Molecular docking method, device and electronic equipment

Abstract

Embodiments of the present invention provide a molecular docking method, apparatus, and electronic device. The method includes: obtaining a first spatial structure of a small molecule compound and a second spatial structure of a target protein; determining a first pharmacophore present in the first spatial structure and a second pharmacophore present in the second spatial structure; establishing a first pharmacophore model for representing the distance between each first pharmacophore and each first pharmacophore, and establishing a second pharmacophore model for representing the distance between each second pharmacophore and each second pharmacophore; and determining, using a quantum approximate optimization algorithm, a combination of the first pharmacophore model and the second pharmacophore model in which the score satisfies a preset screening condition, wherein the score is related to the distance between the bound pharmacophores when the first pharmacophore model and the second pharmacophore model are combined. This can improve the efficiency of virtual screening.

Description

Technical Field

[0001] The present invention relates to the field of quantum computing technology, and in particular to a molecular docking method, device and electronic equipment. Background Art

[0002] During drug development, it's necessary to study the interaction patterns between small molecules and target proteins, thereby identifying small molecules that can rationally bind to the target protein. This process is known as virtual screening. For example, in related technologies, computer simulations can be used to perform molecular docking between small molecules and target proteins. By comprehensively analyzing scores and spatial conformational properties, such as hydrogen bonding, hydrophobic interactions, and van der Waals interactions, the specific interaction and binding modes between ligand small molecules and receptor biomacromolecules can be explored, explaining the reasons for the compound's activity and enabling virtual screening of small molecules.

[0003] However, the screening process of small molecule compounds often involves a large number of small molecules, and the computing performance of traditional computers is low, so molecular docking takes a long time, which leads to low virtual screening efficiency. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a molecular docking method, apparatus, and electronic device to improve the efficiency of virtual screening. The specific technical solution is as follows:

[0005] In a first aspect of the present application, a molecular docking method is provided, comprising:

[0006] Obtain the first spatial structure of the small molecule compound and the second spatial structure of the target protein;

[0007] determining a first pharmacophore present in the first spatial structure and a second pharmacophore present in the second spatial structure;

[0008] establishing a first pharmacophore model for representing each of the first pharmacophores and the distance between each of the first pharmacophores, and establishing a second pharmacophore model for representing each of the second pharmacophores and the distance between each of the second pharmacophores;

[0009] A quantum approximate optimization algorithm is used to determine a combination mode in which the score between the first pharmacophore model and the second pharmacophore model meets a preset screening condition, wherein the score is related to the distance between the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are combined.

[0010] In a possible embodiment, determining the combination of the first pharmacophore model and the second pharmacophore model by a quantum approximate optimization algorithm, wherein the scores of the first pharmacophore model and the second pharmacophore model satisfy a preset screening condition, includes:

[0011] A quantum approximate optimization algorithm is used to determine a binding mode in which the score between the first pharmacophore model and the second pharmacophore model meets a preset screening condition and the binding relationship between each of the first pharmacophore and the second pharmacophore meets a preset restriction condition.

[0012] In a possible embodiment, the preset restriction condition includes any one or more of the following conditions:

[0013] Binding frequency condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, the number of times the first target pharmacophore binds to the other pharmacophore is no greater than a preset number threshold;

[0014] Binding scoring condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, if the second target pharmacophore is bound to the third target pharmacophore, the score item corresponding to the distance between the second target pharmacophore and the third target pharmacophore is a preset score.

[0015] In a possible embodiment, the score of each combination mode is determined by:

[0016] determining a binding distance between the first pharmacophore and the second pharmacophore in each pharmacophore pair when the first pharmacophore model and the second pharmacophore model are combined in the binding manner, wherein the pharmacophore pair is composed of the first pharmacophore and the second pharmacophore of the same type;

[0017] Determine the score item corresponding to each of the combined distances;

[0018] The score of the combination method is determined according to each of the score items.

[0019] In a possible embodiment, the method further includes:

[0020] Determining, based on a preset correspondence between a pharmacophore combination and a penalty term, a penalty term corresponding to the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are bound in the binding manner;

[0021] Determining the score of the combination according to each of the score items includes:

[0022] The score of the combination is determined according to the score item and the determined penalty item.

