Structural analysis system and modeling method for protein degradation targeted chimera complexes

Abstract

The invention discloses a structural analysis system and a modeling method for a protein degradation targeting chimera compound, and belongs to the field of computer-aided drug molecule design. The structural analysis system comprises a chimera component analysis module, a chimera induced complex component analysis module and a chimera linker length distribution analysis module, so that automatic structural splitting and analysis of the protein degradation targeted chimera drug are realized; through protein-protein docking, conformation search, conformation fusion and optimization, conformation clustering and conformation scoring and sorting, efficient and accurate structure modeling is realized. According to the method, automatic component splitting and structure analysis can be carried out, so that a potential drug molecule binding mode is comprehensively explored, the trial and error cost and time of traditional drug structure analysis are remarkably reduced, the design efficiency and success rate of a protein degradation targeting chimera are remarkably improved, and through automatic parameter optimization, the design efficiency of the protein degradation targeting chimera is greatly improved. And the high efficiency and reliability of the structure modeling process are ensured.

Description

Technical Field

[0001] The present invention relates to the field of computer-aided drug molecule design, and in particular to a structural analysis system and a modeling method for a protein degradation targeted chimera complex. Background Art

[0002] Proteolysis-targeting chimeras (PROTACs) are a new type of drug (hereinafter referred to as chimeras) that can achieve target protein degradation, greatly expanding the range of drug-regulated targets. This type of drug can bind to ubiquitin ligases and target proteins at the same time, thereby inducing the target protein to be ubiquitin-tagged, and ultimately recognized and degraded by the proteasome.

[0003] Due to the unique mechanism of this type of drug, it has several advantages over traditional small molecule inhibitors, including: complete inhibition of function, not limited to the function of the enzyme or a specific location on the target; long-lasting action and enhanced selectivity that are proportional to the protein turnover rate; and because a single protein degradation targeting chimeric molecule can degrade multiple copies of the target, it greatly expands the range of drug targets and has therefore become a hot spot for new drug research and development.

[0004] However, the research and development of such drugs is difficult. Their complex mechanism of action makes it difficult to design molecules based on experience or a single protein structure. Therefore, the research and development process of such drugs usually involves large-scale random chemical structure modification, synthesis and biological activity testing of the drugs. Therefore, a rational design method suitable for such drugs is urgently needed.

[0005] For example, the invention application with publication number CN118711703A discloses a PROTAC molecular drug design method based on the DEL platform. Based on the DNA Encoded Library (DEL), bifunctional molecules are quickly generated and evaluated through molecular generation models and computer-aided drug design, which accelerates the construction of the PROTAC-like DEL bifunctional molecule library. Then, by modeling and simulating the protein-bifunctional molecule ternary complex system mediated by bifunctional molecules, the binding mode of the bifunctional molecules in the PROTAC-like DEL bifunctional molecule library is predicted, and the bifunctional molecules are further optimized, making the designed bifunctional molecules more reasonable and reducing the difficulty of molecule synthesis.

[0006] For example, reference 2 (Comprehensive modeling of PROTAC-mediated ternary complexes) discloses a comprehensive modeling method based on FRODOCK and RosettaDock. The structure is initially docked through an effective sampling algorithm of coordinate constraints provided by FRODOCK, and each docked structure is optimized in combination with RosettaDock for predicting the structure of PROTAC-mediated ternary complexes. This comprehensive modeling method models the ternary complex starting from the non-binding structure, making the ternary complex close to the natural structure, which is helpful for the design and optimization of new PROTACs.

[0007] However, in existing work, the structural modeling methods of chimera-induced ternary complexes have some limitations. These structural modeling methods do not provide reliable analysis of chimeras and chimera-induced ternary complex components, making the results of structural modeling unable to be well close to the structure of the native ternary complex, and the structural modeling prediction lacks a reasonable scoring and ranking standard, resulting in insufficient accuracy and low usability of the final modeling results. Summary of the invention

[0008] The purpose of the present invention is to provide a structural analysis system and modeling method for a protein degradation targeted chimera complex, which are used to solve the problems of insufficient accuracy of existing modeling scoring performance and structural modeling results, and low usability.

[0009] To achieve the above-mentioned purpose of the invention, the embodiment provides a structural analysis system for a protein degradation targeted chimera complex, comprising: a chimera component analysis module, a chimera induced complex component analysis module and a chimera linker length distribution analysis module; The chimera component analysis module is used to identify a series of structural fragments in the chimera molecule, and fuse adjacent structural fragments using an iterative method; sort each fused fragment, test the rationality of the connection between the target protein ligand and the ubiquitin ligase ligand and the sorted fused fragments, and regard the rationally sorted fused fragments as the connector part to achieve the splitting and component analysis of the chimera; The chimera-induced complex component analysis module is used to split the chimera-induced ternary complex, and analyze, identify and screen the split target protein, chimera and ubiquitin ligase to achieve component analysis of the chimera-induced complex; The connector length distribution analysis module of the chimera is used to perform molecular conformation search on the connector part extracted from the chimera in the complex, analyze the connector endpoint distance distribution when the chimera is in different three-dimensional conformations, and use it as an indicator of the connector length range for structural analysis of protein degradation targeted chimera complexes.

[0010] In one embodiment, the identifying a series of structural fragments in the chimeric molecule comprises: identifying structural fragments based on a molecular structure pattern or based on a loop matching pattern; The structural fragments matched by the molecular structure pattern include: primary amide, primary thioamide, a carbon atom connected to two hydrogen atoms, a hydroxymethyl group, an alkynyl group, a substructure formed by a ring-containing nitrogen atom connected to a non-cyclic carbonyl group, or an azo group formed by two nitrogen atoms connected by a double bond; The structural fragments of the ring matching pattern include: a six-membered carbon aromatic ring containing two substituents, an aliphatic ring system, or a three- to six-membered aromatic nitrogen heterocycle or oxygen heterocycle containing two substituents, wherein the aliphatic ring system includes a monocyclic heterocycle and a spirocyclic system containing nitrogen and oxygen, and these ring systems must have exactly two substituents, the monocyclic heterocycle is limited to a three- to six-membered aliphatic ring, and the spirocyclic ring can be one of six-membered-six-membered, six-membered-five-membered, and five-membered-five-membered, and at most one substituent in the three- to six-membered aromatic nitrogen heterocycle or oxygen heterocycle containing two substituents can be connected to the peptide bond.

[0011] In one embodiment, the adjacent structural fragments refer to fragments that can be connected by a single chemical bond or have partially overlapping chemical structures.

[0012] In one embodiment, the test of the rationality of the connection between the target protein ligand and the ubiquitin ligase ligand and the linker part includes: if the corresponding ligand target protein ligand and the ubiquitin ligase ligand do not meet the conditions, the fusion fragment is discarded, and the next fusion fragment in the sorting is selected for testing; if the fusion fragment meets the rationality requirements of the corresponding ligand, it is regarded as the linker part of the chimera, wherein the conditions are corresponding to only zero or one ligand, the number of atoms of the ligand is less than 5, and the ligand does not contain any ring structure.

[0013] In one embodiment, the analysis, identification and screening of the target protein, chimera and ubiquitin ligase after splitting include: for the chimera part, identifying each non-protein molecule through non-standard amino acid residues, limiting the number of atoms in the non-protein molecule to less than 150 and not in the form of polypeptides, and sorting them from large to small by the number of atoms, screening these non-protein molecules based on specific features, and selecting the molecule with the largest number of atoms after screening, wherein the specific feature is the ligands on both sides of the chimera molecule, which interact with different protein chains in the target protein and ubiquitin ligase respectively; if no satisfying ligand is found or all ligands are screened out, then more relaxed conditions are used to re-identify and screen the ligands to achieve identification and screening of the chimera part, wherein the more relaxed conditions are that the number of atoms in the non-protein molecule is less than 300 and can be in the form of polypeptides; On the basis of identifying the chimera part, the two main protein chains that interact with the target protein and ubiquitin ligase are identified, the protein chains that interact with these two main protein chains are recursively collected, and finally a set of protein chains on both sides is formed to realize the identification and screening of the target protein and ubiquitin ligase, and complete the analysis, identification and screening of the split target protein, chimera and ubiquitin ligase.

