to obtain affinity information of multiple atoms in the protein-protein complex.

Description

Technical Field

[0001] The invention belongs to the field of computer technology, and in particular to a protein-protein binding affinity prediction method, a computer and a storage medium. Background Art

[0002] Protein-protein interaction refers to the process in which two or more protein molecules form a protein complex through non-covalent bonds, which constitutes the main component of the cellular biochemical reaction network.

[0003] The structure of protein-protein complex is the product and form of protein-protein interaction. Characterizing and elucidating the determinants of protein-protein interaction from a structural perspective will help to more deeply understand the mechanisms of related biological processes, more clearly understand the pathogenesis of diseases, and more efficiently design protein drugs. As one of the closely related research directions, there are more and more studies on the quantitative relationship between geometric properties and binding affinity in the structure of protein-protein complexes. For example, the micromodel for predicting protein-protein binding affinity uses the protein-protein interface area and the square of the root mean square displacement of the Cα atoms of the interface amino acids before and after binding. This model only considers interface information. The global surface model (Global Surface Model): PRODIGY (PROtein binDIngenerGY prediction) for predicting protein-protein binding affinity uses different types of interface atom pairs or interface amino acid contacts and the percentage of different types of surface amino acids. This model considers both interface and surface information, with interface information represented by the number of contacts and surface information represented by the percentage of different types of surface amino acids. A micropredictor of protein-protein binding energy uses the interface area of ​​different types of amino acids. This predictor only considers interface information and belongs to the classical interface model. An Interface Amino Acid Count Model for predicting protein-protein binding affinity. This model only considers interface information and belongs to the Classical Interface Model. A nonlinear model for predicting protein-protein affinity: LISA (Local Interaction Signal Analysis), which uses terms describing preferred and non-preferred contacts on the interface, the contribution of different types or regions of interface amino acids, non-interacting charged amino acids and the contribution of secondary structure. This model considers both interface and surface information. Interface information is characterized by preferred contacts, non-preferred contacts, different types or regions of amino acids and secondary structure, and surface information is expressed by the percentage of charged surface amino acids.

[0004] The above models cannot accurately predict protein-protein binding affinity, and cannot effectively characterize and integrate the interface and surface information of protein-protein complexes. Summary of the invention

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a method for predicting protein-protein binding affinity, comprising:

[0006] Acquiring structural information of a protein-protein complex, and calculating surface information and interface information of atoms in the protein-protein complex based on the structural information;

[0007] dividing the surface information and interface information of atoms in the protein-protein complex according to amino acid types;

[0008] The affinity information of multiple atoms in the protein-protein complex is obtained by integrating the different contributions of the interface information and the surface information to the affinity through a linear model, a nonlinear model, a hybrid model and a neural network model. Further, it includes:

[0009] Calculating the atomic surface area of ​​the protein-protein complex using the Shrake-Rupley algorithm;

[0010] The contact interface area between atoms in the protein-protein complex is calculated using the Voronoi polyhedron calculation method.

[0011] Further, including:

[0012] According to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged, the surface information is divided into: the surface area A1 of the basic amino acids of the receptor, the surface area A2 of the non-polar hydrophobic amino acids of the receptor, the surface area A3 of the polar uncharged amino acids of the receptor, the surface area A4 of the acidic amino acids of the receptor, the surface area A5 of the basic amino acids of the ligand, the surface area A6 of the non-polar hydrophobic amino acids of the ligand, the surface area A7 of the polar uncharged amino acids of the ligand, and the surface area A8 of the acidic amino acids of the ligand.

[0013] Further, including:

[0014] According to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged, the interface information is divided into: the interface area between the receptor and the ligand is A9, the interface area between the non-polar hydrophobic amino acids and the non-polar hydrophobic amino acids is A 10 , the interface area A between polar uncharged amino acids and polar uncharged amino acids 11 , the interface area A between acidic amino acids 12, the interface area A between basic amino acids and nonpolar hydrophobic amino acids 13 , the interface area A between basic amino acids and polar uncharged amino acids 14 , the interface area A between basic amino acids and acidic amino acids 15 , the interface area A between nonpolar hydrophobic amino acids and polar uncharged amino acids 16 , the interface area A between nonpolar hydrophobic amino acids and acidic amino acids 17 , the interface area A between polar uncharged amino acids and acidic amino acids 18 .

