Deep learning algorithm for predicting three-dimensional structure of non-natural cyclic peptide monomer and compound thereof based on AlphaFold

Abstract

The invention discloses a deep learning algorithm for predicting a three-dimensional structure of a non-natural cyclic peptide monomer and a compound thereof based on AlphaFold. The deep learning algorithm specifically comprises the following steps: acquiring a non-natural linear peptide data set; dividing a training set, a verification set and a test set; the method comprises the following steps: adding a neural network used for capturing polypeptide atomic scale characterization in AlphaFold; pooling the obtained atomic scale characterization, and then combining the pooled atomic scale characterization with the original residue scale characterization of AlphaFold; the structure of the non-natural amino acid is predicted by defining a rigid group of the non-natural amino acid and initializing atomic coordinates of the non-natural amino acid; training the modified AlphaFold by using a non-natural line peptide training set and a verification set; testing on a non-natural line peptide test set by using the trained model; acquiring a non-natural cyclic peptide test data set; modifying a relative position coding matrix in the model; and testing on a non-natural cyclic peptide test set by using the modified model. Compared with a previous structure prediction model, the method provided by the invention can be used for accurately predicting the structures of the non-natural cyclic peptide monomer and the compound of the non-natural cyclic peptide monomer.

Description

Technical Field

[0001] The present invention relates to the technical field of cyclic peptide drug research and development, and specifically to a deep learning algorithm based on AlphaFold to predict the three-dimensional structure of non-natural cyclic peptide monomers and their complexes. Background Art

[0002] Peptides are compounds formed by amino acids connected by peptide bonds. Their molecular weight is between small molecules and large molecules, and they can selectively regulate a variety of protein-protein interactions. In drug development, peptide-based therapies have shown unique advantages: compared with small molecule drugs, they have higher specificity and lower toxicity; compared with protein drugs, they show lower immunogenicity and lower production costs, which make peptides candidate drugs for the treatment of various diseases. However, linear peptides are easily hydrolyzed by proteases. By cyclizing linear peptides and introducing non-natural amino acids, not only can their binding affinity and specificity be significantly improved, but also their stability and membrane permeability can be improved. Therefore, the research and development of cyclic peptides containing non-natural amino acids has become a hot spot in peptide drug research.

[0003] Predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes with protein receptors is crucial for structure-based drug design. As one of the core methods of computer-aided drug design, structure-based drug design has been widely used in the discovery and optimization of lead compounds, ADMET property prediction and drug repurposing, and has become an effective tool for developing new drug candidates. The rapid development of artificial intelligence algorithms has greatly improved structure prediction models. Tools such as AlphaFold2, RoseTTAFold and ESMFold have achieved accurate predictions of protein monomers and protein-protein complexes. Based on these models, subsequent studies have extended structure prediction to protein-nucleic acid, protein-line peptide, protein-small molecule and protein-cyclic peptide complexes. Recently, models such as RoseTTAFold All-Atom and AlphaFold3 have further expanded the scope of spatial structure prediction based on deep learning to complex complex structures involving proteins, nucleic acids, small molecules, ions and modified residues. However, none of these methods have solved the problem of structure prediction of non-natural cyclic peptide monomers and their complexes. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention successfully realizes the accurate spatial structure prediction of non-natural cyclic peptide monomers and their complexes based on the AlphaFold model, atomic feature extraction and non-natural amino acid coding information expansion. Considering that polypeptides have certain small molecule characteristics, the present invention integrates the neural network module that characterizes the atomic features of polypeptides into the AlphaFold model to realize multi-scale modeling of polypeptide molecules. In addition, the present invention has also successfully expanded the coding information of non-natural amino acids, so that the model has the ability to predict non-natural amino acids. Although the model can be trained directly on cyclic peptide structures with non-natural amino acids, such structures are relatively scarce in the RCSB ProteinDataBank database. Therefore, the present invention first uses the linear peptide data containing non-natural amino acids to fine-tune the model, and then modifies the relative position coding matrix in the model so that it can accurately predict the cyclic peptide structure. Experimental results show that the model can accurately predict the three-dimensional structure of non-natural cyclic peptide monomers and their complexes.