[0014] The present invention also provides a method for structural modeling of a protein degradation targeted chimera complex, wherein the structural modeling method uses the structural analysis system and comprises the following steps: Based on the chemical structures of target protein, target protein ligand, ubiquitin ligase protein, ubiquitin ligase protein ligand and linker as input, combined with the linker length distribution analysis module of the chimera, the relative positions of target protein and ubiquitin ligase protein are adjusted to perform protein-protein docking, simulate the interaction between proteins, and form a protein-protein docking conformation set; Based on the chimera component analysis module, a series of three-dimensional structural conformations are generated for the chimera to be modeled, and the conformations of the chimera are searched and analyzed based on the conformation search strategy to form a chimera conformation set; Based on the protein-protein docking conformation set and the chimera conformation set, matching and splicing are performed by energy minimization, random sampling and screening conformation fusion, and the matching and splicing results are optimized by conformational structure fine docking and energy optimization to form a chimera-induced ternary complex conformation set; For the chimera-induced ternary complex conformation clustering, specific members in each category were selected to form a structure set. Combined with the chimera-induced complex component analysis module, a conformational scoring and ranking method was constructed, and the structures in the result set were ranked to complete the structural modeling of the chimera complex for protein degradation targeting.

[0015] In one embodiment, the relative positions of the target protein and the ubiquitin ligase protein are adjusted to perform protein-protein docking, including: treating the protein with a larger number of atoms as a fixed protein, keeping the position of the backbone atoms unchanged during the docking process, treating the protein with a smaller number of atoms as a mobile protein, and performing conformational search for protein-protein docking through multiple Monte Carlo steps during the docking process to achieve the lowest total energy of the system and complete protein-protein docking.

[0016] Optionally, multiple Monte Carlo steps randomly select to translate the protein 0 Å~0.1 Å or rotate the protein 0°~5°, and the rotation direction can be one of the X, Y, and Z axes.

[0017] Optionally, a distance limit is imposed in the conformational search of protein-protein docking. The distance limit is defined as the energy addition of each Monte Carlo step, represented by a Gaussian function with distance as the independent variable. The mean and standard deviation of the linker length distribution are calculated based on the chimera-based linker length distribution analysis module to reduce the spatial range of the search.

[0018] Optionally, in one embodiment, the conformational search strategy includes at least one of: conformational search on the complete chimera, conformational search on the linker portion, and conformational search on the ligand portion of the linker.

[0019] In one embodiment, the energy minimization includes: extracting conformations from a protein-protein docking conformation set and analyzing the positions of protein ligands, generating arbitrary conformations of the chimera and placing them at relative positions, and performing a two-step energy minimization process, wherein the two-step energy minimization process includes: step 1 ignoring the protein part in the protein-protein docking conformation for energy minimization, while pulling the ligand in the chimera to move toward the ligand position in the protein-protein docking conformation; step 2 re-adding the protein part in the protein-protein docking conformation for energy minimization.

[0020] Optionally, the step 1 ignores the protein part in the protein-protein docking conformation to perform energy minimization, including: ignoring the protein part in the protein-protein docking conformation, retaining only the ligand part, calculating the maximum common substructure between each ligand and the chimera, and forming a mapping relationship; achieving energy minimization through multiple Monte-Carlo processes, and minimizing the mapped atomic pair distance in the form of energy addition in the process; after energy minimization, calculating the root mean square distance deviation between the chimera and the ligand and setting a threshold for conformation screening.

[0021] Optionally, the step 2 re-adds the protein part in the protein-protein docking conformation for energy minimization, including: re-adding the protein part in the protein-protein docking conformation and ignoring the ligand part in the protein-protein docking conformation, calculating the position conflict between the protein and the chimera and setting a threshold for screening, and realizing energy minimization and iterative optimization of the position through multiple Monte-Carlo processes to achieve conformation screening.

[0022] Further optionally, positional clashes are represented by calculating the overlapping volume of the van der Waals volumes between the protein and the chimera.

[0023] Further optionally, distance constraints are imposed on the chimera, pocket atom set and interface atom set in the Monte Carlo process, wherein the pocket atom set is defined as all residue atoms within 5 Å of the chimera molecule, and the interface atom set is defined as between any two protein chains A and B, there are some protein side chain atoms such that they are located in the A chain and there are B chain atoms within 10 Å.

[0024] Optionally, the iterative optimization of the position by energy minimization through multiple Monte-Carlo processes includes: minimizing the energy through multiple Monte-Carlo processes, checking the overlapping volume of the van der Waals volume, and continuing the energy minimization process if it is higher than a set threshold, otherwise outputting a merged structure.

[0025] In one embodiment, the random sampling and screening include: searching the conformation of the chimera in advance, extracting the conformation from the protein-protein docking conformation set, superimposing the conformation of the ligand and each chimera, calculating the root mean square distance deviation of the corresponding atoms between the ligand and the chimera, setting a threshold for preliminary conformation screening; calculating the position conflict between the protein and the chimera, setting a threshold for secondary screening; achieving energy minimization convergence through multiple Monte Carlo processes to achieve conformation screening.

[0026] Optionally, the setting of a threshold for preliminary conformation screening includes: setting a threshold, and if the atomic root mean square distance deviation exceeds the threshold, discarding the corresponding conformation.

[0027] Optionally, the calculating of the position conflict between the protein and the chimera and setting a threshold for secondary screening include: expressing the position conflict by calculating the overlapping volume of the van der Waals volumes between the protein and the chimera and setting a threshold for conformational screening.

[0028] Further optionally, distance constraints are imposed on the chimera, pocket atom set and interface atom set in the Monte-Carlo process, and the strength of the distance constraints is set differently for different atom sets.

[0029] Optionally, the energy minimization convergence achieved through multiple Monte Carlo processes includes: achieving energy minimization through multiple Monte Carlo processes, checking the overlapping volume of the van der Waals volume, and if the reduction in the overlapping volume compared to the previous energy minimization is less than 20%, the minimization process is deemed to have converged, and a merged structure is output; after the minimization converges, if the overlapping volume is higher than 10 cubic angstroms, the corresponding conformation is screened out.

[0030] In one embodiment, the conformation score is calculated using multiple indicators or scores of the conformation, including one or more of the following: the number of atoms of the chimera, the number of atoms of the linker, the number of movable dihedral angles of the linker, the mean and variance of the length distribution of the linker calculated after Gaussian function fitting, the docking score of the chimera, the total energy of the system after energy minimization, the pairwise binding energy components of the chimera and the two proteins, the number of member conformations of the conformational category, the ranking of the distance center conformation, the interface energy of the protein-protein interaction, the difference in solvent accessible surface area of ​​the linker in the dissociated state and the bound state, and the strain energy of the chimera drug conformation.

[0031] Compared with the prior art, the present invention has the following beneficial effects: (1) By integrating chimera component analysis, ternary complex analysis, and linker length analysis, the system can comprehensively evaluate the structural characteristics of chimeras, significantly improve the design efficiency and success rate of protein degradation targeted chimeras, and help to deeply understand the interaction mechanism between chimeras and target proteins and ubiquitin ligases, accelerate the discovery and optimization of new protein degradation drugs, and promote the development of targeted protein degradation therapies.

