Dehydrogenase thermal stability optimization method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of biological information, in particular to a dehydrogenase thermal stability optimization method and device, electronic equipment and a storage medium. According to the dehydrogenase thermal stability optimization method and device, the electronic equipment and the storage medium provided by the embodiment of the invention, the temperature-sensitive target amino acid residues in the three-dimensional structure of the dehydrogenase are determined according to the state data at different temperatures; and predicting the thermal stability change of the target amino acid residue before and after mutation through the pre-trained protein thermal stability prediction model, so that the prediction efficiency and prediction accuracy of the dehydrogenase mutant with improved thermal stability can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of biological information, and in particular to a dehydrogenase thermal stability optimization method and device, an electronic device and a storage medium. BACKGROUND

[0002] At present, dehydrogenase is a key biological catalyst for redox reaction. Under the dual driving of biological evolution and protein engineering modification, it forms a unique catalytic mechanism. Dehydrogenase has precise substrate specificity recognition ability and stereoisomer selectivity characteristics, and plays an irreplaceable important role in metabolic regulation, biosynthesis and industrial biological catalysis and other key fields. However, most natural dehydrogenases have significant thermal sensitivity. The tertiary structure of dehydrogenase is prone to irreversible denaturation under high temperature conditions, which is manifested as changes in active site conformation, attenuation of coenzyme binding ability and tendency of multi-subunit complex depolymerization. These structural instability phenomena directly lead to a sharp decline in catalytic efficiency. Especially in the industrial catalysis scene that requires high temperature conditions (such as high-temperature fermentation system, non-aqueous phase chemical production, etc.), the thermal instability of dehydrogenase has become a major technical bottleneck restricting its large-scale application.

[0003] Therefore, how to improve the thermal stability of dehydrogenase has become a technical problem to be solved in the field. SUMMARY

[0004] In view of the above problems, the embodiments of the present application provide a dehydrogenase thermal stability optimization method and device, an electronic device and a storage medium to solve the technical problem of how to improve the thermal stability of dehydrogenase.

[0005] In a first aspect, the embodiments of the present application provide a dehydrogenase thermal stability optimization method, comprising: obtaining a dehydrogenase three-dimensional structure, and performing molecular dynamics simulation on the dehydrogenase three-dimensional structure at different temperatures to obtain state change data of the dehydrogenase three-dimensional structure during the molecular dynamics simulation at different temperatures; determining target amino acid residues sensitive to temperature in the dehydrogenase three-dimensional structure according to the state data at different temperatures; performing saturation mutation on the target amino acid residues to obtain dehydrogenase mutant data; inputting the dehydrogenase mutant data into a pre-trained protein thermal stability prediction model to output a thermal stability prediction result of the corresponding dehydrogenase mutant, and obtaining a target dehydrogenase mutant according to the thermal stability prediction result; wherein the protein thermal stability prediction model is obtained by training preset protein mutation data and thermal stability data thereof.

[0006] In a second aspect, the embodiments of the present application provide a dehydrogenase thermal stability optimization device, comprising: The structural data acquisition module is configured to acquire a dehydrogenase three-dimensional structure, perform molecular dynamics simulation on the dehydrogenase three-dimensional structure at different temperatures, and acquire state change data of the dehydrogenase three-dimensional structure during the molecular dynamics simulation at different temperatures. The mutation target determination module is configured to determine target amino acid residues sensitive to temperature in the dehydrogenase three-dimensional structure according to the state data at different temperatures. The mutant data generation module is configured to perform saturated mutation on the target amino acid residues and acquire a plurality of dehydrogenase mutant data. The prediction module is configured to input the dehydrogenase mutant data into a pre-trained protein thermal stability prediction model, output a thermal stability prediction result of a corresponding dehydrogenase mutant, and acquire a target dehydrogenase mutant according to the thermal stability prediction result. The protein thermal stability prediction model is obtained by training a preset protein and thermal stability data of the protein.

[0007] In a third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory coupled with the processor, the memory storing program instructions executable by the processor; and the processor executes the program instructions stored in the memory to implement the dehydrogenase thermal stability optimization method described above.

[0008] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, the computer-readable storage medium storing program instructions, the program instructions being executed by a processor to implement the dehydrogenase thermal stability optimization method described above.

[0009] The dehydrogenase thermal stability optimization method, device, electronic device, and storage medium provided by the embodiments of the present application can determine target amino acid residues sensitive to temperature in a dehydrogenase three-dimensional structure according to state data at different temperatures, and predict changes in thermal stability of the target amino acid residues before and after mutation by using a pre-trained protein thermal stability prediction model, thereby improving the prediction efficiency and prediction accuracy of dehydrogenase mutants with improved thermal stability.

[0010] These aspects and other aspects of the present application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 A flow scenario diagram of the dehydrogenase thermal stability optimization method provided by the embodiments of the present application is shown.

[0012] Figure 2 A schematic diagram of a dehydrogenase three-dimensional structure in the embodiments of the present application is shown.