[0005] The specific technical solutions are as follows:

[0006] A deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes based on AlphaFold, comprising the following steps:

[0007] Step 1: Obtain the spatial structure file containing non-natural linear peptides and perform preprocessing. The linear peptide data containing non-natural amino acids comes from the ModPep dataset, which contains a total of 501 linear peptide samples containing non-natural amino acid residues, most of which are complex structures formed by non-natural linear peptides and proteins. Using the PDB ID provided by it, the corresponding crystal structure was downloaded from the RCSBProteinDataBank database, and duplicates were deleted, and finally 419 unique three-dimensional structures were obtained. The solvent, hydrogen atoms, and heteroatoms in the crystal structure were then deleted, and only the first state of each structure was retained. In order to reduce the redundancy of the crystal structure and reduce the computational requirements during the training process, some chains in the spatial structure were deleted according to the following rules: chains with less than 50 amino acid residues in the three-dimensional structure were classified as polypeptide chains, while other chains were classified as protein chains. If a structure contains at least one protein chain, the sequences of all polypeptide chains are aligned, and for those with the same or partial overlap and a distance greater than The peptide sequence of the crystal structure is analyzed, and only the longest polypeptide chain is retained, and the remaining polypeptide chains are deleted. In addition, if there are at least two protein chains in the structure at this time, only the protein chain with the most contact with the polypeptide chain is retained. Then, the types of non-natural amino acids in the crystal structure are analyzed, and any structure containing non-natural amino acids that are only found in this crystal structure will also be deleted. After this pre-processing process, a total of 382 non-natural linear peptide samples were obtained, which contained 23 different non-natural amino acid residues;

[0008] Step 2: Randomly divide the dataset obtained in step 1 into training set, validation set, and test set in a ratio of 7.5:1.5:1. Make sure that each unnatural amino acid appears at least once in the validation set and the test set. Use ColabFold to generate various features required for training and validation;

[0009] Step three: In the original AlphaFold model, an additional neural network module is added to the polypeptide molecule to capture its atomic-scale characterization. The atomic-scale features come from the atomic elements and the bond information between them, which is extracted using RDKit after converting the FASTA of the polypeptide molecule into SMILES. The atomic element features are encoded using one-hot encoding. In order to reduce the sparsity of the encoding matrix, the present invention only encodes common atoms (B, C, F, I, N, O, P, S, Br, Cl), and rare atoms share the encoding. The generated one-hot matrix is ​​input into the multi-head attention and multi-layer perceptron. The bond information between atoms is also encoded using the one-hot method, which includes a total of five bond types (single bond, double bond, triple bond, aromatic bond and ionic bond). The obtained encoding matrix is ​​transformed through a linear layer;

[0010] Step 4: Pool the atomic-scale representation obtained in step 3 to facilitate combination with the original residue-scale representation of the AlphaFold model to achieve multi-scale modeling of polypeptide molecules. Both the atomic element representation and the representation of bond information between atoms use an attention-based pooling mechanism to map them to amino acid residue-scale embeddings. The pooled atomic element embeddings will be added to the unique-hot matrix encoding the polypeptide type in the AlphaFold model, and then input into the Evoformer module to generate a single representation. The pooled bond information embeddings will be added to the paired activation matrix in the AlphaFold model to generate paired representations in the Evoformer module. The following is the calculation formula for the attention-based pooling mechanism:

[0011]

[0012] A score =Softmax(x atom +F),

[0013]