[0032] (2) The structural modeling method provided by the structural analysis system achieves efficient structural modeling of protein degradation targeted chimera complexes through the steps of protein-protein docking, conformation search, conformation fusion and optimization, conformation clustering, and conformation scoring and ranking. It is accurate, efficient and practical, and is of great significance to drug development and protein degradation mechanism research. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0034] Figure 1 It is a schematic diagram of the structure of the structural analysis system for protein degradation targeted chimera complex provided by the present invention.

[0035] Figure 2 Schematic diagram of the protein degradation targeting chimera and its induced ternary complex and substructures.

[0036] Figure 3 It is a schematic diagram of the structure of the chimera component analysis module provided by the present invention.

[0037] Figure 4 It is a schematic diagram of the structure of the chimera-induced complex component analysis module provided by the present invention.

[0038] Figure 5 It is a flow chart of the structural modeling method for protein degradation targeting chimera complex provided by the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the following is a The present invention is further described in detail in the following examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and do not limit the scope of protection of the present invention.

[0040] In order to improve the accuracy of the structural modeling of the chimera-induced ternary complex, the embodiment provides a structural analysis system for protein degradation targeted chimera complexes, which is used to process the structural data of a large number of protein degradation targeted chimera molecules and their complexes, including splitting, property analysis, structure generation, etc., and is used in related modeling processes or used alone or called by other computer systems for tasks such as data analysis. Figure 1 As shown, the structure analysis system includes: a chimera component analysis module, a chimera induced complex component analysis module and a chimera connector length distribution analysis module. Each part of the structure analysis system will be described in detail below.

[0041] First, the embodiment provides a chimera component analysis module. The chimera component separation mainly refers to analyzing the chemical structure of the chimera and automatically identifying and selecting its various substructures (target protein ligand, ubiquitin ligase ligand, connector). Figure 2 As shown, the protein targeted degradation chimera 100 is generally composed of three parts: a target protein ligand part 101, a ubiquitin ligase ligand part 102, and a linker part 103 connecting the two. In drug design, the target protein ligand part 101 and the ubiquitin ligase ligand part 102 are generally selected from known binding agent structures.

[0042] Therefore, after the two ligands are assembled into a chimera, the induced ternary complex still retains a three-dimensional structure similar to the original binder. In contrast, the linker portion 103 is generally more flexible, and its length and composition may significantly affect the assembly mode of the ternary complex. Therefore, it is necessary to develop an automated component analysis method to parse the chemical structure of the chimera into three substructure parts: target protein ligand, ubiquitin ligase ligand and linker, so as to realize its differentiated processing strategy in the complex modeling process.

[0043] like Figure 3 As shown, in the chimera component analysis module, a series of structural fragments 301 are first identified, specifically, first, the following substructures 302 are matched based on the molecular structure pattern: primary amide, or thioprimary amide, or a carbon atom connected to two hydrogen atoms, or a hydroxymethyl group, or an alkynyl group, or a substructure formed by a ring-containing nitrogen atom connected to a non-cyclic carbonyl group, or an azo group formed by two nitrogen atoms connected by a double bond. The above substructures can be represented by the SMARTS chemical structure search language "[NH]C=O", "[NH]C=S", "[H2&C]", "[H2&C]O", "C#C", "[R1&N][R0&C]=O", "[N]=[N]".

[0044] Second, the following substructures 303 are matched based on the ring matching pattern: a six-membered carbon aromatic ring containing two substituents, or a three- to six-membered aromatic nitrogen heterocycle or oxygen heterocycle containing two substituents, wherein at most one substituent can be connected to the peptide bond; the following substructures 304 are matched based on the ring matching pattern: an aliphatic ring system, including a monocyclic heterocycle and a spirocyclic system containing nitrogen and oxygen, and these ring systems must have exactly two substituents. The monocyclic heterocycle is limited to a three- to six-membered aliphatic ring, and the spirocyclic ring can be one of: six-membered-six-membered, six-membered-five-membered, and five-membered-five-membered.

[0045] Subsequently, an iterative method is used to fuse adjacent structural fragments, wherein the adjacent structural fragments are fragments connected by a single chemical bond or having partial chemical structure overlap. If the outermost fragment in the fused fragment is an aromatic ring, the ring is removed from the fused fragment. Subsequently, each fused fragment 305 is sorted from large to small according to the approximate bond length index, which can be calculated by the sum of the number of ring atoms and the number of non-ring atoms in the fragment: approximate bond length index = 0.5 × number of ring atoms + number of non-ring atoms. This index is used to quickly indicate the approximate length of the connector when it is fully extended. This bond length is only used to quickly determine the approximate length of the candidate connector when the chimera components are split, and it is different from the connector length distribution described later.

[0046] Then, each sorted fusion fragment 305 is tested for rationality in turn, specifically, testing whether the target protein ligand part 101 and the ubiquitin ligase ligand part 102 in the chimera meet specific requirements when the fusion fragment 305 is regarded as a connector part.

[0047] Specifically, first identify the two anchor atoms 306 of the fusion fragment 305 that covalently connect the atoms outside the fragment, and then extract the atom sets that can be extended to by covalent bonds starting from the two anchor atoms. For each anchor atom 306, the extracted atom set is further split according to connectivity, and the larger atom set is selected as the ligand 307 on this side, that is, the target protein ligand or the ubiquitin ligase ligand. In the process of taking each fusion fragment 305 as a connector 103 for testing, if the corresponding ligand 307 meets one of the following conditions, the fusion fragment is discarded, and the next fusion fragment in the sorting is selected for testing: only corresponding to zero or one ligand, or the number of atoms of the ligand is less than 5, or the ligand does not contain any ring structure. If a fusion fragment 305 meets the rationality requirements of the corresponding ligand, it is regarded as the connector part of the chimera.

[0048] In order to verify the effectiveness of the chimera component analysis module for automated chimera splitting, the chimera structure and substructure information collected manually in the existing public database ProtacDB were compared, and the error items with large differences in the number of substructure atoms and the failed items that did not recognize some substructures were analyzed, and the applicability of the above chimera component analysis module to various chimera drug molecular structures was quantitatively analyzed. As shown in Table 1, the generalizability of the chimera automated splitting method provided by the chimera component analysis module is explained, and the criticality of each step and parameter is proved by deleting some key steps or modifying some key parameters. Table 1

[0049] The embodiment provides a chimera-induced complex component analysis module, and the chimera ternary complex component separation mainly refers to analyzing the three-dimensional structure of the chimera drug-induced ternary complex, and automatically identifying and selecting its various substructures (target protein, chimera, ubiquitin ligase).

[0050] like Figure 2 As shown, the protein degradation targeting chimera ternary complex 200 is formed by the combination of target protein 201, chimera drug 100 and ubiquitin ligase 202, wherein both target protein 201 and ubiquitin ligase 202 may participate in the form of multi-protein complexes, containing multiple protein chains, and it is difficult to separate them only according to the protein chains in the structure. For example, von Hippel-Lindau (VHL) E3 ubiquitin ligase usually functions in the form of VHL-ElongB-ElangC complex. Therefore, it is necessary to develop an automated component analysis method to separate the three-dimensional molecular structure model of the chimera ternary complex into three substructure parts: target protein, chimera drug and ubiquitin ligase, each of which may contain several molecules and protein chains, so as to realize its differentiated processing strategy in the complex modeling process.

[0051] like Figure 4 As shown, the chimera part is identified in the chimera-induced complex component analysis module. Specifically, first, each non-protein molecule is identified by non-standard amino acid residues, the number of atoms in the chimera molecule is limited to less than 150, and the chimera molecule is not in the form of a polypeptide, and is sorted from large to small by the number of atoms (301). Second, these chimera molecules are screened based on a specific feature, and the chimera molecule with the largest number of atoms after screening is selected. The specific feature is the ligands on both sides of the chimera molecule, which interact with different protein chains in the target protein and ubiquitin ligase respectively.