[0015] A device for predicting protein-protein binding affinity, comprising:

[0016] An acquisition module, used to acquire structural information of a protein-protein complex, and calculate surface information and interface information of atoms in the protein-protein complex according to the structural information;

[0017] A processing module, for dividing the surface information and interface information of atoms in the protein-protein complex according to amino acid types;

[0018] The execution module is used to integrate the different contributions of the interface information and the surface information to the affinity through a linear model, a nonlinear model, a hybrid model and a neural network model to obtain the affinity information of multiple atoms in the protein-protein complex.

[0019] Furthermore, it also includes:

[0020] A first processing submodule is used to calculate the atomic surface area in the protein-protein complex using a Shrake-Rupley algorithm;

[0021] The second processing submodule is used to calculate the contact interface area between atoms in the protein-protein complex by using the Voronoi polyhedron calculation method.

[0022] Furthermore, it also includes:

[0023] The third processing submodule is used to divide the surface information according to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged into: the surface area A1 of the basic amino acids of the receptor, the surface area A2 of the non-polar hydrophobic amino acids of the receptor, the surface area A3 of the polar uncharged amino acids of the receptor, the surface area A4 of the acidic amino acids of the receptor, the surface area A5 of the basic amino acids of the ligand, the surface area A6 of the non-polar hydrophobic amino acids of the ligand, the surface area A7 of the polar uncharged amino acids of the ligand, and the surface area A8 of the acidic amino acids of the ligand.

[0024] Furthermore, it also includes:

[0025] The fourth processing submodule is used to divide the interface information into the following types according to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged: the interface area A9 between the basic amino acids and the basic amino acids between the receptor and the ligand, the interface area A9 between the non-polar hydrophobic amino acids and the non-polar hydrophobic amino acids 10 , the interface area A between polar uncharged amino acids and polar uncharged amino acids 11 , the interface area A between acidic amino acids 12 , the interface area A between basic amino acids and nonpolar hydrophobic amino acids 13 , the interface area A between basic amino acids and polar uncharged amino acids 14 , the interface area A between basic amino acids and acidic amino acids 15 , the interface area A between nonpolar hydrophobic amino acids and polar uncharged amino acids 16 , the interface area A between nonpolar hydrophobic amino acids and acidic amino acids 17 , the interface area A between polar uncharged amino acids and acidic amino acids 18 .

[0026] A computer device comprises a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the method for predicting protein-protein binding affinity as described above.

[0027] A storage medium storing computer-readable instructions, when the computer-readable instructions are executed by one or more processors, causes the one or more processors to perform the steps of the method for predicting protein-protein binding affinity as described above.

[0028] The embodiment of the present invention predicts the affinity between proteins by dividing and integrating the interface and surface area of ​​proteins by the types of amino acids. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0030] Figure 1 A schematic diagram of a method for predicting protein-protein binding affinity provided by an embodiment of the present invention;

[0031] Figure 2 A two-dimensional schematic diagram of calculating the atomic surface area provided by an embodiment of the present invention;

[0032] Figure 3A two-dimensional schematic diagram of calculating the atomic interface area provided by an embodiment of the present invention;

[0033] Figure 4 A basic structural diagram of a device for predicting protein-protein binding affinity provided by an embodiment of the present invention;

[0034] Figure 5 A basic structural block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0036] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0038] Please refer to Figure 1 , Figure 1 A method for predicting protein-protein binding affinity is provided in the embodiments of the present invention. Figure 1 As shown, the method specifically comprises the following steps:

[0039] S1. Acquiring structural information of a protein-protein complex, and calculating surface information and interface information of atoms in the protein-protein complex according to the structural information;

[0040] The structural information includes: the structure file of the protein-protein complex, for example, the chain number of the receptor, the chain number of the ligand.

[0041] In an embodiment of the present invention, calculating the surface information and interface information of atoms in the protein-protein complex according to the structural information specifically includes:

[0042] Step 1, calculating the atomic surface area of ​​the protein-protein complex using the Shrake-Rupley algorithm;

[0043] Step 2: Calculate the contact interface area between atoms in the protein-protein complex using the Voronoi polyhedron calculation method.