[0023] In a possible embodiment, the corresponding relationship is determined in advance in the following manner:

[0024] The penalty term corresponding to each pharmacophore combination is determined based on the binding mode of each pharmacophore in the cocrystal.

[0025] In a possible embodiment, the small molecule compound is a quercetin molecule, and the target protein is a human ACE2 protein.

[0026] In a second aspect of the present application, a molecular docking device is provided, comprising:

[0027] A spatial structure acquisition module is used to obtain the first spatial structure of the small molecule compound and the second spatial structure of the target protein;

[0028] a pharmacophore determination module, configured to determine a first pharmacophore present in the first spatial structure and a second pharmacophore present in the second spatial structure;

[0029] a model building module, configured to establish a first pharmacophore model for representing each of the first pharmacophores and the distance between each of the first pharmacophores, and to establish a second pharmacophore model for representing each of the second pharmacophores and the distance between each of the second pharmacophores;

[0030] A quantum approximate optimization module is used to determine, through a quantum approximate optimization algorithm, a combination mode between the first pharmacophore model and the second pharmacophore model in which the score satisfies a preset screening condition, wherein the score is related to the distance between the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are combined.

[0031] In a possible embodiment, the quantum approximate optimization module determines, by using a quantum approximate optimization algorithm, a combination mode in which the scores of the first pharmacophore model and the second pharmacophore model satisfy a preset screening condition, including:

[0032] A quantum approximate optimization algorithm is used to determine a binding mode in which the score between the first pharmacophore model and the second pharmacophore model meets a preset screening condition and the binding relationship between each of the first pharmacophore and the second pharmacophore meets a preset restriction condition.

[0033] In a possible embodiment, the preset restriction condition includes any one or more of the following conditions:

[0034] Binding frequency condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, the number of times the first target pharmacophore binds to the other pharmacophore is no greater than a preset number threshold;

[0035] Binding scoring condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, if the second target pharmacophore is bound to the third target pharmacophore, the score item corresponding to the distance between the second target pharmacophore and the third target pharmacophore is a preset score.

[0036] In a possible embodiment, the apparatus further includes a scoring module configured to determine a score for each of the combination methods in the following manner:

[0037] determining a binding distance between the first pharmacophore and the second pharmacophore in each pharmacophore pair when the first pharmacophore model and the second pharmacophore model are combined in the binding manner, wherein the pharmacophore pair is composed of the first pharmacophore and the second pharmacophore of the same type;

[0038] Determine the score item corresponding to each of the combined distances;

[0039] The score of the combination method is determined according to each of the score items.

[0040] In a possible embodiment, the scoring module is further configured to determine, based on a preset correspondence between a pharmacophore combination and a penalty term, a penalty term corresponding to the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are bound to each other in the binding manner;

[0041] The scoring module determines the score of the combination method according to each of the scoring items, including:

[0042] The score of the combination is determined according to the score item and the determined penalty item.

[0043] In a possible embodiment, the corresponding relationship is determined in advance in the following manner:

[0044] The penalty term corresponding to each pharmacophore combination is determined based on the binding mode of each pharmacophore in the cocrystal.

[0045] In a possible embodiment, the small molecule compound is a quercetin molecule, and the target protein is a human ACE2 protein.

[0046] In a third aspect of the present application, an electronic device is provided, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0047] Memory for storing computer programs;

[0048] The processor is configured to implement any of the method steps described in the first aspect when executing a program stored in the memory.

[0049] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any of the method steps described in the first aspect above is performed.