[0052] In order to quickly calculate this specific feature, it is necessary to represent each atom and bond in the chimera molecule as an undirected graph, and use the minimum eccentricity algorithm to calculate the central atom 313 and peripheral atom set of the chimera molecule. Subsequently, the shortest path length between all peripheral atom node pairs is calculated, and the node pair corresponding to the longest length is selected as the terminal atom pair 314. Next, the shortest path length between the central atom and the two terminal atoms a and b is calculated, which is recorded as and Then, the shortest path length to the terminal atoms a and b is calculated for each atom in the molecule, recorded as and For each atom, the following formula can be used to calculate which end of the molecule the atom is biased towards: , according to The positive or negative value can split each atom in the chimera molecule into a set A biased towards the terminal atom a and a set B biased towards the terminal atom b. Then, for each set, all non-drug atoms within 4 angstroms of each atom in the set are counted, and the protein chain to which the most non-drug atoms belong is counted as the main interacting protein chain of the terminal atom set. If the main interacting protein chains of the two atom sets are different, it can be determined that the specific feature is met (315).

[0053] The first ligand that meets the specified characteristics will be identified as the chimera part in the structural model. If no ligand that meets the requirements is found or all ligands are screened out, a more relaxed condition is selected, that is, the number of atoms in the molecule is less than 300 and it can be in the form of a polypeptide, and the ligand identification and screening are repeated.

[0054] It should be noted that if this specific feature is not used for screening and the number of molecular interaction chains is directly determined, unexpected molecules may be mistakenly identified, especially peptide fragments, molecular glues or protein-protein interaction interfaces occupied by drugs that are analyzed due to precision limitations in protein structures.

[0055] Based on the identification of the chimera part, the two main protein chains with which the drug interacts can be identified in a similar manner as described above, and then the protein chains with which the two chains interact can be recursively collected, and finally a collection of protein chains on both sides is formed (316) as the target protein and ubiquitin ligase substructure parts.

[0056] The embodiment also provides a chimera connector length distribution analysis module. The connector part in the protein degradation targeting chimera is usually highly flexible and greatly affects the distance limit between the target protein and the ubiquitin ligase protein during the assembly of the ternary complex. Therefore, it is necessary to develop a chimera connector length distribution analysis method for analyzing the connector endpoint distance distribution of the chimera drug in different three-dimensional conformations as an indicator of the connector length range.

[0057] First, the molecular conformation search is performed using the linker part extracted from the chimera in the complex. Specifically, a method based on distance geometry is used to generate multiple three-dimensional conformations of the molecule. First, the interatomic distance range is defined by the distance constraint matrix, then random sampling is used to generate the initial conformation that satisfies the constraints, and finally the optimization algorithm is used to minimize the energy of the conformation. For each three-dimensional conformation, the straight-line distance between the endpoints of the linker on both sides is calculated and statistically analyzed.

[0058] Typically, the distance distribution is Gaussian, 305 but specific linker chemistry may affect the distribution. Fitting the distribution with a Gaussian function allows calculation of the mean and variance of the distance distribution as an indicator of linker length and length variability.

[0059] The ternary complex is the key structure for protein degradation targeted chimeric drugs to exert their degradation effects. The chimera forms a ternary complex by binding to the target protein and the ubiquitin ligase at the same time, allowing the ubiquitin on the ubiquitin ligase to be transferred to the cysteine ​​or lysine of the target protein and recognized and degraded by the proteasome. Therefore, ternary complex modeling is the key to the design of protein degradation targeted chimeric drugs.

[0060] Based on the above structure analysis system, the embodiment provides a structure modeling method for protein degradation targeting chimera complex. Figure 5 The following steps are shown: S1. Based on the chemical structures of target protein, target protein ligand, ubiquitin ligase protein, ubiquitin ligase protein ligand and linker as input, combined with the linker length distribution analysis module of the chimera, the relative positions of target protein and ubiquitin ligase protein are adjusted to perform protein-protein docking, simulate the interaction between proteins, and form a protein-protein docking conformation set.

[0061] In the embodiment, the chemical structures of the target protein, the target protein ligand, the ubiquitin ligase protein, the ubiquitin ligase protein ligand and the linker are used as input, and the chemical structure of a target chimera is specified in the modeling process to generate the corresponding ternary complex. The strength of the protein-protein interaction significantly affects the structural stability of the ternary complex, and thus affects the degradation activity of the protein degradation targeting chimera.

[0062] Therefore, the first step in modeling the ternary complex is to perform protein-protein docking on the target protein and the ubiquitin ligase protein. Both the target protein and the ubiquitin ligase protein used for docking should contain their respective ligand parts. The docking process can be described as placing the two protein three-dimensional structures in reasonable relative positions so that the energetic evaluation of the interaction between the proteins reaches a high level. In this process, the relative three-dimensional positions of the backbone atoms of the same protein (including the nitrogen atoms of the amino groups of each amino acid, the carbon atoms connected to the nitrogen atoms, and the carbonyl atoms in the peptide bond formed by each amino acid and the next amino acid) usually remain unchanged, while the relative three-dimensional positions of the side chain atoms (i.e., atoms other than the backbone atoms) can remain unchanged or be redistributed.

[0063] In protein-protein docking, the protein with a larger number of atoms in the target protein and the ubiquitin ligase protein will be regarded as a fixed protein, and the position of the backbone atoms will remain unchanged during the docking process, while the protein with a smaller number of atoms will be regarded as a mobile protein, and random translation or rotation will occur during the docking process. The conformational search process includes multiple Monte-Carlo steps, that is, randomly selecting a translational mobile protein of 0 Å~0.1 Å or a rotational mobile protein of 0°~5°, and the rotation direction can be one of the X, Y, and Z axes. After each transformation, the energy value based on the physical force field of the current docking conformation is calculated. If the energy of the new conformation is lower, it is accepted and added to the structure collection to continue the search process.

[0064] After several steps, the search process is stopped when the energy value decreases to a certain threshold value, and each conformation in the current structure set is selected and output in order of the total energy value of the system.

[0065] Optionally, a distance restriction is imposed during the search process of protein-protein docking to reduce the spatial scope of the search. This is because the chimera limits the distance and binding angle between the target protein and the ubiquitin ligase protein, and the distance restriction will be imposed between the ligand anchor atoms of the two proteins, and the anchor atoms can be described as: the boundary atoms between the ligand and the molecular connector of the modeled chimera. When the restriction is imposed, the mean and standard deviation of the length distribution are first calculated based on the connector length distribution analysis module of the chimera. Subsequently, the distance restriction is defined as the addition of the energy value of each Monte Carlo step, in the form of a Gaussian function with distance as the independent variable, and the mean and standard deviation are equivalent to the length mean and standard deviation of the modeled connector. The energy value addition can make the anchor atom distance in the Monte Carlo step gradually biased towards the length mean of the modeled connector, forming a protein-protein docking conformation set.

[0066] S2. Based on the chimera component analysis module, a series of three-dimensional structural conformations are generated for the chimera to be modeled, and the conformations of the chimera are searched and analyzed based on the conformation search strategy to form a chimera conformation set.

[0067] In the embodiment, a series of three-dimensional structural conformations are generated for the protein degradation targeting chimera to be modeled, and unreasonable atom-to-atom distances and three-dimensional structures inconsistent with the chemical substructure are avoided. This process is usually achieved by a series of Monte Carlo steps.

[0068] In the three-dimensional structure of the chimera, the conformation of the ligand part is usually very similar to the binding mode of the corresponding independent ligand, which can be predetermined and avoided from changing during the conformational search, while the conformation of the linker part can be freely searched. Therefore, when performing a conformational search of the chimera, any of the following strategies can be optionally used.