[0013] Figure 3A comparison chart of RMSD variation curves of the dehydrogenase in the embodiments of the present application and the dehydrogenase mutants is shown.

[0014] Figure 4 A comparison chart of prediction accuracy of different protein thermal stability prediction tools is shown.

[0015] Figure 5 A structural schematic diagram of the dehydrogenase thermal stability optimization device provided in the embodiments of the present application is shown.

[0016] Figure 6 A structural schematic diagram of the electronic device provided in the embodiments of the present application is shown.

[0017] Figure 7 A structural schematic diagram of the storage medium provided in the embodiments of the present application is shown. DETAILED DESCRIPTION

[0018] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the drawings, in which the same or similar notations represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below are exemplary only, and are used only for explaining the present application, and cannot be understood as limiting the present application.

[0019] In order to enable persons skilled in the art to better understand the schemes of the present application, the technical schemes in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without making creative efforts fall within the scope of protection of the present application.

[0020] In the embodiments of the present application, it should be noted that, in this document, relational terms such as first and second and the like are used solely to distinguish one entity or action from another entity or action, without necessarily requiring or implying any actual such relationship or order between or among the entities or actions.

[0021] Moreover, the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusions, so that processes, methods, articles, or devices including a series of elements not only include those elements, but also include other elements not explicitly listed, or further include elements inherent in such processes, methods, articles, or devices. Without more limitations, the element defined by the phrase "including a" does not exclude the presence of additional identical elements in the process, method, article, or device including the element.

[0022] In the description of the embodiments of the present application, the words "example" or "for example" or similar words are used to represent that example, illustration, or description. Any embodiment or design scheme described as "example" or "for example" in the embodiments of the present application is not interpreted as more preferred or having more advantages than another embodiment or design scheme. The words "example" or "for example" or similar words are used to present the relative concept in a clear manner.

[0023] In addition, "multiple" in the embodiments of the present application refers to two or more than two, and therefore "multiple" in the embodiments of the present application can also be understood as "at least two". "At least one" can be understood as one or more, for example, understood as one, two or more. For example, including at least one means including one, two or more, and does not limit which ones are included, for example, including at least one of A, B and C, and the included can be A, B, C, A and B, A and C, B and C, or A and B and C.

[0024] It should be noted that in the embodiments of the present application, the association relationship of the associated objects described by "and / or" can represent three kinds of relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone. In addition, the character " / ", if not specially stated, generally represents a "or" relationship between the associated objects before and after.

[0025] It should be noted that in the embodiments of the present application, "connection" can be understood as electrical connection, and the connection between two electrical elements can be direct or indirect connection between the two electrical elements. For example, A and B are connected, which can be direct connection between A and B, or indirect connection between A and B through one or more other electrical elements.

[0026] Proteins can only play specific functions by adopting specific three-dimensional structures, and the ability to fold into corresponding three-dimensional structures is affected by the stability of the protein.

[0027] An embodiment of the present application provides a method for optimizing the thermal stability of a dehydrogenase, please refer to Figure 1 As shown in the figure, the method for optimizing the thermal stability of the dehydrogenase includes the following steps S11-S14: Step S11: Obtain the three-dimensional structure of the dehydrogenase, and perform molecular dynamics simulation on the three-dimensional structure of the dehydrogenase at different temperatures to obtain state change data of the three-dimensional structure of the dehydrogenase during the molecular dynamics simulation at different temperatures.

[0028] The tool for the molecular dynamics simulation can be, for example, GROMACS (GROningen MAchine for Chemical Simulations). GROMACS supports various simulation algorithms such as molecular dynamics, energy minimization, conformational search, and free energy calculation, and provides functions such as molecular construction, force field parameterization, simulation setting, post-processing, and analysis. The state change data can include, for example, the change of atomic coordinates over time, the structural change (secondary structure change, conformation change, etc.) of the dehydrogenase three-dimensional structure, the kinetic parameters (velocity, kinetic energy, potential energy, and temperature of atoms, etc.), viscosity, and diffusion coefficient. Exemplarily, the state data can include a trajectory file generated by GROMACS. In some embodiments, the number of hydrogen bonds, the root mean square deviation, and / or the root mean square fluctuation can be calculated based on the state change data.

[0029] A plurality of simulation temperatures can be set, and the molecular dynamics simulation of the dehydrogenase three-dimensional structure is performed at each simulation temperature to obtain the state change data at different temperatures.

[0030] Exemplarily, the dehydrogenase three-dimensional structure is as shown in FIG. 1. Figure 2

[0031] Step S12: determining the target amino acid residues in the dehydrogenase three-dimensional structure that are temperature-sensitive according to the state data at different temperatures.

[0032] In this case, the target index of each amino acid residue can be calculated based on the state data at the corresponding temperature, and then whether each amino acid residue is temperature-sensitive can be determined based on the target index of the amino acid residue at different temperatures. The amino acid residues that are temperature-sensitive are used as the target for mutation screening, so as to improve the thermal stability of the dehydrogenase by mutation of the target amino acid residues.

[0033] Step S13: performing saturated mutation on the target amino acid residues to obtain a plurality of dehydrogenase mutant data.