[0050] Beneficial effects of the embodiments of the present invention:

[0051] The molecular docking method, apparatus, and electronic device provided by the embodiments of the present invention can construct a pharmacophore model based on the spatial structure of a small molecule compound and a target protein, using a pharmacophore that can effectively characterize the interaction between the small molecule compound and the target protein. Since the pharmacophore model is used to represent pharmacophores and the distances between pharmacophores, it can effectively characterize the original spatial structure. At the same time, if the pharmacophores are regarded as nodes in a graph theory model, the distances between pharmacophores can be equivalently regarded as the relationship between nodes, that is, edges in the graph theory model. Therefore, the pharmacophore model is equivalent to a node-edge graph theory model, and the molecular docking problem is transformed into an optimal combination problem of two graph theory models. Therefore, the QAOA algorithm can be used to determine the optimal combination of the first pharmacophore model and the second pharmacophore model. The combination of the graph theory models can be regarded as the binding mode of the small molecule compound and the target protein. Therefore, the binding mode determined by the present application that meets the pre-screening conditions can be regarded as the most likely binding mode of the small molecule compound and the target protein, that is, the present application effectively achieves molecular docking. Furthermore, since the present application realizes molecular docking through the QAOA algorithm, it can fully utilize the powerful computing power of quantum computers, effectively shorten the time consumption of molecular docking, and thus improve the efficiency of virtual screening.

[0052] Of course, it is not necessary to achieve all of the advantages described above simultaneously in order to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0054] Figure 1 A schematic diagram of a process of the molecular docking method provided in this application;

[0055] Figure 2 Schematic diagram of the docking of quercetin and human ACE2 protein provided in this application;

[0056] Figure 3a A schematic diagram of the pharmacophore present in quercetin provided in this application;

[0057] Figure 3b Another schematic diagram of the pharmacophore present in quercetin provided for this application;

[0058] Figure 4 A schematic diagram of the pharmacophore present in the human ACE2 protein provided in this application;

[0059] Figure 5a A schematic diagram of the interaction between quercetin and the pharmacophore in human ACE2 protein provided in this application;

[0060] Figure 5b A schematic diagram of the interaction between quercetin and the pharmacophore in human ACE2 protein provided in this application at another scale;

[0061] Figure 5c A schematic diagram of the interaction between quercetin and the pharmacophore in human ACE2 protein provided in this application at another scale;

[0062] Figure 6a A topological diagram of the first pharmacophore model provided in this application;

[0063] Figure 6b A topological diagram of the second pharmacophore model provided in this application;

[0064] Figure 6c A topological diagram of the interaction between HD and hd provided in this application;

[0065] Figure 6d A topological diagram of the interaction between HA and HA provided in this application;

[0066] Figure 6e A topological diagram of the interaction between CR and cr provided in this application;

[0067] Figure 7 A schematic diagram of a process for determining the score provided for this application;

[0068] Figure 8a A schematic diagram of the basis for the scoring items provided for this application;

[0069] Figure 8b Another schematic diagram of the basis for the scoring items provided for this application;

[0070] Figure 8c Another schematic diagram of the basis for the scoring items provided in this application;

[0071] Figure 9 A schematic structural diagram of the molecular docking device provided in this application;

[0072] Figure 10 A schematic diagram of the structure of the electronic device provided in this application. DETAILED DESCRIPTION

[0073] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field based on this application are within the scope of protection of the present invention.

[0074] To more clearly illustrate the molecular docking method provided in this application, the following briefly describes the relevant concepts involved in this article:

[0075] Quantum Approximate Optimization Algorithm (QAOA) algorithm: An algorithm suitable for quantum computers for solving combinatorial optimization problems, which can handle graph theory models such as the maximum cut problem.

[0076] Molecular Docking: Molecular docking is a method for drug design that uses the characteristics of receptor proteins and the interaction between receptors and drug molecules. It is a theoretical simulation method that studies the interaction between ligand molecules and receptor molecules and predicts their binding mode and affinity. In recent years, molecular docking has become an important technology in the field of computer-assisted drug research.

[0077] A pharmacophore is a combination of characteristic three-dimensional structural elements. It is a model built on pharmacophoric characteristic elements. These elements are primarily categorized into seven types: hydrogen bond donor (HD), hydrogen bond acceptor (HA), positively charged center (PO), negatively charged center (NE), aromatic ring center (AR), hydrophobic group, hydrophilic group, and geometric conformational volume clash. When a drug molecule interacts with a receptor target, it must form an active conformation that is geometrically and energetically compatible with the target. However, the groups within a drug molecule have varying effects on activity, and changes in these groups can significantly influence the interaction between the drug and target.