[0014] Where M is a mask matrix used to indicate which amino acid residue the atom belongs to, ∈ is a very small positive value to prevent division by zero errors, and x atom represents the atomic scale representation, x residueIt is the scale representation of the amino acid residues after pooling; Step 5: Predict the spatial structure of non-natural amino acids by defining rigid groups and initializing the atomic coordinates of non-natural amino acids. The Structure module in the AlphaFold model uses the single representation and paired representation obtained from the Evoformer module to predict the torsion angle of each amino acid residue in the protein. These predicted torsion angles, together with the predefined rigid groups and initialized atomic coordinates, construct the coordinates of all atoms in the spatial structure of the protein. According to the dependence of each atom in the non-natural amino acid residue on a specific torsion angle, they are divided into five rigid groups, namely the backbone rigid group, ω rigid group, Φ rigid group, ψ rigid group and χ rigid group. The four atoms C, Cα, Cβ, and N on the main chain are the backbone rigid group, the hydrogen atoms connected to Cα are classified as the ω rigid group, the Φ rigid group only contains the hydrogen atoms on the amino group, the ψ rigid group only contains the oxygen atoms in the carboxyl group, and the χ rigid group contains all the side chain atoms. Since the present invention does not predict hydrogen atoms, the ω rigid group and the Φ rigid group can be ignored. After determining the rigid group, the initialization coordinates of each atom of the non-natural amino acid are determined according to the crystal structure of the non-natural amino acid. For the backbone rigidity group, Cα is taken as the origin, C is placed on the positive x-axis, and N is placed on the xy plane to establish a coordinate system, and the initial coordinates of Cα, Cβ, C, and N atoms are calculated respectively. In the ψ rigidity group, C is taken as the origin, Cα is on the negative x-axis, and N in the next amino acid residue is on the xy plane to establish a coordinate system to determine the initial coordinates of O. The χ rigidity group contains all atoms that depend on the torsion angle of the side chain. There are four groups in total. According to the true coordinates of the four atoms in each group, the third atom is set as the origin, the second atom is placed on the negative x-axis, and the first atom is placed on the xy plane to establish a coordinate system, thereby determining the relative coordinates of the fourth atom in the coordinate system, and then the fourth atom is rotated around the x-axis to the xy plane to determine the initial coordinates of the fourth atom. The calculation formula is as follows:

[0015]

[0016] Coordinate init =R x ·Coordinate relative ,

[0017] Among them, Coordinate relative is the relative coordinate of the fourth atom, and θ is the angle between it and the xy plane, R x is the rotation matrix, Coordinate init represents the initialization coordinates of the fourth atom;

[0018] Step 6: Use the training set and validation set obtained in step 2 to train the above-mentioned non-natural line peptide monomer and its complex structure prediction model. During training, the amino acid residues in each training sample are trimmed to 220, while the validation set remains uncropped. For samples with breakpoints in the training set, the loss value is halved. In order to prevent overfitting, an early stopping strategy is implemented based on the loss of the validation set. The five sets of parameters of the AlphaFold model are fine-tuned using non-natural line peptide structures;

[0019] Step 7: Use the five trained models to test on the test set obtained in step 2;

[0020] Step 8: Obtain a test data set of non-natural cyclic peptide monomers and their complexes. Part of the cyclic peptide data containing non-natural amino acids comes from cPEPmatch. In order to effectively evaluate the accuracy of the model on the cyclic peptide data, we filtered the data set, excluded samples containing unencoded non-natural amino acids, and limited the cyclization forms to head-to-tail cyclization and disulfide bond cyclization. In addition, a certain search was conducted on the RCSB ProteinDataBank database to determine the cyclic peptide structures that meet the standards. After downloading the corresponding non-natural cyclic peptide crystal structures from the RCSB ProteinDataBank database, the solvent molecules, hydrogen atoms, and heteroatoms were removed, and only the first state of each structure was retained, and samples containing nucleic acids and repeats were deleted. Finally, a total of 34 cyclic peptide test samples containing non-natural amino acids were obtained;

[0021] Step 9: Modify the relative position encoding matrix in the structure prediction model of the non-natural linear peptide monomer and its complex, so that the relative positions of the two loop residues in the model can be accurately characterized and adjacent to each other;