[0034] In this case, for each target amino acid residue, it is replaced by other amino acid residues, and the “other amino acid” is not limited as long as it is different from the amino acid corresponding to each position. Specifically, the other amino acid in the embodiments of the present application can be one or more amino acids selected from the following: non-polar amino acids glycine G, alanine A, valine V, leucine L, isoleucine I, methionine M, phenylalanine F, tryptophan W, and proline P; neutral polar amino acids serine S, threonine T, cysteine C, tyrosine Y, asparagine N, glutamine Q, and histidine H; acidic amino acids aspartic acid D and glutamic acid E; and basic amino acids lysine K and arginine R, but is not limited thereto.

[0035] ​Wherein at least one target amino acid residue is replaced by other amino acid residues to obtain a dehydrogenase mutant, and a corresponding dehydrogenase mutant data is constructed according to the dehydrogenase mutant.

[0036] Step S14: inputting the dehydrogenase mutant data into the pre-trained protein thermal stability prediction model to output a thermal stability prediction result of the corresponding dehydrogenase mutant, and obtaining the target dehydrogenase mutant according to the thermal stability prediction result.

[0037] Wherein the thermal stability prediction result can be used to reflect the influence degree of the mutation of at least one target amino acid residue on the thermal stability of the dehydrogenase, and the influence degree of the thermal stability may, for example, be significantly improved thermal stability, improved thermal stability, flat thermal stability, reduced thermal stability, and significantly reduced thermal stability. The dehydrogenase mutant with a thermal stability prediction result reflecting significantly improved thermal stability or improved thermal stability can be used as the target dehydrogenase mutant.

[0038] Wherein the protein thermal stability prediction model is obtained by training the preset protein mutation data and its thermal stability data. Specifically, the preset protein mutation data can include the amino acid sequence of the protein, the three-dimensional structure of the protein, and the mutation information of at least one target amino acid residue, which may, for example, include the position of the target amino acid residue and the mutated amino acid residue; and the thermal stability data may, for example, include the change amount of the folding free energy before and after mutation, which may, for example, be the difference between the folding free energy of the protein before mutation and the folding free energy of the protein after mutation as the change amount of the folding free energy before and after mutation (ΔΔG) when constructing the training data.

[0039] In this embodiment, the temperature-sensitive target amino acid residue in the dehydrogenase three-dimensional structure is determined according to the state data at different temperatures, and the change of the thermal stability of the target amino acid residue before and after mutation is predicted by the pre-trained protein thermal stability prediction model, which can improve the prediction efficiency and accuracy of the dehydrogenase mutant with improved thermal stability.

[0040] As an implementation manner, step S12 specifically includes the following steps: Step S21: for each temperature, calculating the RMSF of each amino acid residue in the dehydrogenase three-dimensional structure according to the state data at the corresponding temperature.

[0041] The target index can be RMSF (Root Mean Square Fluctuation), and the RMSF value of each atom can be calculated by using the GROMACS software calculation tool, specifying a trajectory file, a reference structure file, and an output file, and the software can automatically calculate the RMSF value of each atom. Then, according to the index of the amino acid residue, the RMSF values of all atoms in the residue are averaged to obtain the RMSF value of each residue.

[0042] Step S22: obtaining a candidate target amino acid residue from the amino acid residues according to the RMSF of each amino acid residue at different temperatures.

[0043] The RMSF of the amino acid residue at different temperatures can reflect the difference of the amino acid residue at different temperatures. If the RMSF at different temperatures is large, it means that the amino acid residue is sensitive to temperature. For example, the RMSF maximum value and the RMSF minimum value of the amino acid residue at different temperatures can be determined according to the RMSF of the amino acid residue at different temperatures. The range is calculated according to the difference between the RMSF maximum value and the RMSF minimum value. The difference between the RMSF of the amino acid residue at different temperatures is characterized by the range. The amino acid residue with a range greater than a preset range threshold is taken as a candidate target amino acid residue. Alternatively, the standard deviation of the RMSF can be calculated according to the RMSF of the amino acid residue at different temperatures. The difference between the RMSF of the amino acid residue at different temperatures is characterized by the standard deviation. The amino acid residue with a standard deviation greater than a preset standard deviation threshold is taken as a candidate target amino acid residue.

[0044] Step S23: obtaining the distance between the candidate target amino acid residue and the substrate, and taking the candidate target amino acid residue with a distance greater than a first preset distance threshold and located in the loop region as a target amino acid residue.

[0045] The molecular structure of the substrate of the dehydrogenase can be obtained, the three-dimensional structure of the substrate can be constructed according to the molecular structure of the substrate, the three-dimensional structure of the dehydrogenase-substrate complex can be obtained according to the three-dimensional structure of the dehydrogenase and the molecular structure of the substrate, and the distance between the candidate target amino acid residue and the substrate can be calculated based on the three-dimensional structure of the dehydrogenase-substrate complex. For example, the distance between the candidate target amino acid residue and the substrate can be the distance between the alpha carbon atom of the candidate target amino acid residue and the center of the substrate. The distance between the candidate target amino acid residue and the substrate can reflect the influence of the candidate target amino acid residue on the binding of the substrate. For example, the first preset distance can be 6.0 Å.