[0078] The molecular docking method provided in this application is described below. Figure 1 , Figure 1 The figure shows a flow chart of the molecular docking method provided by the present application, which may include:

[0079] S101, obtaining the first spatial structure of the small molecule compound and the second spatial structure of the target protein.

[0080] S102, determining a first pharmacophore existing in the first spatial structure and a second pharmacophore existing in the second spatial structure.

[0081] S103 , establishing a first pharmacophore model for representing each first pharmacophore and the distance between each first pharmacophore, and establishing a second pharmacophore model for representing each second pharmacophore and the distance between each second pharmacophore.

[0082] S104, determining, by the QAOA algorithm, a combination mode in which the scores of the first pharmacophore model and the second pharmacophore model meet a preset screening condition.

[0083] The score is related to the distance between the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are combined.

[0084] By selecting this embodiment, a pharmacophore model can be constructed based on the spatial structure of the small molecule compound and the target protein, using a pharmacophore that can effectively characterize the interaction between the small molecule compound and the target protein. Since the pharmacophore model is used to represent the pharmacophore and the distance between each pharmacophore, the pharmacophore model can effectively characterize the original spatial structure. At the same time, if the pharmacophore is regarded as a node in a graph theory model, the distance between the pharmacophores can be equivalently regarded as the relationship between the nodes, that is, the edge in the graph theory model. Therefore, the pharmacophore model is equivalent to a node-edge graph theory model, and the molecular docking problem is transformed into an optimal combination problem of two graph theory models. Therefore, the QAOA algorithm can be used to determine the optimal combination of the first pharmacophore model and the second pharmacophore model. The combination of the graph theory models can be regarded as the binding mode of the small molecule compound and the target protein. Therefore, the binding mode determined by the present application that meets the pre-screening conditions can be regarded as the most likely binding mode of the small molecule compound and the target protein, that is, the present application effectively achieves molecular docking. Furthermore, since the present application realizes molecular docking through the QAOA algorithm, it can fully utilize the powerful computing power of quantum computers, effectively shorten the time consumption of molecular docking, and thus improve the efficiency of virtual screening.

[0085] The aforementioned S101-S104 will be described in detail below. In S101, the small molecule compound and the target protein may vary depending on the application scenario. For example, in one possible embodiment, the small molecule compound is quercetin and the target protein is human ACE2 protein. In another possible embodiment, the small molecule compound may also be other small molecule compounds such as kaempferol and baicalenol. For convenience of description, the following description is based on the example of quercetin as the small molecule compound and human ACE2 protein as the target protein. The principles are the same for other application scenarios, so they will not be repeated here.

[0086] The second spatial structure of human ACE2 protein can be obtained from the protein database PDB, with ID (identity document) 1R4L. The first spatial structure of quercetin is derived from the pubchem database and optimized using Gaussian16. Molecular docking tools such as AutoDock Tools are used to process the first spatial structure and the second spatial structure, and the semi-flexible docking method of AutoDock Vina is used to simulate the docking of the first spatial structure and the second spatial structure. For example, Figure 2 As shown, Figure 2 The helical structure in the middle is the second spatial structure of the target protein, and the multi-ring connected structure is the first spatial structure of quercetin. Figure 2 In the example shown, the version used by AutoDock Tools is version 1.5.6, and the version used by AutoDock Vina is version 1.2.3. In other possible embodiments, the versions of the molecular docking tools used may be different.

[0087] In S102, AncPhore can be used to determine the pharmacophores of small molecules and target proteins. For example, quercetin can be used to identify the pharmacophores of small molecules and target proteins. Figure 3a 、 Figure 3b , Figure 3a 、 Figure 3b Shown are the pharmacophores present in quercetin. Figure 3a The positions of the pharmacophores are marked with circular shadows. Figure 3b The pharmacophores present in quercetin are marked, including: HD1, HD2, HD3, HA1, HA2, CR (Cation-πInteractions, pharmacophore formed by conjugation of cations and aromatic rings) 1, CR2, and CR3.