[0044] In the embodiment of the present invention, the interface and surface information of the protein-protein complex are both described by area, and the atomic surface area and the contact interface of the interfacial amino acid pairs in the protein-protein complex are calculated by dr_sasa and Qcontact, respectively.

[0045] Among them, dr_sasa uses a modified Shrake-Rupley algorithm ('rolling ball' algorithm) to calculate the atomic surface area (solvent accessible area). Figure 2 As shown, Figure 2 This is a two-dimensional diagram of dr_sasa's calculation of atomic surface area. Figure 2 In this example, A, B, and C represent atoms (with atoms B and C acting as probes), and the radius of the circle is the van der Waals radius of the atom. The surface area of ​​atom A is composed of areas 1, 2, and 3. Areas 1 and 2 are the contact areas between atoms A and B, and areas 2 and 3 are the contact areas between atoms A and C. In order to avoid repeated calculation of area 2 when calculating the surface area of ​​atom A, the contact area between atoms A and B is the sum of area 1 and half of area 2, and the contact area between atoms A and C is the sum of area 3 and half of area 2.

[0046] Qcontact uses a method based on Voronoi polyhedra to calculate the contact area of ​​interface atoms. Figure 3 This is a two-dimensional schematic diagram of Qcontact calculating the atomic interface area. Figure 3 In the figure, a, b, c, d, e, and f represent atoms. The inner circle (yellow or green) has the van der Waals radius of the atom as its radius, and the outer circle (black or orange) has the van der Waals radius of the atom as its radius. is the radius. The contact areas between the central atom a and different neighboring atoms are the different side areas of a polyhedron. Areas 1, 2, and 3 are the contact areas between atoms a and b, c, and f, respectively. In addition, Qcontact also considers the contact areas mediated by water molecules.

[0047] S2. dividing the surface information and interface information of atoms in the protein-protein complex according to amino acid types;

[0048] According to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged, the surface information is divided into: the surface area A1 of the basic amino acids of the receptor, the surface area A2 of the non-polar hydrophobic amino acids of the receptor, the surface area A3 of the polar uncharged amino acids of the receptor, the surface area A4 of the acidic amino acids of the receptor, the surface area A5 of the basic amino acids of the ligand, the surface area A6 of the non-polar hydrophobic amino acids of the ligand, the surface area A7 of the polar uncharged amino acids of the ligand, and the surface area A8 of the acidic amino acids of the ligand.

[0049] According to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged, the interface information is divided into: the interface area between the receptor and the ligand is A9, the interface area between the non-polar hydrophobic amino acids and the non-polar hydrophobic amino acids is A 10 , the interface area A between polar uncharged amino acids and polar uncharged amino acids 11 , the interface area A between acidic amino acids 12 , the interface area A between basic amino acids and nonpolar hydrophobic amino acids 13 , the interface area A between basic amino acids and polar uncharged amino acids 14 , the interface area A between basic amino acids and acidic amino acids 15 , the interface area A between nonpolar hydrophobic amino acids and polar uncharged amino acids 16 , the interface area A between nonpolar hydrophobic amino acids and acidic amino acids 17 , the interface area A between polar uncharged amino acids and acidic amino acids 18 .

[0050] S3. Integrate the different contributions of the interface information and the surface information to the affinity through a linear model, a nonlinear model, a hybrid model and a neural network model to obtain the affinity information of multiple atoms in the protein-protein complex.

[0051] The embodiments of the present invention use linear models, nonlinear models and hybrid models to integrate interface and surface information to predict protein-protein binding affinity. At present, 60 models are selected as representatives from the trained or constructed models to integrate different interface and surface areas to predict protein-protein binding affinity. These 60 representative models include:

[0052] 1. 12 linear models: The specific forms are shown in Table 1, K is the dissociation constant or inhibition constant, A1, A2…A18 are different interfaces and surface areas.

[0053] Table 1. Forms of the 12 linear models

[0054]

[0055]

[0056] 2. 6 constructed nonlinear models: The specific forms are shown below.