[0046] The dehydrogenase-substrate complex three-dimensional structure can show the active center of the enzyme, the catalytic residues, the active pocket of the enzyme, the binding site of the substrate, the domain of the enzyme, the loop region of the enzyme, etc. Whether the candidate target amino acid residue is located in the loop region can be determined by the dehydrogenase three-dimensional structure or the dehydrogenase-substrate complex three-dimensional structure.

[0047] The target amino acid residue is close to the substrate, which may have a certain influence on the binding of the substrate, and the target amino acid residue is located in the loop region, which may have a certain influence on the folding of the enzyme.

[0048] As an implementation manner, the dehydrogenase mutant data includes the amino acid sequence of the dehydrogenase, the three-dimensional structure of the dehydrogenase, and the mutation information of at least one target amino acid residue. The protein thermal stability prediction model can be a neural network model, which includes a first feature extraction network, a second feature extraction network, a third feature extraction network, and a prediction network. Step S13 specifically includes the following steps: Step S31: obtaining the amino acid sequence of the dehydrogenase mutant according to the amino acid sequence of the dehydrogenase and the mutation information of at least one target amino acid residue.

[0049] Each dehydrogenase mutant data corresponds to the amino acid sequence of a dehydrogenase mutant.

[0050] Step S32: using the first feature extraction network to extract the first sequence feature of the amino acid sequence of the dehydrogenase and the second sequence feature of the amino acid sequence of the dehydrogenase mutant, respectively, and splicing the first sequence feature and the second sequence feature to obtain the first extracted feature.

[0051] The first feature extraction network can learn the sequence features of the amino acid sequence before mutation and the sequence features of the amino acid sequence after mutation, respectively.

[0052] Exemplarily, the first feature extraction network can be a deep learning network based on the Transformer architecture, which can capture long-range dependencies and evolutionary information in the amino acid sequence and generate high-dimensional protein feature vectors. Exemplarily, the protein semantic network can be an ESM (Evolutionary Scale Modeling) model, for example, an ESM-1bTransformer model.

[0053] Exemplarily, the first feature extraction network can also use other pre-trained models to extract the first sequence feature and the second sequence feature. The pre-trained models can be at least one of UniRef, ProteinBert, TAPE, ProtGPT2, ProtTXL, ProtBert, ProtXLNet, ProtAlbert, ProtElectra, ProtT5-XL, and ProtT5-XXL.

[0054] Step S33: Using the second feature extraction network, the second extraction feature is obtained according to the dehydrogenase three-dimensional structure and the mutation information of the at least one target amino acid residue.

[0055] The second extraction feature is a structure feature related to the mutation site, and the second feature extraction network can learn the structure information near the target amino acid residue. Exemplarily, the second extraction feature can be constructed according to at least one of other amino acid residues near the target amino acid residue, secondary structure near the target amino acid residue, and structure of the loop region where the target amino acid residue is located. The second feature network can learn the three-dimensional structure feature of the temperature-sensitive amino acid residue.

[0056] Step S34: Using the third extraction network, the third extraction feature is obtained according to the mutation information of the at least one target amino acid residue.

[0057] The third extraction feature can be the sum of the features of the mutated amino acid residues. The feature vector of the mutated amino acid residue can be constructed according to at least one description information of the mutated amino acid residue, and then the feature vectors of all the mutated amino acid residues are spliced to form the third extraction feature. The description information is used to describe the physicochemical characteristics, three-dimensional structure characteristics, topological characteristics, evolutionary characteristics, or biological function characteristics of the mutated amino acid residue. The third extraction network can learn the features of the mutated amino acid residue.

[0058] Exemplarily, the description information of the mutated amino acid residue can include at least one of VHSE, PSSM, T-scal, ST-scal or Z-scal. Specifically, VHSE (Van der Waals, Hydrophobic, Steric, Electrostatic) is a physical and chemical property descriptor, which is classified based on the characteristics of van der Waals force, hydrophobicity, steric hindrance and electrostatic interaction of amino acids; PSSM (Position-Specific Scoring Matrix) is a sequence conservation descriptor, which is generated by multiple sequence alignment and reflects the conservation of amino acids at a specific position; T-scal (Topological Scaling) is a topological structure descriptor, which is quantified based on the topological characteristics (such as contact distance, secondary structure propensity) of amino acids in the three-dimensional structure of the protein; ST-scal (Secondary Structure Scaling) is a secondary structure propensity descriptor, which measures the propensity of amino acids to form secondary structures such as alpha-helix and beta-sheet; Z-scal (Z-score Scaling) is a statistical standardization descriptor, which standardizes the physical and chemical properties (such as hydrophobicity, charge) of amino acids by Z-score.

[0059] Step S35: inputting the first extracted feature, the second extracted feature and the third extracted feature into the prediction network to output the thermal stability prediction result.