[0088] Taking human ACE2 protein as an example, see Figure 4 , Figure 4 The following diagram shows the pharmacophores present in the human ACE2 protein, specifically ha1, ha2, ha3, hd1, hd2, and cr1. HA and ha in this article refer to hydrogen bond receptors. Capitalization is used to distinguish the pharmacophores in small molecule compounds from those in the target protein. The same applies to HD, hd, CR, and cr.

[0089] The interactions between quercetin and the pharmacophores in human ACE2 protein are as follows Figure 5a 、 Figure 5b as well as Figure 5c As shown, Figure 5a 、 Figure 5b as well as Figure 5cSchematic diagram of the interaction between quercetin and various pharmacophores in human ACE2 protein at different scales.

[0090] In S103, the pharmacophore model may be represented in different ways depending on the application scenario, including but not limited to a table, a matrix, a topological diagram, etc. For example, taking quercetin as an example, quercetin includes eight pharmacophores, namely HD1, HD2, HD3, HA1, HA2, CR1, CR2, and CR3. Therefore, in one possible embodiment, the first pharmacophore model may be represented in the form of a table as shown in Table 1:

[0091] Table 1

[0092] HD1 HD2 HD3 HA1 HA2 CR1 CR2 CR3 HD1 HD2 7.8 HD3 11.2 6.3 HA1 4.9 3.7 6.5 HA2 10.3 5.3 2.9 5.7 CR1 2.8 5.0 9.2 2.8 8.2 CR2 5.0 2.8 7.1 1.4 6.2 2.4 CR3 8.6 3.9 2.8 3.8 2.8 6.5 4.3

[0093] Each entry is used to represent the distance between the center of mass of the corresponding two pharmacophores. For example, the value 7.8 in the HD1 column and HD2 row in Table 1 indicates that the distance between the center of mass of the pharmacophore HD1 and the pharmacophore HD2 is The 11.2 in the HD1 column and HD3 row in Table 1 indicates that the center-of-mass distance between the pharmacophore HD1 and the pharmacophore HD3 is (Angstroms), and so on.

[0094] In another possible embodiment, the first pharmacophore model can also be represented in the form of a matrix, where each row in the matrix corresponds to a pharmacophore, and each column corresponds to a pharmacophore, and the value of each element in the matrix is the center of mass distance between the pharmacophore corresponding to the row to which the element belongs and the pharmacophore corresponding to the column to which the element belongs.

[0095] Furthermore, in another possible embodiment, the first pharmacophore model may also be represented in the form of a topological graph. For example, assuming that the first pharmacophore model includes pharmacophores HD1, HA1, CR1, and CR2, and the center-of-mass distance between CR1 and CR2 is The center-of-mass distance between CR1 and HD1 is The centroid distance between CR1 and HA1 is The center-of-mass distance between CR2 and HD1 is The centroid distance between CR2 and HA1 is The distance between the centroids of HD1 and HA1 is Then the first pharmacophore model can be expressed as Figure 6aThe topological diagram shown includes multiple nodes and multiple edges, each node is used to represent a pharmacophore, and each edge is used to represent the center-of-mass distance between the pharmacophores represented by the two nodes connected by the edge. For example, the edge between the nodes representing CR1 and CR2 represents the center-of-mass distance between CR1 and CR2. Specifically, the edge can be enabled to represent the distance by setting the edge weight of the edge to the center-of-mass distance between CR1 and CR2.

[0096] Similar to the first pharmacophore model, the second pharmacophore model can be expressed as a table, matrix, or topological diagram depending on the application scenario. Taking human ACE2 protein as an example, the second pharmacophore model can be expressed as shown in Table 2:

[0097] Table 2

[0098] hd1 hd2 ha1 ha2 ha3 cr1 hd1 hd2 7.4 ha1 4.6 9.7 ha2 7.8 10.3 10.8 ha3 7.0 6.6 6.0 10.4 cr1 9.4 5.0 13.1 8.9 10.5

[0099] The meanings of the items in Table 2 are the same as those in Table 1 and are not repeated here.