[0057]

[0058] log(K)=-0.3042859089144x-4.313861052

[0059]

[0060] log(K)=6173.14626206173x-10.9451091646642

[0061]

[0062] log(K)=-0.00194823682341253x-3.60632288257436

[0063]

[0064] log(K)=-0.000995978533158535x-3.40762240628879

[0065]

[0066] log(K)=-0.000297405827677228x-3.10137180970196

[0067]

[0068] log(K)=-0.0494567261584352x+1.02656054246774

[0069] 3. 5 mixed models based on the constructed nonlinear model: The specific form is shown below.

[0070]

[0071] log(K)=2123.51554728111x1-0.00076154614x2-0.009117371x3-6.81901

[0072]

[0073] log(K)=11101.3438625x1-0.0024382248x2-10.14654806

[0074]

[0075] log(K)=9324.87836043513x1-1.0264057483x2-0.02888991233x3

[0076] -29.8937019x4+0.000038582x5+0.6532806x6

[0077] -0.000086271984x7-10.96338

[0078]

[0079] log(K)=10653.84007381x1-0.844635959x2-0.663819162574532x3

[0080] +0.010170716x4-56.13299101x5+1.6289367716x6

[0081] -0.018383451x7-11.3611942

[0082]

[0083] log(K)=8448.04346688343x1-0.00133007562587674x2

[0084] -0.473048255651898x3+0.199215088637466x4

[0085] -0.03044412798198x5-1756.51996145466x6

[0086] -0.00000825159763657977x7-10.0642209228433

[0087] Among them, A1, A2…A 18 are different interfaces and surface areas, A 19 , A 20 and A 21 They are the total receptor surface area, the total ligand surface area, and the total interface area in the protein-protein complex, respectively.

[0088] 4. 11 trained nonlinear models (with explicit expressions): The specific form is shown below.

[0089]

[0090] log(K)=-0.568888001101073x+0.901517361877311

[0091]

[0092] log(K)=-0.166074171452944x-0.0116527283887838

[0093]

[0094] log(K)=-0.210565325409778x+0.0608136869974672

[0095]

[0096] log(K)=-0.00515262504749831x-3.94782837626315

[0097]

[0098] log(K)=-0.0803933862888868x-4.03271669648362

[0099]

[0100] log(K)=-0.0227956750271275x-4.02948970526830

[0101]

[0102] log(K)=-0.0864140410834600x+0.292459454308228

[0103]

[0104] log(K)=-0.128259073621144x+0.241234090799209

[0105]

[0106] log(K)=-0.0286352567962234x+0.269415725339909

[0107]

[0108] log(K)=-0.0392121632941331x+0.257453328050034

[0109]

[0110] log(K)=-0.0456985331945409x+0.277608285189444

[0111] 5. A hybrid model of 5 trained nonlinear models: The specific form is shown below.

[0112]

[0113]

[0114] log(K)=-0.00972554008088431x1-0.00309176968287047x2-6.04187921736417

[0115]

[0116]

[0117]

[0118]

[0119] log(K)=-2.77332582241019x1+2.20971087799709x2

[0120] -0.0504913337487757x3-0.199007425298938x4+7.00208678164769

[0121]

[0122]

[0123] log(K)=-0.111577505746261x1-0.0150171443156653x2+0.675198126328139

[0124]

[0125]

[0126]

[0127] log(K)=-2.63084468100610x1

[0128] +2.15374864513813x2-0.0528842342604257x3

[0129] +4.70482049561711

[0130]

[0131]

[0132]

[0133] log(K)=-1.90921650787848x1+2.07530866172609x2

[0134] -0.0737497215603200x3+4.75625877556473

[0135] Among them, A1, A2…A18 are different interface and surface areas, and A19, A20 and A21 are the total surface area of ​​the receptor, the total surface area of ​​the ligand and the total interface area in the protein-protein complex, respectively.