[0060] The first extracted feature, the second extracted feature and the third extracted feature are spliced to form a to-be-predicted feature. A plurality of convolution network layers in the prediction network respectively process the to-be-predicted feature through matrix multiplication and aggregation operation, and then output. The features output by the plurality of convolution network layers are merged and then output to a pooling layer. The pooling layer performs a pooling operation on the merged features and then outputs to an output layer. The output layer performs thermal stability prediction on the features output by the pooling layer and outputs a thermal stability prediction result.

[0061] In some embodiments, step S33 specifically includes the following steps: Step S41: obtaining the distance between the target amino acid residue and each other amino acid residue by using the second feature extraction network.

[0062] The distance between the target amino acid residue and each other amino acid residue can be calculated according to the distance between the alpha carbon of the target amino acid residue and the alpha carbon of the other amino acid residue.

[0063] Step S42: taking the other amino acid with a distance less than a second preset distance threshold as a neighboring amino acid residue of the target amino acid residue.

[0064] wherein, when the distance between the target amino acid residue and another amino acid residue is less than the second preset distance threshold, the other amino acid residue is a neighboring amino acid residue of the target amino acid residue. Exemplarily, the second preset distance can be 10 Å or 6 Å.

[0065] Step S43: generating a second extraction feature according to the target amino acid residue, the corresponding neighboring amino acid residue thereof, and the distance between the target amino acid residue and the neighboring amino acid residue.

[0066] In some embodiments, the thermal stability prediction result includes the amount of change in folding free energy before and after mutation. Exemplarily, in step S35, the calculation function of the output layer performs DDG score calculation on the features output by the pooling layer, and outputs the DDG score as the thermal stability prediction result.

[0067] Correspondingly, in step S14, the target dehydrogenase mutant is obtained according to the thermal stability prediction result, including the following steps: obtaining a dehydrogenase mutant with an amount of change in folding free energy before and after mutation less than a change amount threshold as the target dehydrogenase mutant.

[0068] wherein, the smaller the amount of change in folding free energy before and after mutation, the higher the thermal stability of the dehydrogenase mutant relative to the dehydrogenase, and exemplarily, the change amount threshold can be -1.0.

[0069] As an embodiment, after step S12, the method further includes the following steps: Step S51: generating a dehydrogenase mutant three-dimensional structure using the PyMOL tool according to the dehydrogenase three-dimensional structure in the dehydrogenase mutant data and the mutation information of at least one target amino acid residue.

[0070] wherein, the mutagenesis tool of PyMOL can conveniently perform single-point or multi-site amino acid replacement on the dehydrogenase three-dimensional structure to obtain the dehydrogenase mutant three-dimensional structure.

[0071] Step S52: performing molecular dynamics simulation on the dehydrogenase mutant three-dimensional structure to obtain state change data of the dehydrogenase mutant three-dimensional structure during the molecular dynamics simulation.

[0072] wherein, the dehydrogenase mutant PDB file generated by PyMOL is used as input, and a molecular dynamics simulation software such as GROMACS is used for short-time dynamics simulation. Exemplarily, a topology file and a parameter file can be generated for each PDB file, the simulation parameters (such as 100 ps) are configured, and then the simulation is run and the trajectory is output.

[0073] Step S53: obtaining the state data of the dehydrogenase mutant three-dimensional structure and the comparison result of the state data of the dehydrogenase three-dimensional structure, and screening the dehydrogenase mutant data according to the comparison result.

[0074] For example, the comparison result can be a comparison graph of the RMSD change curve of the dehydrogenase and the RMSD change curve of the dehydrogenase mutant, and the change trend of the thermal stability of the mutant is determined according to the comparison of the RMSD curves of the proteins before and after mutation. As shown in FIG. 5, a comparison graph of the RMSD change curve of the dehydrogenase (red curve) and the RMSD change curve of the dehydrogenase mutant (black curve) is shown. Figure 3

[0075] As an implementation, after step S12, the following steps are further included: Step S61: using at least one protein mutation stability prediction tool to make a prediction according to the dehydrogenase mutant data, and screening the dehydrogenase mutant data according to the prediction result.

[0076] The protein mutation stability prediction tool can be FoldX. FoldX constructs an energy equation by linearly adding various interaction forces (such as hydrogen bonds, van der Waals forces, and electrostatic interactions) and entropy changes in the folding process. In the calculation of the folding free energy, FoldX first fixes the protein backbone and the surrounding environment, and only samples the mutation site to obtain the optimal conformation, and then uses the energy equation to calculate the influence of the mutation on the folding free energy, thereby obtaining the DDG value. The DDG value represents the change in the folding free energy of the protein complex before and after mutation, and a positive value indicates that the protein stability decreases after mutation, and a negative value indicates that the stability increases.

[0077] The protein mutation stability prediction tool can also be I Mutant2.0, ESM-1v, or ESM2.

[0078] In this embodiment, the multiple protein mutation stability prediction tools and the pre-trained protein thermal stability prediction model are used in combination to increase the accuracy of the thermal stability prediction of the mutant.