[0069] (1) Perform a conformational search on the complete protein chimera drug molecule without imposing other restrictions.

[0070] (2) Split the linker part for conformational search, and then assemble the ligand part on each conformation of the linker. Specifically, the linker part in the chimera is first extracted using the chimera component analysis module and split into independent molecules for conformational search. Subsequently, the three-dimensional conformations of the ligand structures on both sides are attached from the bond positions that were broken during the splitting.

[0071] (3) Search only the linker part. Specifically, the chimera component analysis module is first used to extract the atoms of the linker part in the chimera, and all atoms that do not belong to any ring system are screened. Subsequently, the dihedral angles involved in these atoms are calculated, and multiple Monte Carlo steps based on dihedral angles are used to implement conformational search.

[0072] (4) Conformational constraints are imposed on the ligand part. Specifically, the chimera component analysis module is first used to extract the atoms of the ligand parts on both sides of the molecule. In order to keep the conformation of the ligand part unchanged during the search process, the dihedral angles involved in these atoms are calculated as the constraint value, and the difference between each constrained dihedral angle and the constraint value is calculated after the Monte-Carlo step of the search as the added value of the energy value of the Monte-Carlo step. This strategy will make the conformation of the ligand part in the conformational results biased towards the original conformation, while achieving slight changes to accommodate the possible changes in the binding mode after the ligand is assembled into a chimera drug.

[0073] The three-dimensional structure collection of protein degradation targeting chimeras obtained by the above method forms a chimera conformation set.

[0074] S3. Based on the protein-protein docking conformation set and the chimera conformation set, matching and splicing are performed by conformational fusion with energy minimization, random sampling and screening, and the matching and splicing results are optimized by conformational structure fine docking and energy optimization to form a chimera-induced ternary complex conformation set.

[0075] Conformational fusion refers to matching and merging the protein-protein docking conformation set and the chimera conformation set after obtaining them to form a chimera-induced ternary complex conformation. In the protein-protein docking conformation set, each protein-protein docking conformation contains its binding ligand, which is used to find the chimera molecular conformation so that the atomic position of the ligand part is close to the atomic position of the binding ligand in the protein-protein docking conformation. There are two main ways to implement conformational fusion based on the protein-protein docking conformation set and the chimera conformation set: based on energy minimization traction and based on random sampling and screening.

[0076] Both methods can adapt to most common protein degradation targeting chimera patterns, but there are large performance differences for different chimera molecular sizes. For molecules with smaller atomic weights and fewer torsional bonds, random sampling and screening are more efficient. Conversely, energy minimization traction can avoid the problem of sampling a large number of torsional bonds at the same time and has significantly higher performance. In the examples, the methods and parameters for these two implementations are described in detail.

[0077] (1) Conformation fusion based on energy minimization: This method does not require a prior conformational search of the chimera molecule. Protein conformations are taken out from the protein-protein docking conformation set one by one, and the position of the binding ligand is analyzed. Then, 64 arbitrary conformations of the chimera molecule (including planar conformations or arbitrary three-dimensional conformations) are generated, and their center of mass is placed at the center of the center of mass of the binding ligand. Subsequently, a two-step energy minimization process is performed.

[0078] In the first step, the protein part in the protein-protein docking conformation is ignored, and only the ligand part is retained. The maximum common substructure between each ligand and the chimera molecule is calculated, and the atomic mapping relationship between the two molecules is formed. Subsequently, the Monte Carlo process of energy minimization is performed 100 times, and the distance limit between the atomic pairs in the mapping is imposed during the process. The distance limit is usually a Gaussian function with distance as the independent variable, and is used as a value-added energy value to affect the Monte-Carlo minimization process. The purpose of this restriction is to pull the ligand part in the chimera in the above process so that it moves to the ligand position in the docking conformation. After minimization, it is necessary to calculate the root mean square distance deviation between the two, and screen out with 3 angstroms as the threshold.

[0079] In the second step, the protein part of the protein-protein docking conformation is re-added, and the ligand part of the protein-protein docking conformation is ignored. First, the position conflict between the protein structure and the chimera molecule structure is calculated, calculated by the overlapping volume of the van der Waals volumes of the two, and filtered out with a threshold of 150 cubic angstroms. Subsequently, several atom sets are defined to define different distance constraints: the pocket atom set is defined as all residue atoms within 5 Å of the chimera molecule, and the interface atom set is defined as between any two protein chains A and B, there are some protein side chain atoms, so that they are located in the A chain and there are B chain atoms within 10 angstroms. The system is subjected to 100 energy minimization Monte Carlo processes, and the distance constraints for the chimera molecule, pocket atom set and interface atom set are optionally applied during the process so that they do not deviate significantly from the initial position. Subsequently, the overlapping volume of the van der Waals volume is re-checked. If it is higher than the threshold of 5 cubic angstroms, the above energy minimization process is performed again based on this structure. Otherwise, the merged structure is output as the ternary complex conformation. If the overlap volume threshold was still not reached after 10 energy minimization processes, the conformation was discarded.

[0080] This method is particularly suitable for cases where the complexity of the binding interface cavity is low and the chimera molecule has a large number of rotatable bonds. In such cases, the efficiency can be improved by 10 to 100 times compared with the random sampling and screening-based methods.

[0081] (2) Conformational fusion based on random sampling and screening: This method requires a conformational search of the chimera molecule in advance. Conformations are taken out from the protein-protein docking conformation set one by one, and then the ligand part and the conformation of each chimera molecule are superimposed. Superposition is a method to minimize the distance between specific atomic pairs in two structures. In this process, each structure is a rigid body, that is, the conformation remains unchanged. Subsequently, the root mean square distance deviation of the corresponding atoms between the ligand and the chimera molecule in the protein-protein docking conformation is calculated. If the deviation is higher than 1 Å, the chimera molecule conformation is discarded.

[0082] Subsequently, the positional conflicts between the protein structure and the chimera molecular structure were calculated based on the overlapping volume of their van der Waals volumes and screened out with a threshold of 300 cubic angstroms.

[0083] Define pocket side chain and backbone atom sets for defining different distance restraints: The pocket atom set is defined as all residue atoms within 5 Å of the chimera molecule, where side chain atoms and backbone atoms are assigned to different atom sets, respectively.

[0084] The system is subjected to up to 5 energy minimizations, each containing 20 Monte Carlo processes. During the process, distance constraints are optionally applied to the chimera molecule, pocket side chain atom set, and backbone atom set so that they do not deviate significantly from the initial position. The strength of the distance constraint can be set differently for different atom sets. Usually, the backbone atoms are frozen, a constraint of 1.0 relative units is imposed on the chimera molecule, and a constraint of 0.5 relative units is imposed on the side chain atoms. After each energy minimization, the overlapping volume between the protein structure and the chimera molecule structure is rechecked. If the reduction in the overlapping volume compared to the previous energy minimization is less than 20%, the minimization process is considered to have converged and the next energy minimization is not performed.

[0085] After the minimization process converges, or after exceeding the maximum number of energy minimization thresholds, the overlapping volume between the protein structure and the chimera molecular structure is recalculated. If it is higher than 10 cubic angstroms, the conformation is discarded. Otherwise, the merged structure is output as the chimera-induced ternary complex conformation.

[0086] This method is particularly suitable for cases where the binding interface is narrow and the chimera molecule has few rotatable bonds. In such cases, the energy minimization-based traction method may not be able to search for conformations in a short time, and this method is more efficient in finding reasonable conformations.

[0087] Further structural optimization and energy minimization can be performed on the ternary complex conformation induced by the fused chimera.

[0088] Define pocket side chain and backbone atom sets for defining different distance restraints: The pocket atom set is defined as all residue atoms within 5 angstroms of the chimera molecule, where side chain atoms and backbone atoms are assigned to different atom sets, respectively.