[0136] Note: According to the processing method in training, when a certain type of area does not exist in the complex, the size of this area will be set to 1. In addition, in the Qcontact calculation results, the contact area of ​​each interface amino acid pair appears twice. Therefore, the interface area used by the 16 models listed above and the models trained by neural networks (and hybrid models based on models trained by neural networks) is twice the real interface area. The interface areas used by 12 linear models, 6 constructed nonlinear models, and 5 hybrid models based on constructed nonlinear models have been corrected to the real interface area. After correcting the corresponding coefficients, the prediction results of these models are the same as those of the original models.

[0137] 6. 18 nonlinear models (trained with neural networks, without explicit expressions): no specific form.

[0138] 7. 3 hybrid models (non-linear models based on neural network training): no specific form.

[0139] The present invention predicts protein-protein binding affinity by classifying and integrating interface and surface areas by amino acid type. The present invention may have a certain impact on the area-energy relationship of protein-protein interaction, interface area pairing rules, surface area contribution, etc.

[0140] Example 1 Protein-protein complex (PDB code: 1acb)

[0141] #Enter the scripts folder

[0142] #Calculate 18 different interfaces and surface areas. calculate_sasa_and_contact.pl is a perl script, 1acb.pdb is the structure file of the protein-protein complex, E is the chain number of the receptor, I is the chain number of the ligand, and output is the output file, which saves 18 different interfaces and surface areas of the complex.

[0143] #Based on the output file output in the previous step, predict protein-protein binding affinity. matlab-nodesktop-nosplash-r predict_binding_affinity is to run the matlab program predict_binding_affinity.m in the command line, and output the file predicted_affinity_60, which saves the prediction results of 60 models.

[0144] #Rename the file output storing 18 different interfaces and surface areas to 1acb.pdb.area

[0145] #Rename the file predicted_affinity_60 storing 60 model prediction results to 1acb.pdb.predicted_affinity_60

[0146] Example 2 Protein-Protein Complex (PDB code: 1a2k)

[0147] #Enter the scripts folder

[0148] #Calculate 18 different interfaces and surface areas. calculate_sasa_and_contact.pl is a perl script, 1a2k.pdb is a protein-protein complex structure file, A, B are the chain numbers of the receptor, C is the chain number of the ligand, and output is the output file, which saves 18 different interfaces and surface areas of the complex

[0149] #Based on the output file output in the previous step, predict protein-protein binding affinity. matlab-nodesktop-nosplash-r predict_binding_affinity is to run the matlab program predict_binding_affinity.m in the command line, and output the file predicted_affinity_60, which saves the prediction results of 60 models.

[0150] #Rename the file output storing 18 different interfaces and surface areas to 1a2k.pdb.area

[0151] #Rename the file predicted_affinity_60 storing 60 model prediction results to 1a2k.pdb.predicted_affinity_60.

[0152] Example 3 Protein-protein complex (PDB code: 1akj)

[0153] #Enter the scripts folder

[0154] #Calculate 18 different interfaces and surface areas. calculate_sasa_and_contact.pl is a perl script, 1akj.pdb is the structure file of protein-protein complex, A, B are the chain numbers of receptor, D, E are the chain numbers of ligand, output is the output file, which saves 18 different interfaces and surface areas of the complex.

[0155] # Based on the output file output of the previous step, predict protein-protein binding affinity. matlab-nodesktop-nosplash-r predict_binding_affinity is to run the matlab program predict_binding_affinity.m in the command line, and output the file predicted_affinity_60, which saves the prediction results of 60 models

[0156] #Rename the file output storing 18 different interfaces and surface areas to 1akj.pdb.area

[0157] #Rename the file predicted_affinity_60 storing 60 model prediction results to 1akj.pdb.predicted_affinity_60.

[0158] like Figure 4As shown, in order to solve the above problems, an embodiment of the present invention further provides a prediction device for protein-protein binding affinity, including: an acquisition module 2100, a processing module 2200 and an execution module 2300, wherein the acquisition module 2100 is used to acquire structural information of a protein-protein complex, and calculate surface information and interface information of atoms in the protein-protein complex according to the structural information; the processing module 2200 is used to divide the surface information and interface information of atoms in the protein-protein complex according to amino acid types; the execution module 2300 is used to integrate different contributions of the interface information and surface information to affinity through a linear model, a nonlinear model, a hybrid model and a neural network model, so as to obtain affinity information of multiple atoms in the protein-protein complex.