[0079] In some embodiments, as shown in FIG. 6, a comparison graph of the accuracy of thermal stability prediction using FoldX alone, I Mutant2.0 alone, ESM-1v alone, ESM2 alone, and the protein thermal stability prediction model alone is shown. Figure 4

[0080] As an implementation, in step S11, the dehydrogenase three-dimensional structure is obtained, and specifically includes the following steps: ​​Step S71: input the amino acid sequence of the dehydrogenase and the molecular structure of the substrate into the pre-trained protein structure prediction model, output the dehydrogenase-substrate complex three-dimensional structure, and obtain the dehydrogenase three-dimensional structure according to the dehydrogenase-substrate complex three-dimensional structure.

[0081] wherein the amino acid sequence of the enzyme can be provided in FASTA format; the molecular structure of the substrate can be provided in the form of a SMILES string or a molecular graph. The protein structure prediction model is used to predict and generate the complex three-dimensional structure of cellobiose epimerase-substrate, and the dehydrogenase-substrate complex three-dimensional structure with a plddt (Predicted Local Distance Difference Test) score greater than a score threshold or the dehydrogenase-substrate complex three-dimensional structure with the highest plddt score can be selected for output. Illustratively, the protein structure prediction model can employ an AlphaFold 3 model or a Protenix deep learning model.

[0082] Correspondingly, in step S11, the dehydrogenase-substrate complex three-dimensional structure is subjected to molecular dynamics simulation at different temperatures. The tool for molecular dynamics simulation is, for example, GROMACS (GROningen MAchine for Chemical Simulations, GROMACS).

[0083] Specifically, at the corresponding temperature, the step of molecular dynamics simulation is as follows: First, the three-dimensional structure of the enzyme-substrate complex is established and energy minimized to obtain the initial structure of the enzyme-substrate complex.

[0084] wherein the initial structure of the enzyme-substrate complex is obtained based on the catalytic disability of the enzyme and the binding site of the substrate; and the initial structure of the enzyme-substrate complex is energy minimized to eliminate unreasonable conformations in the initial structure.

[0085] Then, the initial structure of the enzyme-substrate complex is subjected to equilibrium simulation to obtain the enzyme-substrate complex in the equilibrium state.

[0086] wherein the molecular dynamics simulation parameters, including temperature, pressure, solvent environment, etc., are set to ensure that the experimental conditions match the physiological environment, the initial structure of the enzyme-substrate complex is subjected to equilibrium simulation, and each parameter is gradually adjusted to make the system reach a thermodynamic equilibrium state. After energy minimization and equilibrium simulation, the system reaches a stable state, reducing the energy deviation of the initial structure of the initial structure of the enzyme-substrate complex.

[0087] Finally, the enzyme-substrate complex in the equilibrium state is simulated in the production phase to analyze the structural stability and energy change, and the results of the structural stability and energy change are taken as state change data in the process of molecular dynamics simulation.

[0088] In the production simulation phase, the enzyme-substrate interaction trajectory and conformation change are recorded, the structural stability and energy change are analyzed according to the simulation trajectory and conformation change, and the state change data of the production simulation phase are obtained. The state change data can include multiple target indicators, which can include but are not limited to binding energy, number of hydrogen bonds, root mean square deviation, or root mean square fluctuation. Exemplarily, the time of the production simulation phase is less than or equal to 100 nanoseconds, for example, the production simulation process can last for 10 nanoseconds to 100 nanoseconds.

[0089] An embodiment of the present application provides a kind of dehydrogenase thermal stability optimization device, as shown in Figure 5 The dehydrogenase thermal stability optimization device 200 includes a structure data acquisition module 21, a mutation target determination module 22, a mutant data generation module 23, and a prediction module 24. The structure data acquisition module 21 is configured to obtain a three-dimensional structure of a dehydrogenase, perform molecular dynamics simulation on the three-dimensional structure of the dehydrogenase at different temperatures, and obtain state change data of the three-dimensional structure of the dehydrogenase during the molecular dynamics simulation at different temperatures. The mutation target determination module 22 is configured to determine target amino acid residues in the three-dimensional structure of the dehydrogenase that are sensitive to temperature based on the state data at different temperatures. The mutant data generation module 23 is configured to perform saturation mutation on the target amino acid residues and obtain a plurality of dehydrogenase mutant data. The prediction module 24 is configured to input the dehydrogenase mutant data into a pre-trained protein thermal stability prediction model, output a thermal stability prediction result of a corresponding dehydrogenase mutant, and obtain a target dehydrogenase mutant based on the thermal stability prediction result. The protein thermal stability prediction model is trained based on pre-set proteins and their thermal stability data.

[0090] As an implementation, the mutation target determination module 22 is further configured to: for each temperature, calculate the RMSF of each amino acid residue in the three-dimensional structure of the dehydrogenase based on the state data at the corresponding temperature; obtain candidate target amino acid residues from the amino acid residues based on the RMSF of each amino acid residue at different temperatures; and obtain the distance between the candidate target amino acid residues and the substrate. The candidate target amino acid residues with a distance greater than a first pre-set distance threshold and located in a loop region are taken as the target amino acid residues.