[0089] Define interface side chain and backbone atom sets for defining different distance constraints: The side chain atom set is defined as any two protein chains A and B, where there are some residue atoms such that they are located in chain A and there are atoms in chain B within 10 Å, where side chain atoms and backbone atoms are assigned to different atom sets, respectively.

[0090] For each ternary complex conformation after fusion, a Monte Carlo process of energy minimization of up to 100 times is performed, and distance constraints are optionally applied to the ligand part and the linker part of the chimera molecule, the pocket side chain and the backbone atom set, and the interface side chain and the backbone atom set so that they do not deviate significantly from the initial position. The strength of the distance constraint can be set differently for different sets of atoms, usually freezing the backbone atoms of the pocket and the interface, applying a constraint of 1.0 relative units to the ligand part of the chimera molecule, and applying a constraint of 0.5 relative units to the side chain atoms of the pocket and the interface and the linker part of the chimera molecule.

[0091] This energy minimization step outputs the optimized structure and total system energy, thereby forming a set of chimera-induced ternary complex conformations.

[0092] S4. Cluster the chimera-induced ternary complex conformations, select specific members in each category to form a structure set, combine the chimera-induced complex component analysis module, construct a conformational scoring and ranking method, rank the structures in the result set, and complete the structural modeling of the chimera complex for protein degradation targeting.

[0093] In the embodiment, the chimera-induced ternary complex conformation set is clustered using the common contact point ratio clustering algorithm (FCC) to select representative conformations therein. Specifically, during clustering, 75% common contact similarity is used as the class threshold, and the minimum number of conformations in each conformation class is limited to 4. After clustering, the number of members in each class is calculated, and the central conformation of each class is selected, wherein the central conformation of a class is defined as the conformation with the highest sum of common contact similarities with other conformations in the class.

[0094] The conformation of the chimera molecule is not considered in the conformational clustering process because the linker part is highly flexible and the ligand part is anchored to the protein position, so no other clustering indicators are set outside the protein position.

[0095] The structural modeling method provided in the embodiment predicts 10 to 1000 three-dimensional structures. Among them, it can be expected that there are reasonable results similar to the structure obtained by the crystal diffraction experiment of the chimera-induced ternary complex (hereinafter referred to as the crystal structure), and there are also unreasonable results with large deviations. Therefore, a conformation scoring and sorting method is constructed to sort the various structures in the result set, and the modeling prediction results can be further optimized, and the final result conformation can be used in other new drug development tasks such as molecular design and optimization to complete the structural modeling of the chimera complex targeted for protein degradation.

[0096] In the embodiment, the conformation score needs to be calculated using multiple indicators or scores of the conformation, including one or more of the following: the number of atoms of the chimeric drug, the number of atoms of the linker, the number of movable dihedral angles of the linker, the mean and variance calculated after the length distribution of the linker is fitted by a Gaussian function, the docking score of the chimeric drug, the total energy of the system after energy minimization, the pairwise binding energy components of the chimeric drug and the two proteins, the number of member conformations of the conformational class, the ranking of the distance center conformation, the interface energy of the protein-protein interaction, the difference in solvent accessible surface area between the linker in the dissociated state and the bound state, and the strain energy of the chimeric drug molecule conformation. The sources or calculation methods of these indicators or scores are described in detail below: (1) The number of atoms in the chimera can be intuitively determined from the specified chemical structure, and hydrogen atoms are not included in the calculation.

[0097] (2) The number of atoms in the linker can be extracted by the chimera component analysis module, and the number of atoms can be intuitively calculated based on the chemical structure, without hydrogen atoms in the calculation.

[0098] (3) The number of movable dihedral angles of the linker is achieved through a publicly available chemical structure analysis method. The principle is to calculate whether the bonds of the aromatic ring system and the conjugated system are twistable, thereby determining the movable dihedral angle.

[0099] (4) The length distribution of the linker and its fitting can be obtained according to the linker length distribution analysis module of the aforementioned chimera.

[0100] (5) The docking score of the chimera is calculated by using the protein-drug molecular docking method to calculate the docking score of the chimera molecule in situ docking to the target protein-ubiquitin ligase complex structure. In situ docking means that the molecular conformation search in the docking process is not performed, but the molecular docking score is directly performed using the position of the chimera molecule in the ternary complex conformation.

[0101] (6) The total energy of the system after energy minimization can be calculated and output during the conformational structure optimization and energy minimization process.

[0102] (7) The calculation of binding energy components can be described as follows: For two structures A and B, each structure may include one or more protein chains or molecules. When there is only one structure in the calculation system, the total energy of the system is and , and then calculate the total energy of the system when the two are combined , then the binding energy component between the two can be defined as Four sets of binding energy components between the chimera and ubiquitin ligase, the chimera and the target protein, the ubiquitin ligase and the target protein, and the complex of the chimera and the ubiquitin ligase and the target protein can be calculated.

[0103] (8) The number of member conformations of a conformational class can be calculated and output during the conformational clustering process.

[0104] (9) The calculation of the ranking of the distance to the central conformation during the clustering process can refer to the calculation method of the central conformation during the conformational clustering process.

[0105] (10) The calculation of protein-protein interaction interface energy is usually performed after conformational structure optimization and energy minimization. Compared with the protein-protein binding energy component, this indicator has a stronger adaptability to the predicted structure of protein-protein docking.

[0106] (11) The target protein and the ubiquitin ligase usually form a cavity when they bind to each other. The linker of the chimera is usually located in the cavity. The difference in solvent accessible surface area between the dissociated state and the bound state can be used to determine the degree to which the linker part is close to the surface of the cavity pocket. The solvent accessible surface area refers to the surface area of ​​the protein molecule that can be contacted by the solvent. It can be calculated by the rolling ball algorithm, which uses a sphere of a specific radius to simulate the solvent to detect the surface of the biological molecule.

[0107] (12) The calculation of the strain energy of the chimera molecular conformation can be described as follows: first, a molecular conformation search is performed, simulating water as a solvent, the total energy of the system of the lowest energy conformation is calculated, and the difference between the energy of the current molecular conformation and the energy of the current molecular conformation is calculated. This energy difference can be used as strain energy and characterizes the possibility of the molecular conformation occurring naturally in the polar solvent state.

[0108] In the embodiment, the conformation score is used to characterize the similarity between the conformation and the crystal structure. Specifically, it is characterized by the difference in the binding mode between the target protein and the ubiquitin ligase protein. In order to further quantify this similarity, three key interface quality indicators need to be calculated: the root mean square deviation of the interface main chain atoms (iRMS), the root mean square deviation of the ligand protein main chain atoms (LRMS), and the reproducibility of protein-protein interactions (Fnat). These indicators evaluate the degree of match between the conformation and the crystal structure from different angles.

[0109] Among them, iRMS measures the main chain atoms (such as A lower iRMS value indicates that the conformation has a higher similarity with the crystal structure in the main chain conformation and a better match in the interface region. LRMS measures the main chain atomic deviation of the ligand protein (i.e., ubiquitin ligase protein) in the entire complex. It not only focuses on the interface region, but also evaluates the deviation of the main chain conformation of the ligand protein as a whole from the crystal structure. A lower LRMS value indicates that the overall conformation of the ligand protein is relatively consistent with the crystal structure. Fnat measures the reproduction rate of the real contact between the target protein and the ubiquitin ligase protein in the conformation. It is evaluated by calculating the matching ratio of the predicted interface atom pairs to the actual interface atom pairs in the crystal structure. A higher Fnat value indicates that the conformation better reproduces the real interaction mode between proteins.