[0022] Step 10: Use the model obtained in step 9 to test on the non-natural cyclic peptide test set obtained in step 8. As a technical solution of the present invention, in the process of constructing multiple sequence alignment features, all non-natural amino acids are regarded as unknown amino acids. By integrating the atomic scale embedding information of the polypeptide into the AlphaFold model, the model can distinguish various non-natural amino acid residues and realize multi-scale modeling of polypeptide molecules. As a technical solution of the present invention, after defining the corresponding rigid groups and initialized atomic coordinates for non-natural amino acids, the modeling of each protein atom can be completed by predicting the seven torsion angles of each amino acid, which greatly reduces the complexity of the prediction and enables the model to predict their atomic coordinates in an end-to-end manner. As a technical solution of the present invention, five sets of AlphaFold model parameter sets were fine-tuned using non-natural linear peptide data, and the relative position encoding matrix in the model was modified to enable it to have the ability to predict the structure of non-natural cyclic peptides.

[0023] The present invention provides a deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes based on the AlphaFold model, atomic feature extraction and non-natural amino acid encoding information expansion, which has the following beneficial effects:

[0024] (1) The present invention proposes a multi-scale cyclic peptide molecule characterization method, which has better feature extraction capabilities. Compared with previous methods, the present invention can better characterize cyclic peptide molecules, and this method is also applicable to the atomic characterization of linear peptide molecules.

[0025] (2) The present invention proposes a method for predicting the structure of non-natural amino acids. By defining the corresponding rigid groups and initializing the atomic coordinates of non-natural amino acids, the prediction range of the structure prediction model is successfully expanded to non-natural linear peptides and non-natural cyclic peptide structures.

[0026] (3) The present invention designs a zero-shot learning method for predicting the structure of non-natural cyclic peptides. The model is first trained on non-natural linear peptide data, and then the relative position encoding matrix is ​​modified to make it suitable for predicting the structure of non-natural cyclic peptides. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a flowchart of the non-natural peptide three-dimensional structure prediction algorithm based on AlphaFold;

[0028] Figure 2 Diagram of the training and inference framework of the non-natural cyclic peptide three-dimensional structure prediction model. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present invention are described clearly and completely below, and the described embodiments are only a part of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present invention.

[0030] like Figure 1 and Figure 2 As shown, this embodiment provides a deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes. The method comprises the following steps:

[0031] Step 1: Obtain the spatial structure files containing non-natural linear peptides from the RCSB ProteinDataBank database and perform preprocessing. The linear peptide data containing non-natural amino acids comes from the ModPep data set, which contains a total of 501 linear peptide samples containing non-natural amino acid residues, most of which are complex structures formed by non-natural linear peptides and proteins. Using the PDB ID provided by it, download the corresponding crystal structure from the RCSB ProteinDataBank database, delete duplicates, and finally obtain 419 unique three-dimensional structures. Then delete the solvent, hydrogen atoms and heteroatoms in the crystal structure, and only retain the first state of each structure. In order to reduce the redundancy of the crystal structure and reduce the computational requirements during the training process, some chains in the spatial structure are deleted according to the following rules: chains with less than 50 amino acid residues in the three-dimensional structure are classified as polypeptide chains, while other chains are classified as protein chains. If a structure contains at least one protein chain, the sequences of all polypeptide chains are aligned, and for those with the same or partial overlap and a distance greater than The peptide sequence is then sorted, and only the longest polypeptide chain is retained, and the remaining polypeptide chains are deleted. In addition, if there are at least two protein chains in the structure at this time, only the protein chain with the most contact with the polypeptide chain is retained. Then, the types of unnatural amino acids in the crystal structure are analyzed, and any structure containing unnatural amino acids that are only found in this crystal structure will also be deleted. After this preprocessing process, a total of 382 unnatural linear peptide samples were obtained, containing 23 different unnatural amino acid residues;

[0032] Step 2: Randomly divide the data set obtained in step 1 into training set, validation set, and test set in a ratio of 7.5:1.5:1, ensuring that each unnatural amino acid appears at least once in the validation set and the test set, and use ColabFold to generate various features required for training and validation;