[0100] Assume that the second pharmacophore model includes pharmacophores cr1, hd1, and ha1, and the center-of-mass distance between cr1 and ha1 is The centroid distance between ha1 and hd1 is The centroid distance between cr1 and hd1 is Then the second pharmacophore model can be expressed as Figure 6b The topology diagram shown.

[0101] In S104, the preset screening conditions may vary depending on the application scenario. For example, in one possible embodiment, the preset screening condition may be the combination method with the highest score. In another possible embodiment, the preset screening condition may be the combination method with a score higher than a preset score threshold. However, the preset screening condition should satisfy the following requirements: the combination methods with relatively high scores should satisfy the preset screening condition as much as possible, and the combination methods with relatively low scores should not satisfy the preset screening condition as much as possible.

[0102] It is understood that during the molecular docking process, the pharmacophore in the compound molecule will interact with the pharmacophore in the target protein, that is, bind to each other. In one possible embodiment, to save computational effort, only the binding between pharmacophores of the same type is considered.

[0103] Still taking quercetin and human ACE2 protein as an example, the interaction between HD and HD can be as follows: Figure 6c Similarly, the combination between HA and ha can be expressed as follows Figure 6d The mutual combination between CR and cr can be expressed in the form of Figure 6eAs explained above, in this embodiment, only the mutual binding between pharmacophores of the same type is considered, so the Figure 6c 、 Figure 6d as well as Figure 6e It can effectively reflect that the pharmacophore in the compound molecule will interact with the pharmacophore in the target protein, that is, the graph theory model constructed in the aforementioned S103 can effectively characterize the interaction between the compound molecule and the target protein. Therefore, the QAOA algorithm is used to optimize the combination of graph theory models to determine the binding mode between the compound molecule and the target protein, thereby realizing virtual screening based on the determined binding mode.

[0104] It is understandable that the first pharmacophore model and the second pharmacophore model, as graph theory models, are not constrained by the laws of nature when combined. However, in reality, compound molecules are constrained by various natural laws during the process of interacting with target proteins. Therefore, if the combination of the first pharmacophore model and the second pharmacophore model is not restricted, the determined binding mode may not conform to the laws of nature, that is, the determined binding mode is inaccurate.

[0105] Based on this, in a possible embodiment, the aforementioned S104 includes:

[0106] S1041, determining, by the QAOA algorithm, a binding mode in which the scores between the first pharmacophore model and the second pharmacophore model meet a preset screening condition and the binding relationship between the first pharmacophore and the second pharmacophore meet a preset restriction condition.

[0107] By selecting this embodiment, the binding pattern determined by the QAOA algorithm can be made to meet specific constraints by introducing restriction conditions, thereby reducing the possibility that the determined binding pattern does not conform to the laws of nature and effectively improving the accuracy of the determined binding pattern.

[0108] The constraints may include different conditions depending on the application scenario. The constraints may be set by the user based on natural laws or experience. For example, if the user deduces from natural laws that pharmacophore 1 and pharmacophore 2 cannot combine, the preset constraints may include: pharmacophore 1 and pharmacophore 2 are not combined in this combination mode. For another example, if the user finds based on their own experimental experience that the possibility of pharmacophore 3 and pharmacophore 4 combining is very low, the preset constraints may include: if pharmacophore 3 and pharmacophore 4 combine in this combination mode, the score item corresponding to the combination should be sufficiently low, that is, lower than the preset score threshold.

[0109] For example, in a possible embodiment, the restriction condition may include one or more of the following conditions:

[0110] Binding number condition: when the first pharmacophore binds to the second pharmacophore in this binding mode, the number of times the first target pharmacophore binds to the other pharmacophore is no greater than a preset number threshold.

[0111] Binding scoring condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, if the second target pharmacophore is bound to the third target pharmacophore, the score item corresponding to the distance between the second target pharmacophore and the third target pharmacophore is a preset score.

[0112] The restriction condition may include only the combination number condition, only the combination score condition, or both the combination number condition and the combination score condition. The restriction condition may include one combination number condition or multiple combination number conditions, and the restriction condition may include one combination score condition or multiple combination score conditions.

[0113] For example, taking quercetin and human ACE2 protein as examples, the restrictions include:

[0114] 1. The number of times any HD combines with HD is no more than 1.