[0110] By calculating these indicators, the degree of accurate matching of the conformation relative to the crystal structure can be obtained. However, iRMS, LRMS or Fnat alone may not fully reflect the quality of the conformation because they focus on different aspects. Therefore, the DockQ algorithm is used to calculate the comprehensive score to more comprehensively evaluate the quality of the conformation. The calculation formula is as follows: , , , in, and is a predefined threshold parameter, represents the root mean square deviation normalized by a predefined parameter, Therefore, it can be calculated by iRMS, LRMS and Fnat and used to evaluate the similarity between the modeled conformation and the crystal structure. The result is between 0 and 1, with a larger The value indicates that the modeled conformation has a higher similarity with the crystal structure.

[0111] For each conformation output by a modeling, The scoring training target value of the conformation It can be expressed as the following formula, that is, the conformation score Cubic and Optimal Conformation The ratio of the cube: , Similarity to the crystal structure can also include conformational and positional differences in the chimeric molecule, but since the linker is usually highly flexible and the ligand part of the chimeric molecule is already anchored in the original ligand binding site of the protein, this is usually not considered additionally.

[0112] In order to train the weights of each indicator in the conformation score through the model, conformation score training data needs to be collected in the embodiment. The ternary complex structure of the known crystal structure in Table 2 below is pre-split, the chimera-induced ternary complex is modeled, and each indicator or score of each conformation is calculated according to the above method, and the similarity between the conformation and the crystal structure is calculated at the same time. The known crystal structure 701 can be obtained from PDB (Protein Data Bank) and has a corresponding database number.

[0113] Table 2

[0114] For each known crystal structure, the complex structure needs to be automatically split to generate input data for confirmatory modeling. Based on the chimera-induced complex component analysis module, the target protein, ubiquitin ligase, and protein targeted degradation chimera in the ternary complex are identified. The above-mentioned chimera component analysis module is further used to identify the linker part in the protein targeted degradation chimera and remove it, leaving only the ligand part. The two proteins and their ligand parts are randomly moved and rotated so that the distance between them is at least 30 Å to ensure that the original binding mode between the proteins cannot be passed to the modeling algorithm through the input data.

[0115] Subsequently, the generated input data was used for validation modeling. For each conformation, each indicator and score was calculated and stored, and its similarity with the crystal structure was calculated and stored as a training data set for constructing the conformation score.

[0116] Since there are a large number of conformations in the data set that are not similar to the crystal structure, but there are fewer similar conformations, there is an imbalance in the data set, which may affect the training accuracy. Therefore, the molecular dynamics method is used for data enhancement. For each ternary complex modeling result set, the three conformations with the highest similarity to the crystal structure are selected for molecular dynamics simulation, and the structure after dynamic perturbation is extracted.

[0117] Specifically, the structure was placed in an aqueous solution environment, 0.15 mol / L sodium chloride ions were added to simulate the physiological environment, and a periodic water box was constructed using the TIP3P water model. The system first performed 5000 steps of energy minimization, followed by a 10 nanosecond molecular dynamics simulation under the NPT ensemble. During this process, a Langevin thermostat was used to maintain the temperature at 300 Kelvin and the pressure at 1 atmosphere. The dynamics calculation time step was 2 femtoseconds, and all hydrogen bonds were constrained using the SHAKE algorithm. In the last 5 nanoseconds of the simulated dynamic trajectory, a frame of conformation was taken every 10 picoseconds, and these conformations were clustered for RMSD (Root Mean Square Deviation). The number of clusters was set to 10, and the central conformation of each class was selected as the representative conformation. These representative conformations were energy minimized, and the resulting conformations were used as supplementary samples for data enhancement, thereby expanding the number of conformations in the data set that were similar to the crystal structure. This collection process can form a data set containing 8524 modeled conformations and their indicators.

[0118] Next, the random forest algorithm was used to train the conformational score. Before training, the conformational index dataset was standardized, and the Z-score standardization method was used to eliminate the dimensional differences between different features to ensure that each feature had the same weight in the model training. During model training, the number of decision trees was set to 500, the maximum number of features was set to the square root of the total number of features, the maximum depth was set to 10, and the minimum number of sample leaf nodes was set to 2 to avoid too small leaf nodes causing the model to be sensitive to noise. Stratified cross-validation technology was used to ensure that the model could achieve good performance on different data subsets. An early stopping strategy was used to avoid overfitting.

[0119] Finally, through the above-mentioned training process of step-by-step iteration, optimization and evaluation, a robust random forest model with good generalization ability was constructed, and the mean square error (MSE) of conformation score prediction was 0.11. This model can accurately predict the multi-factor score of new conformations, thus providing reliable data support for the ranking of ternary complex conformation sets.

[0120] To determine the relationship between conformational scores and ranking accuracy, a ternary complex model was modeled using the PDB database entry numbered 6HR2 as a reference, which is a ternary complex structure and is not involved in the data set. Ranking conformations by their predicted scores has a stronger ranking performance than ranking only by protein-protein interaction energy or score, or only by the size of ternary complex conformation clusters, as shown in Table 3.

[0121] Table 3

[0122] The data in Table 3 show that the first few models ranked using conformational scoring have a higher structural similarity with the 6HR2 reference structure, especially in the key interaction regions, and the ranking results of conformational scoring can better reflect the true structural characteristics. This result shows that conformational scoring can not only effectively distinguish the energy states of different conformations, but also comprehensively consider factors such as structural rationality and interaction complexity, thereby providing more accurate ranking results.

[0123] Further analysis shows that the method of sorting using conformational scores can effectively distinguish the quality of different conformations, especially in the complex interactions of ternary complexes, and can capture more subtle structural differences. In contrast, the method that relies solely on protein-protein interaction energy or scores for sorting, although it can provide reasonable sorting results in some cases, often fails to fully consider the impact of all interactions when dealing with complex ternary complexes, resulting in a decrease in the accuracy of the sorting results. In addition, the method of sorting only by cluster size, although it can reflect the prevalence of certain conformations, cannot accurately evaluate the structural rationality and energy characteristics of the conformations.

[0124] In summary, conformational scoring shows significant advantages in the ranking of ternary complexes and can provide a more reliable tool for structural biology research, helping researchers to more accurately identify the optimal conformational model in complex protein-protein interactions for structural modeling of protein degradation-targeted chimera complexes.

[0125] In order to verify the effectiveness of the structural analysis system and modeling method for protein degradation targeted chimera complex provided by the present invention, the examples were used to develop chimera drugs for protein targeted degradation.

[0126] The structural modeling method for protein degradation targeting chimera complexes of the present invention can be applied to the research and development of protein targeting degradation chimera drugs for protein kinase B (AKT). In the initial stage, a lead compound with weak degradation activity was obtained. In order to improve the degradation efficiency of the compound, the structural modeling method of the present invention was used to conduct an in-depth analysis of the interaction between AKT and ubiquitin ligase.

[0127] By modeling the ternary complex of the lead compound with AKT and E3 ligase, it was found that there were phenylalanine groups on the surface of the adjacent protein in the spatial path of the linker, which could form potential π-π interactions through chemical structure design. This interaction is considered to be a key factor in improving the degradation activity of the drug. Therefore, the structure of the linker part of the drug was optimized, and a benzene ring structure was introduced at a suitable position, thereby enhancing the interaction between the linker and the protein-protein interaction cavity and improving the stability of the ternary complex.

[0128] The optimized chimera showed significantly improved degradation activity in cell experiments. Compared with the unoptimized lead compound, the optimized molecule can achieve effective degradation of AKT at a lower concentration. These results fully demonstrate the effectiveness of the ternary complex modeling method of the present invention in guiding the molecular structure optimization of protein targeted degradation chimeras and improving degradation efficiency.