[0033] Step three: In the original AlphaFold model, an additional neural network is added to the polypeptide molecule to capture its atomic-scale characterization. The atomic-scale features come from the atomic elements and the bond information between them. This information is extracted using RDKit after converting the FASTA of the polypeptide molecule into SMILES. The atomic element features are encoded using one-hot encoding. In order to reduce the sparsity of the encoding matrix, the present invention only encodes common atoms (B, C, F, I, N, O, P, S, Br, Cl), and rare atoms share the encoding. The generated one-hot matrix is ​​input into the multi-head attention and multi-layer perceptron. The bond information between atoms is also encoded using the one-hot method, which includes a total of five bond types (single bond, double bond, triple bond, aromatic bond and ionic bond). The obtained encoding matrix is ​​transformed through a linear layer;

[0034] Step 4: Pool the atomic-scale representation obtained in step 3 and combine it with the original residue-scale representation of the AlphaFold model to achieve multi-scale modeling of polypeptide molecules. The atomic element representation and the representation of bond information between atoms both use an attention-based pooling mechanism to map them to amino acid residue-scale embeddings. The pooled atomic element embeddings will be added to the unique-hot matrix encoding the polypeptide type in the AlphaFold model, and then input into the Evoformer module to generate a single representation. The pooled bond information embeddings will be added to the paired activation matrix in the AlphaFold model to generate paired representations in the Evoformer module. The following is the calculation formula for the attention-based pooling mechanism:

[0035]

[0036] A score =Softmax(x atom +F),

[0037]

[0038] Where M is a mask matrix used to indicate which amino acid residue the atom belongs to, ∈ is a very small positive value to prevent division by zero errors, and x atom represents the atomic scale representation, x residueIt is the scale characterization of amino acid residues after pooling; Step 5: The prediction of non-natural polypeptide structure is realized by defining the rigid group of non-natural amino acids and initializing their atomic coordinates. The Structure module in the AlphaFold model uses the single characterization and paired characterization obtained from the Evoformer module to predict the torsion angle of each amino acid residue in the protein. These predicted torsion angles, together with the predefined rigid group and the initialized atomic coordinates, construct the coordinates of all atoms in the spatial structure of the protein. According to the dependence of each atom in the non-natural amino acid residue on a specific torsion angle, they are divided into five rigid groups, namely the backbone rigid group, the ω rigid group, the Φ rigid group, the ψ rigid group and the χ rigid group. The four atoms C, Cα, Cβ and N on the main chain are the backbone rigid group, the hydrogen atoms connected to Cα are classified as the ω rigid group, the Φ rigid group only contains the hydrogen atoms on the amino group, the ψ rigid group only contains the oxygen atoms in the carboxyl group, and the χ rigid group contains all the side chain atoms. Since the present invention does not predict hydrogen atoms, the ω rigid group and the Φ rigid group can be ignored. After determining the rigid group, the initialization coordinates of each atom of the non-natural amino acid are determined according to the crystal structure of the non-natural amino acid. For the backbone rigidity group, Cα is taken as the origin, C is placed on the positive x-axis, and N is placed on the xy plane to establish a coordinate system, and the initial coordinates of Cα, Cβ, C, and N atoms are calculated respectively. In the ψ rigidity group, C is taken as the origin, Cα is on the negative x-axis, and N in the next amino acid residue is on the xy plane to establish a coordinate system to determine the initial coordinates of O. The χ rigidity group contains all atoms that depend on the torsion angle of the side chain. There are four groups in total. According to the true coordinates of the four atoms in each group, the third atom is set as the origin, the second atom is placed on the negative x-axis, and the first atom is placed on the xy plane to establish a coordinate system, thereby determining the relative coordinates of the fourth atom in the coordinate system, and then the fourth atom is rotated around the x-axis to the xy plane to determine the initial coordinates of the fourth atom. The calculation formula is as follows:

[0039]

[0040] Coordinate init =R x ·Coordinate relative ,

[0041] Among them, Coordinate relative is the relative coordinate of the fourth atom, and θ is the angle between it and the xy plane, R x is the rotation matrix, Coordinate init represents the initialization coordinates of the fourth atom;