[0091] As an implementation form, the dehydrogenase mutant data comprises an amino acid sequence of the dehydrogenase, a three-dimensional structure of the dehydrogenase, and mutation information of at least one target amino acid residue; the prediction module 24 is further configured to: acquire an amino acid sequence of the dehydrogenase mutant according to the amino acid sequence of the dehydrogenase and the mutation information of the at least one target amino acid residue; extract a first sequence feature of the amino acid sequence of the dehydrogenase and a second sequence feature of the amino acid sequence of the dehydrogenase mutant respectively by using a first feature extraction network, and splice the first sequence feature and the second sequence feature to obtain a first extracted feature; acquire a second extracted feature according to the three-dimensional structure of the dehydrogenase and the mutation information of the at least one target amino acid residue by using a second feature extraction network; acquire a third extracted feature according to the mutation information of the at least one target amino acid residue by using a third extraction network; and input the first extracted feature, the second extracted feature, and the third extracted feature into a prediction network to output the thermal stability prediction result.

[0092] In some embodiments, the prediction module 24 is further configured to: acquire distances between the target amino acid residue and each other amino acid residue by using the second feature extraction network; take other amino acids with distances less than a second preset distance threshold as neighboring amino acid residues of the target amino acid residue; and generate a second extracted feature according to the target amino acid residue, the corresponding neighboring amino acid residues, and the distances between the target amino acid residue and the neighboring amino acid residues.

[0093] In some embodiments, the thermal stability prediction result comprises a change amount of folding free energy before and after mutation; and the prediction module 24 is further configured to: acquire a dehydrogenase mutant with a change amount of folding free energy before and after mutation less than a change amount threshold as the target dehydrogenase mutant.

[0094] As an implementation form, the prediction module 24 is further configured to: generate a three-dimensional structure of the dehydrogenase mutant according to the three-dimensional structure of the dehydrogenase and the mutation information of the at least one target amino acid residue in the dehydrogenase mutant data by using a PyMOL tool; perform molecular dynamics simulation on the three-dimensional structure of the dehydrogenase mutant to acquire state change data of the three-dimensional structure of the dehydrogenase mutant during the molecular dynamics simulation; acquire a comparison result of the state data of the three-dimensional structure of the dehydrogenase mutant and the state data of the three-dimensional structure of the dehydrogenase, and filter the dehydrogenase mutant data according to the comparison result.

[0095] As an implementation form, the prediction module 24 is further configured to: perform prediction according to the dehydrogenase mutant data by using at least one protein mutation stability prediction tool, and filter the dehydrogenase mutant data according to a prediction result.

[0096] Figure 6is a structural schematic diagram of an electronic device according to an embodiment of the present application. As shown in Figure 5 The electronic device 30 includes a processor 31 and a memory 32 coupled to the processor 31.

[0097] The memory 32 stores program instructions for implementing the method for optimizing the thermal stability of a dehydrogenase according to any of the above embodiments.

[0098] The processor 31 is configured to execute the program instructions stored in the memory 32 to optimize the thermal stability of a dehydrogenase.

[0099] The processor 31 can also be referred to as a CPU (Central Processing Unit). The processor 31 can be an integrated circuit chip with processing capability. The processor 31 can also be a general purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0100] Referring to Figure 7 Figure 7 is a structural schematic diagram of a computer readable storage medium according to an embodiment of the present application. The storage medium according to the embodiment of the present application stores program instructions 41 capable of implementing all the above methods. The storage medium can be non-volatile or volatile. The program instructions 41 can be stored in the above storage medium in the form of a software product, including a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of the present application. The above storage medium includes a U disk, a mobile hard disk, a ROM (Read-Only Memory), a RAM (Random Access Memory), a magnetic disk or an optical disk, etc. various media capable of storing program codes, or a computer, a server, a mobile phone, a tablet, etc. terminal device.

[0101] ​In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the division of the units is only a logical function division, and there can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between the units can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0102] In addition, each function unit in the various embodiments of the present application can be integrated in a processing unit, or each unit can exist alone physically, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware, or in the form of a software function unit. The above is only an embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation using the content of the present application specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

[0103] The above is only an embodiment of the present application, and it should be noted that for those skilled in the art, without departing from the creative concept of the present application, improvements can be made, but these are all within the protection scope of the present application.

Claims

1. A method for optimizing the thermostability of a dehydrogenase, characterized in that, The method comprises the following steps: acquiring a dehydrogenase three-dimensional structure, performing molecular dynamics simulation on the dehydrogenase three-dimensional structure at different temperatures, and acquiring state change data of the dehydrogenase three-dimensional structure during the molecular dynamics simulation at different temperatures; determining target amino acid residues sensitive to temperature in the dehydrogenase three-dimensional structure according to the state data at different temperatures; performing saturation mutation on the target amino acid residues to obtain dehydrogenase mutant data; inputting the dehydrogenase mutant data into a pre-trained protein thermal stability prediction model to output a thermal stability prediction result of the corresponding dehydrogenase mutant, and acquiring a target dehydrogenase mutant according to the thermal stability prediction result, wherein the protein thermal stability prediction model is obtained by training preset protein mutation data and thermal stability data.