[0129] The specific implementation methods described above provide a detailed description of the technical solutions and beneficial effects of the present invention. It should be understood that the above is only the most preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, supplements and equivalent substitutions made within the scope of the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A structural analysis system for a protein degradation targeted chimera complex, characterized in that: The structural analysis system includes: a chimera component analysis module, a chimera induced complex component analysis module and a chimera connector length distribution analysis module; The chimera component analysis module is used to identify a series of structural fragments in the chimera, and fuse adjacent structural fragments using an iterative method; sort each fused fragment, test the rationality of the connection between the target protein ligand and the ubiquitin ligase ligand and the sorted fused fragments, and regard the rationally sorted fused fragments as the connector part to achieve the splitting and component analysis of the chimera; The chimera-induced complex component analysis module is used to split the chimera-induced ternary complex, and analyze, identify and screen the split target protein, chimera and ubiquitin ligase to achieve component analysis of the chimera-induced complex; The connector length distribution analysis module of the chimera is used to perform molecular conformation search on the connector part extracted from the chimera in the complex, analyze the connector endpoint distance distribution when the chimera is in different three-dimensional conformations, and use it as an indicator of the connector length range for structural analysis of protein degradation targeted chimera complexes.

2. The structural analysis system according to claim 1, characterized in that: The identification of a series of structural fragments in the chimera molecule comprises: identifying structural fragments based on a molecular structure pattern or based on a loop matching pattern; The structural fragments matched by the molecular structure pattern include: primary amide, primary thioamide, a carbon atom connected to two hydrogen atoms, a hydroxymethyl group, an alkynyl group, a substructure formed by a ring-containing nitrogen atom connected to a non-cyclic carbonyl group, or an azo group formed by two nitrogen atoms connected by a double bond; The structural fragments of the ring matching pattern include: a six-membered carbon aromatic ring containing two substituents, an aliphatic ring system, or a three- to six-membered aromatic nitrogen heterocycle or oxygen heterocycle containing two substituents, wherein the aliphatic ring system includes a monocyclic heterocycle and a spirocyclic system containing nitrogen and oxygen, and these ring systems must have exactly two substituents, the monocyclic heterocycle is limited to a three- to six-membered aliphatic ring, and the spirocyclic ring can be one of six-membered-six-membered, six-membered-five-membered, and five-membered-five-membered, and at most one substituent in the three- to six-membered aromatic nitrogen heterocycle or oxygen heterocycle containing two substituents can be connected to the peptide bond.

3. The structural analysis system according to claim 1, characterized in that: The adjacent structural fragments refer to fragments that can be connected by a single chemical bond or have partially overlapping chemical structures.

4. The structural analysis system according to claim 1, characterized in that: The test of the rationality of the connection between the target protein ligand and the ubiquitin ligase ligand and each fusion fragment includes: if the corresponding ligand target protein ligand and the ubiquitin ligase ligand do not meet the conditions, the fusion fragment is discarded, and the next fusion fragment in the sorting is selected for testing; if the fusion fragment meets the rationality requirements of the corresponding ligand, it is regarded as the connector part of the chimera, wherein the conditions are corresponding to only zero or one ligand, the number of atoms of the ligand is less than 5, and the ligand does not contain any ring structure.

5. The structural analysis system according to claim 1, characterized in that: The analysis, identification and screening of the target protein, chimera and ubiquitin ligase after splitting include: for the chimera part, identifying each non-protein molecule through non-standard amino acid residues, limiting the number of atoms in the non-protein molecule to less than 150 and not in the form of polypeptides, and sorting them from large to small by the number of atoms, screening these non-protein molecules based on specific features, and selecting the molecule with the largest number of atoms after screening, wherein the specific feature is the ligands on both sides of the chimera molecule, which interact with different protein chains in the target protein and ubiquitin ligase respectively; if no satisfying ligand is found or all ligands are screened out, re-identifying and screening the ligands under more relaxed conditions to achieve identification and screening of the chimera part, wherein the more relaxed condition is that the number of atoms in the non-protein molecule is less than 300 and can be in the form of polypeptides; On the basis of identifying the chimera part, the two main protein chains that interact with the target protein and ubiquitin ligase are identified, the protein chains that interact with these two main protein chains are recursively collected, and finally a set of protein chains on both sides is formed to realize the identification and screening of the target protein and ubiquitin ligase, and complete the analysis, identification and screening of the split target protein, chimera and ubiquitin ligase.

6. A method for structural modeling of a protein degradation targeted chimera complex, characterized in that: The structural modeling method uses the structural analysis system according to any one of claims 1 to 5, comprising the following steps: Based on the chemical structures of target protein, target protein ligand, ubiquitin ligase protein, ubiquitin ligase protein ligand and linker as input, combined with the linker length distribution analysis module of the chimera, the relative positions of target protein and ubiquitin ligase protein are adjusted to perform protein-protein docking, simulate the interaction between proteins, and form a protein-protein docking conformation set; Based on the chimera component analysis module, a series of three-dimensional structural conformations are generated for the chimera to be modeled, and the conformations of the chimera are searched and analyzed based on the conformation search strategy to form a chimera conformation set; Based on the protein-protein docking conformation set and the chimera conformation set, matching and splicing are performed by energy minimization, random sampling and screening conformation fusion, and the matching and splicing results are optimized by conformational structure fine docking and energy optimization to form a chimera-induced ternary complex conformation set; For the chimera-induced ternary complex conformation clustering, specific members in each category were selected to form a structure set. Combined with the chimera-induced complex component analysis module, a conformational scoring and ranking method was constructed, and the structures in the result set were ranked to complete the structural modeling of the chimera complex for protein degradation targeting.

7. The structural modeling method according to claim 6, characterized in that: The adjustment of the relative positions of the target protein and the ubiquitin ligase protein for protein-protein docking includes: treating the protein with a larger number of atoms as a fixed protein, keeping the position of the skeleton atoms unchanged during the docking process, treating the protein with a smaller number of atoms as a mobile protein, performing random translation or rotation during the docking process, and performing position transformation through multiple Monte-Carlo steps to achieve the lowest total energy of the system and complete the protein-protein docking.

8. The structural modeling method according to claim 6, characterized in that: The energy minimization includes: extracting conformations from the protein-protein docking conformation set and analyzing the ligand position, generating an arbitrary conformation of the chimera and placing the center of mass of the chimera at the center point of the ligand center of mass, and performing a two-step energy minimization process, wherein the two-step energy minimization process includes: step 1 ignoring the protein part in the protein-protein docking conformation for energy minimization, and at the same time pulling the ligand in the chimera to move to the ligand position in the protein-protein docking conformation; step 2 re-adding the protein part in the protein-protein docking conformation for energy minimization.

9. The structural modeling method according to claim 6, characterized in that: The random sampling and screening include: searching the conformation of the chimera in advance, extracting the conformation from the protein-protein docking conformation set, superimposing the conformation of the ligand and each chimera, calculating the root mean square distance deviation of the corresponding atoms between the ligand and the chimera, setting a threshold for preliminary conformation screening; calculating the position conflict between the protein and the chimera, setting a threshold for secondary screening; performing multiple Monte-Carlo processes to achieve energy minimization convergence, and completing the screening.

10. The structural modeling method according to claim 6, characterized in that: The conformational score is calculated using multiple indicators or scores of the conformation, including one or more of the following: the number of atoms in the chimera, the number of atoms in the linker, the number of movable dihedral angles of the linker, the mean and variance of the length distribution of the linker calculated after Gaussian function fitting, the docking score of the chimera, the total energy of the system after energy minimization, the pairwise binding energy components of the chimera and the two proteins, the number of member conformations in the conformational category, the ranking of the distance to the central conformation, the protein-protein interaction interface energy, the difference in solvent accessible surface area of ​​the linker in the dissociated and bound states, and the strain energy of the chimera conformation.

Images