[0042] Step 6: Use the obtained non-natural peptide training set and validation set to train the non-natural peptide monomer and complex structure prediction model. During training, the amino acid residues in each training sample are trimmed to 220, while the validation set remains uncropped. For samples with breakpoints in the training set, the loss value is halved. In order to prevent overfitting, an early stopping strategy is implemented based on the loss of the validation set. The five sets of parameters of the AlphaFold model are fine-tuned using the non-natural peptide structure;

[0043] Step 7: Use the trained model to test on the non-natural peptide test set;

[0044] Step 8: Obtain the test data set of non-natural cyclic peptide monomers and their complexes from the RCSB ProteinDataBank database, and obtain a total of 34 cyclic peptide test samples containing non-natural amino acids;

[0045] Step 9: Modify the relative position encoding matrix algorithm in the structure prediction model of the non-natural linear peptide monomer and its complex, so that the relative positions of the two loop residues in the model can be accurately characterized and adjacent to each other;

[0046] Step 10: Use the modified structure prediction model to test on the non-natural cyclic peptide test set.

[0047] The present invention evaluates the predictive performance of the model by calculating the root mean square deviation (RMSD) between the predicted structure and the true structure. Specifically, the RMSD of the pocket-aligned peptide all atom , pocket-aligned peptide RMSD Cα RMSD of peptide aligned with pocket unaa Used to evaluate the structural accuracy of non-native peptide complexes. RMSD all atom , RMSD Cα and RMSD unaa It is used to evaluate the structural accuracy of non-natural peptide monomers and is specifically defined as follows:

[0048] For each crystal structure, the chains with less than 25 residues were defined as polypeptide chains, and the remaining chains were defined as protein chains. The protein chain backbone atoms (N, C, Cα) within the pocket are then aligned using the Kabsch algorithm to align the backbone atoms in the predicted structure and the crystal structure. The same rotation and translation matrix is ​​then used to align the polypeptide chains. The RMSD of the pocket-aligned polypeptide all atom is the RMSD between all heavy atoms in the peptide molecule, while the RMSD of the pocket-aligned peptide unaaOnly the RMSD between heavy atoms of unnatural amino acid residues is calculated. Pocket-aligned peptide RMSD Cα A similar procedure is followed, with the main difference being that only the distance to the Cα atom of the polypeptide chain is selected. The protein chain backbone atoms within are aligned, and the RMSD is calculated using only the Cα atoms in the polypeptide chain. all atom It is determined by aligning all heavy atoms in the predicted structure and the crystal structure using the Kabsch algorithm and then calculating the RMSD between all heavy atoms. When the calculation is limited to unnatural amino acid residues, the RMSD is unaa RMSD Cα It is obtained by aligning the Cα atoms in the predicted structure and the crystal structure and calculating the RMSD between these atoms.

[0049] In one embodiment of the present invention, the structure prediction model of the non-natural line peptide monomer and its complex obtained by training in the present invention was tested on the non-natural line peptide test set. The results are shown in Table 1. RMSD Cα The median is 0.994 (for the complex, RMSD Cα Refers to the RMSD of the peptides aligned with the pocket Cα , the same below), RMSD all atom The median is 1.906 (RMSD for the complex all atom Refers to the RMSD of the peptides aligned with the pocket all atom , the same below), RMSD unaa The median is 1.971 (for the complex, RMSD unaa Refers to the RMSD of the peptides aligned with the pocket unaa , the same below), confirming the powerful ability of the model in predicting the three-dimensional structures of non-natural linear peptide monomers and their complexes.

[0050] Table 1 Performance of the structure prediction model of non-natural peptide monomers and their complexes on the non-natural peptide test set

[0051]

[0052]

[0053] In one embodiment of the present invention, the trained non-natural linear peptide monomer and its complex structure prediction model is applied to the prediction of non-natural cyclic peptide structure by modifying the relative position encoding matrix. The results are shown in Table 2. RMSD Cα The median is 1.891, and the RMSD of most test samples is Cα Less than This indicates that the model is highly accurate in predicting the backbone of non-natural cyclic peptides. RMSD all atom The median is 2.872, and there are 6 samples with a RMSD of all atom Less than This indicates that the model also has excellent side chain structure prediction capabilities. RMSD unaa The median score was 2.579, which indicates that the model can also accurately predict the three-dimensional structures of most unnatural amino acid residues.