[0159] In some embodiments, the method further includes: a first processing submodule for calculating the atomic surface area in the protein-protein complex using the Shrake-Rupley algorithm; and a second processing submodule for calculating the contact interface area between atoms in the protein-protein complex using the Voronoi polyhedron calculation method.

[0160] In some embodiments, it also includes: a third processing submodule, which is used to divide the surface information according to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged into: the surface area A1 of the basic amino acids of the receptor, the surface area A2 of the non-polar hydrophobic amino acids of the receptor, the surface area A3 of the polar uncharged amino acids of the receptor, the surface area A4 of the acidic amino acids of the receptor, the surface area A5 of the basic amino acids of the ligand, the surface area A6 of the non-polar hydrophobic amino acids of the ligand, the surface area A7 of the polar uncharged amino acids of the ligand, and the surface area A8 of the acidic amino acids of the ligand.

[0161] In some embodiments, the fourth processing submodule is further included, which is used to divide the interface information into the following types according to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged: the interface area A9 between the basic amino acids and the basic amino acids between the receptor and the ligand, the interface area A9 between the non-polar hydrophobic amino acids and the non-polar hydrophobic amino acids, and the interface area A9 between the non-polar hydrophobic amino acids and the non-polar hydrophobic amino acids. 10 , the interface area A between polar uncharged amino acids and polar uncharged amino acids 11 , the interface area A between acidic amino acids 12 , the interface area A between basic amino acids and nonpolar hydrophobic amino acids 13 , the interface area A between basic amino acids and polar uncharged amino acids 14 , the interface area A between basic amino acids and acidic amino acids 15 , the interface area A between nonpolar hydrophobic amino acids and polar uncharged amino acids 16 , the interface area A between nonpolar hydrophobic amino acids and acidic amino acids17 , the interface area A between polar uncharged amino acids and acidic amino acids 18 .

[0162] To solve the above technical problems, the embodiment of the present invention also provides a computer device. Figure 5 , Figure 5 This is a basic structural block diagram of the computer device in this embodiment.

[0163] like Figure 5 As shown in Figure 1, the internal structure diagram of a computer device. Figure 5 As shown, the computer device includes a processor, a non-volatile storage medium, a memory and a network interface connected via a system bus. Among them, the non-volatile storage medium of the computer device stores an operating system, a database and computer-readable instructions, and the database may store a control information sequence. When the computer-readable instructions are executed by the processor, the processor can implement an image processing method. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute an image processing method. The network interface of the computer device is used to connect and communicate with a terminal. Those skilled in the art can understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0164] In this embodiment, the processor is used to execute Figure 4 The specific contents of the acquisition module 2100, the processing module 2200 and the execution module 2300 are stored in the memory, and the program code and various data required to execute the above modules are stored in the memory. The network interface is used to transmit data between the user terminal or the server. The memory in this embodiment stores the program code and data required to execute all sub-modules in the image processing method, and the server can call the program code and data of the server to execute the functions of all sub-modules.

[0165] The computer device provided in the embodiment of the present invention predicts protein-protein binding affinity by dividing and integrating interface and surface areas by amino acid type. The present invention may have a certain impact on the area-energy relationship of protein-protein interaction, interface area pairing rules, surface area contribution, etc.

[0166] The present invention also provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, enables the one or more processors to execute the steps of the method for predicting protein-protein binding affinity described in any of the above embodiments.

[0167] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0168] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0169] The above descriptions are only some embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for predicting protein-protein binding affinity, characterized in that: include: Acquiring structural information of a protein-protein complex, and calculating surface information and interface information of atoms in the protein-protein complex based on the structural information; dividing the surface information and interface information of atoms in the protein-protein complex according to amino acid types; The affinity information of multiple atoms in the protein-protein complex is obtained by integrating the different contributions of the interface information and the surface information to the affinity through linear models, nonlinear models, hybrid models and neural network models.

2. The prediction method according to claim 1, characterized in that: include: Calculating the atomic surface area of ​​the protein-protein complex using the Shrake-Rupley algorithm; The contact interface area between atoms in the protein-protein complex is calculated using the Voronoi polyhedron calculation method.