2. The method of thermostability optimization of a dehydrogenase enzyme according to claim 1, wherein, The method comprises the following steps: for each temperature, calculating the RMSF of each amino acid residue in the dehydrogenase three-dimensional structure according to the state data at the corresponding temperature; acquiring candidate target amino acid residues from the amino acid residues according to the RMSF of each amino acid residue at different temperatures; acquiring the distance between the candidate target amino acid residues and the substrate, and taking the candidate target amino acid residues with a distance greater than a first preset distance threshold and located in a loop region as the target amino acid residues.

3. The method of dehydrogenase thermostability optimization of claim 1, wherein, The dehydrogenase mutant data comprises the amino acid sequence of the dehydrogenase, the three-dimensional structure of the dehydrogenase, and mutation information of at least one target amino acid residue. The method comprises the following steps: acquiring the amino acid sequence of the dehydrogenase mutant according to the amino acid sequence of the dehydrogenase and the mutation information of at least one target amino acid residue; extracting a first sequence feature of the amino acid sequence of the dehydrogenase and a second sequence feature of the amino acid sequence of the dehydrogenase mutant by using a first feature extraction network, and splicing the first sequence feature and the second sequence feature to obtain a first extraction feature; acquiring a second extraction feature according to the three-dimensional structure of the dehydrogenase and the mutation information of at least one target amino acid residue by using a second feature extraction network; acquiring a third extraction feature according to the mutation information of at least one target amino acid residue by using a third extraction network; inputting the first extraction feature, the second extraction feature, and the third extraction feature into a prediction network to output the thermal stability prediction result.

4. The method of dehydrogenase thermostability optimization of claim 3, wherein, The method comprises the following steps: acquiring the distance between the target amino acid residues and each other amino acid residue by using the second feature extraction network; taking other amino acids with a distance less than a second preset distance threshold as adjacent amino acid residues of the target amino acid residues. According to the target amino acid residue, its corresponding adjacent amino acid residue, and the distance between the target amino acid residue and the adjacent amino acid residue, a second extraction feature is generated.

5. The method of dehydrogenase thermostability optimization of claim 3, wherein, The thermal stability prediction result includes the change in folding free energy before and after mutation; Accordingly, the target dehydrogenase mutant obtained according to the thermal stability prediction result comprises: The dehydrogenase mutant with a change in folding free energy before and after mutation less than a change threshold is obtained as the target dehydrogenase mutant.

6. The method of dehydrogenase thermostability optimization of claim 1, wherein, After the saturation mutation of the target amino acid residue and the obtaining of the plurality of dehydrogenase mutant data, the method further comprises: A dehydrogenase mutant three-dimensional structure is generated by using a PyMOL tool according to the dehydrogenase three-dimensional structure in the dehydrogenase mutant data and the mutation information of at least one target amino acid residue; Molecular dynamics simulation is performed on the dehydrogenase mutant three-dimensional structure to obtain state change data of the dehydrogenase mutant three-dimensional structure during the molecular dynamics simulation; The state data of the dehydrogenase mutant three-dimensional structure and the state data of the dehydrogenase three-dimensional structure are compared, and the dehydrogenase mutant data is screened according to the comparison result.

7. The method of dehydrogenase thermostability optimization of claim 1, wherein, After the saturation mutation of the target amino acid residue and the obtaining of the plurality of dehydrogenase mutant data, the method further comprises: At least one protein mutation stability prediction tool is used to predict according to the dehydrogenase mutant data, and the dehydrogenase mutant data is screened according to the prediction result.

8. A device for optimizing the thermostability of a dehydrogenase, characterized in that, Comprise: The structure data acquisition module is used to acquire a dehydrogenase three-dimensional structure, perform molecular dynamics simulation on the dehydrogenase three-dimensional structure at different temperatures, and acquire state change data of the dehydrogenase three-dimensional structure during the molecular dynamics simulation at different temperatures; The mutation target determination module is used to determine a target amino acid residue sensitive to temperature in the dehydrogenase three-dimensional structure according to the state data at different temperatures; The mutant data generation module is used to perform saturation mutation on the target amino acid residue to obtain a plurality of dehydrogenase mutant data. The prediction module is used to input the dehydrogenase mutant data into a pre-trained protein thermal stability prediction model, output a thermal stability prediction result of the corresponding dehydrogenase mutant, and obtain a target dehydrogenase mutant according to the thermal stability prediction result. The protein thermal stability prediction model is obtained by training a preset protein and its thermal stability data.

9. An electronic device, comprising: The processor and a memory coupled to the processor are included, and the memory stores program instructions executable by the processor; the processor executes the program instructions stored in the memory to implement the dehydrogenase thermal stability optimization method in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program instructions, and the program instructions are executed by the processor to implement the dehydrogenase thermal stability optimization method in any one of claims 1-7.

Images