[0054] Table 2 Performance of the non-natural linear peptide monomer and its complex structure prediction model with modified relative position encoding matrix on the non-natural cyclic peptide test set

[0055]

[0056]

[0057] The above description is only a preferred specific embodiment of the present invention and does not limit the present invention in any way. Any technician in the relevant technical field, without departing from the scope of the technical solution of the present invention, makes any form of equivalent replacement or modification to the technical solution and technical content disclosed in the present invention, or directly or indirectly applies it to other related technical fields, which belongs to the content that does not depart from the technical solution of the present invention and is still included in the protection scope of the present invention.

Claims

1. A deep learning algorithm based on AlphaFold to predict the three-dimensional structure of non-natural cyclic peptide monomers and their complexes, characterized in that: The following steps are involved: Step 1: Obtain the linear peptide structures containing unnatural amino acids from the RCSB ProteinDataBank database, remove the solvent, hydrogen atoms, and heteroatoms from the crystal structure, and only retain the first state of each structure; Step 2: Divide the non-natural peptide dataset obtained in step 1 into training set, validation set and test set and construct features using ColabFold; Step 3: Add a neural network module to the original AlphaFold model to capture the atomic-scale representation of peptide molecules. The atomic-scale features come from the atomic elements and the bond information between them. Step 4: Pool the atomic-scale representation obtained in step 3 to facilitate its combination with the original residue-scale representation of the AlphaFold model to achieve multi-scale modeling of peptide molecules. The atomic element representation and the representation of the bond information between atoms are mapped to amino acid residue-scale embedding using an attention-based pooling mechanism. Step 5: Predict the spatial structure of non-natural amino acids by defining rigid groups and initializing the atomic coordinates of non-natural amino acids. The Structure module in the AlphaFold model uses the single representation and paired representation obtained from the Evoformer module to predict the torsion angle of each amino acid residue in the protein. These predicted torsion angles, together with the predefined rigid groups and initialized atomic coordinates, construct the coordinates of all atoms in the spatial structure of the protein. Step 6: Use the training set and validation set obtained in step 2 to train the above-mentioned non-natural linear peptide monomer and its complex structure prediction model. During training, the amino acid residues in each training sample are trimmed to 220, while the validation set remains untrimmed. For samples with breakpoints in the training set, the loss value is halved. In order to prevent overfitting, an early stopping strategy is implemented based on the loss of the validation set; Step 7: Use the trained model to test on the test set obtained in step 2; Step 8: Obtain a test data set of non-natural cyclic peptide monomers and their complexes, remove solvent molecules, hydrogen atoms, and heteroatoms, retain only the first state of each structure, and delete samples containing nucleic acids and repeats; Step 9: Modify the relative position encoding matrix in the structure prediction model of the non-natural linear peptide monomer and its complex, so that the relative positions of the two loop residues in the model can be accurately characterized and adjacent to each other; Step 10: Use the model obtained in step 9 to test on the non-natural cyclic peptide test set obtained in step 8.

2. A deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes based on AlphaFold as claimed in claim 1, characterized in that: The specific process of obtaining the linear peptide structure containing unnatural amino acids from the RCSB ProteinDataBank database in step 1 is as follows: using the PDB ID provided by the ModPep dataset, download the corresponding crystal structure from the RCSBProteinDataBank database and delete duplicates.

3. A deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes based on AlphaFold as claimed in claim 2, characterized in that: After retaining the first state of each structure in step 1, in order to reduce the redundancy of crystal structures and reduce the computational requirements during training, some chains in the spatial structure were deleted according to the following rules: chains with less than 50 amino acid residues in the three-dimensional structure were classified as polypeptide chains, while other chains were classified as protein chains. If a structure contained at least one protein chain, the sequences of all polypeptide chains were aligned, and sequences with the same or partial overlap and a distance greater than The polypeptide sequence of the structure is analyzed, and only the longest polypeptide chain is retained, and the remaining polypeptide chains are deleted. If there are at least two protein chains in the structure at this time, only the protein chain with the most contact with the polypeptide chain is retained. The types of unnatural amino acids in the crystal structure are analyzed, and any structure containing unnatural amino acids that are only found in this crystal structure will also be deleted.