3. The prediction method according to claim 1, characterized in that: include: According to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged, the surface information is divided into: the surface area A1 of the basic amino acids of the receptor, the surface area A2 of the non-polar hydrophobic amino acids of the receptor, the surface area A3 of the polar uncharged amino acids of the receptor, the surface area A4 of the acidic amino acids of the receptor, the surface area A5 of the basic amino acids of the ligand, the surface area A6 of the non-polar hydrophobic amino acids of the ligand, the surface area A7 of the polar uncharged amino acids of the ligand, and the surface area A8 of the acidic amino acids of the ligand.

4. The prediction method according to claim 1, characterized in that: include: According to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged, the interface information is divided into: the interface area between the receptor and the ligand is A9, the interface area between the non-polar hydrophobic amino acids and the non-polar hydrophobic amino acids is A 10 , the interface area A between polar uncharged amino acids and polar uncharged amino acids 11 , the interface area A between acidic amino acids 12 , the interface area A between basic amino acids and nonpolar hydrophobic amino acids 13 , the interface area A between basic amino acids and polar uncharged amino acids 14 , the interface area A between basic amino acids and acidic amino acids 15 , the interface area A between nonpolar hydrophobic amino acids and polar uncharged amino acids 16 , the interface area A between nonpolar hydrophobic amino acids and acidic amino acids 17 , the interface area A between polar uncharged amino acids and acidic amino acids 18 .

5. A device for predicting protein-protein binding affinity, characterized in that: include: An acquisition module, used to acquire structural information of a protein-protein complex, and calculate surface information and interface information of atoms in the protein-protein complex according to the structural information; A processing module, for dividing the surface information and interface information of atoms in the protein-protein complex according to amino acid types; The execution module is used to integrate the different contributions of the interface information and the surface information to the affinity through a linear model, a nonlinear model, a hybrid model and a neural network model to obtain the affinity information of multiple atoms in the protein-protein complex.

6. The device according to claim 5, characterized in that Also includes: A first processing submodule is used to calculate the atomic surface area in the protein-protein complex using a Shrake-Rupley algorithm; The second processing submodule is used to calculate the contact interface area between atoms in the protein-protein complex by using the Voronoi polyhedron calculation method.

7. The device according to claim 6, characterized in that Also includes: The third processing submodule is used to divide the surface information according to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged into: the surface area A1 of the basic amino acids of the receptor, the surface area A2 of the non-polar hydrophobic amino acids of the receptor, the surface area A3 of the polar uncharged amino acids of the receptor, the surface area A4 of the acidic amino acids of the receptor, the surface area A5 of the basic amino acids of the ligand, the surface area A6 of the non-polar hydrophobic amino acids of the ligand, the surface area A7 of the polar uncharged amino acids of the ligand, and the surface area A8 of the acidic amino acids of the ligand.

8. The device according to claim 6, characterized in that Also includes: The fourth processing submodule is used to divide the interface information into the following types according to the polarity, hydrophobicity, acidity and alkalinity of the amino acids and whether they are charged: the interface area A9 between the basic amino acids and the basic amino acids between the receptor and the ligand, the interface area A9 between the non-polar hydrophobic amino acids and the non-polar hydrophobic amino acids 10 , the interface area A between polar uncharged amino acids and polar uncharged amino acids 11 , the interface area A between acidic amino acids 12 , the interface area A between basic amino acids and nonpolar hydrophobic amino acids 13 , the interface area A between basic amino acids and polar uncharged amino acids 14 , the interface area A between basic amino acids and acidic amino acids 15 , the interface area A between nonpolar hydrophobic amino acids and polar uncharged amino acids 16 , the interface area A between nonpolar hydrophobic amino acids and acidic amino acids 17 , the interface area A between polar uncharged amino acids and acidic amino acids 18 .

9. A computer device comprising a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor performs the step of predicting protein-protein binding affinity according to any one of claims 1 to 4.

10. A storage medium storing computer-readable instructions, wherein when the computer-readable instructions are executed by one or more processors, the one or more processors are caused to perform the step of predicting protein-protein binding affinity according to any one of claims 1 to 4.

Images