4. A deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes based on AlphaFold as claimed in claim 3, characterized in that: The atomic element information and bond information in step 3 were extracted using RDKit after converting the FASTA of the peptide molecule into SMILES.

5. A deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes based on AlphaFold as claimed in claim 4, characterized in that: The specific operation process of step 4 is that the pooled atomic element embedding will be added to the one-hot matrix encoding the peptide type in the AlphaFold model, and then input into the Evoformer module to generate a single representation. The pooled key information embedding will be added to the paired activation matrix in the AlphaFold model to generate the paired representation in the Evoformer module. The following is the calculation formula of the attention-based pooling mechanism: A score =Softmax(x atom +F), Where M is a mask matrix used to indicate which amino acid residue the atom belongs to, ∈ is a very small positive value to prevent division by zero errors, and x atom represents the atomic scale representation, x residue It is the scale representation of the amino acid residues after pooling.

6. A deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes based on AlphaFold as claimed in claim 5, characterized in that: The specific process of defining the rigidity group and initializing the atomic coordinates of the non-natural amino acid in step 5 is to divide them into five rigidity groups according to the dependence of each atom in the non-natural amino acid residue on a specific torsion angle, namely, the backbone rigidity group, the ω rigidity group, the Φ rigidity group, the ψ rigidity group and the χ rigidity group. The four atoms C, Cα, Cβ and N on the main chain are the backbone rigidity group, the hydrogen atoms connected to Cα are classified as the ω rigidity group, the Φ rigidity group only includes the hydrogen atoms on the amino group, the ψ rigidity group only includes the oxygen atoms in the carboxyl group, and the χ rigidity group includes all the side chain atoms. After determining the rigidity group, the initialization coordinates of each atom of the non-natural amino acid are determined according to the crystal structure of the non-natural amino acid. For the backbone rigidity group, Cα is taken as the origin, C is placed on the positive x-axis, and N is placed In the xy plane, a coordinate system is established to calculate the initial coordinates of Cα, Cβ, C and N atoms respectively. In the ψ rigid group, C is used as the coordinate origin, Cα is on the negative x axis, and N in the next amino acid residue is on the xy plane. In this way, a coordinate system is established to determine the initial coordinates of O. The χ rigid group contains all atoms that depend on the torsion angle of the side chain. There are four groups in total. According to the true coordinates of the four atoms in each group, the third atom is set as the origin, the second atom is placed on the negative x axis, and the first atom is placed on the xy plane. In this way, a coordinate system is established to determine the relative coordinates of the fourth atom in the coordinate system. Then the fourth atom is rotated around the x axis to the xy plane to determine the initial coordinates of the fourth atom. The calculation formula is as follows: Coordinate init =R x ·Coordinate relative , Among them, Coordinate relative is the relative coordinate of the fourth atom, and θ is the angle between it and the xy plane, R x is the rotation matrix, Coordinate init Represents the initial coordinates of the fourth atom.

7. A deep learning algorithm for predicting the three-dimensional structure of non-natural cyclic peptide monomers and their complexes based on AlphaFold as claimed in claim 6, characterized in that: The specific operation process of step eight is to filter the non-natural cyclic peptide data from cPEPmatch, exclude samples containing uncoded non-natural amino acids, and limit the cyclization form to head-to-tail cyclization and disulfide bond cyclization. At the same time, the RCSB ProteinDataBank database is searched to determine the cyclic peptide structure that meets the standards. After downloading the corresponding non-natural cyclic peptide crystal structure from the RCSB ProteinDataBank database, the solvent molecules, hydrogen atoms and heteroatoms are removed, and only the first state of each structure is retained, and samples containing nucleic acids and repeats are deleted.

Images