[0115] 2. The number of combinations of any HD and HD is no more than 2 times

[0116] 3. The number of times any HA binds to HA is no more than 1.

[0117] 4. The number of times any HA binds to HA is no more than 2 times

[0118] 5. When any hd is combined with HD1 or HD3, the score item corresponding to the distance between the hd and HD1 or HD3 is set to -0.2

[0119] 6. When ha2 is combined with any HA, the score item corresponding to the distance between ha2 and the HA is set to 0

[0120] 7. When ha3 is combined with any HA, the score corresponding to the distance between each cr and CR is set to 0, and the score corresponding to the distance between each hd and HD is set to 0.

[0121] Conditions 1-4 are for the number of bindings, and conditions 5-7 are for the score. For condition 1, the first pharmacophore is any HD, and the number threshold is 1. The same applies to conditions 2-4. For condition 5, the second pharmacophore is any HD, the third pharmacophore is HD1 or HD3, and the default score is -0.2. The same applies to conditions 6-7.

[0122] The score in molecular docking is an important basis for the QAOA algorithm to find the optimal combination. Therefore, how to reasonably design a scoring function to accurately determine the score of the binding mode so that the QAOA algorithm can accurately determine the reasonable binding mode has become a technical problem that needs to be solved urgently.

[0123] Based on this, this application provides a scoring method, such as Figure 7 Shown, including:

[0124] S701 , determining the binding distance between the first pharmacophore and the second pharmacophore in each pharmacophore pair when the first pharmacophore model and the second pharmacophore model are combined in the binding manner.

[0125] Among them, the pharmacophore pair is composed of the first pharmacophore and the second pharmacophore of the same type. Taking quercetin and human ACE2 protein as an example, the binding distance between HD and hd is shown in Table 3:

[0126] Table 3

[0127] HD1 HD2 HD3 hd1 6.4 5.9 5.5 hd2 8.6 3.1 6.9

[0128] The binding distance between HA and HA is shown in Table 4:

[0129] Table 4

[0130] HA1 HA2 ha1 6.9 3.0 ha2 6.6 11.1 ha3 6.5 4.4

[0131] The binding distance between HA and HA is shown in Table 5:

[0132] Table 5

[0133] CR1 CR2 CR3 cr1 5.4 5.7 8.7

[0134] It can be understood that the values of the items in Tables 3-5 are only values in one possible example. Depending on the application scenario, the values in the items in Tables 3-5 may be different, and the above examples do not impose any limitations on this.

[0135] S702: Determine the score items specifically corresponding to each combination.

[0136] The basis for scoring items may vary depending on the application scenario. For example, only hydrogen bonding, hydrophobic interaction, and van der Waals interaction may be considered without considering electrostatic interaction. Alternatively, electrostatic interaction may be considered while considering hydrogen bonding, hydrophobic interaction, and van der Waals interaction. For example, Figure 8a-8c As shown, Figure 8a-8c Shown is a schematic diagram of the basis for determining the scoring items provided in this application.

[0137] Taking the above example of S701 as an example, the score items corresponding to the distance between hd and HD are shown in Table 6:

[0138] Table 6

[0139] HD1 HD2 HD3 hd1 0.4677 0.4677 0.4677 hd2 0.5244 0.5244 0.5244

[0140] The score items corresponding to the distance between ha and HA are shown in Table 7:

[0141] Table 7

[0142] HA1 HA2 ha1 0.4365 0.5478 ha2 0.5136 0.6847 ha3 0.4365 0.5478

[0143] The score items corresponding to the distance between cr and CR are shown in Table 8:

[0144] Table 8

[0145] CR1 CR2 CR3 cr1 0.1555 0.1 0.096

[0146] It can be understood that the values of the items in Tables 6 to 8 are only values in one possible example. Depending on the application scenario, the values in the items in Tables 6 to 8 may be different, and the above examples do not impose any limitations on this.

[0147] S703: Determine the score of the combination method based on each score item.

[0148] The score of the binding mode is positively correlated with each score item, that is, the higher the score item, the higher the score of the binding mode. In one possible embodiment, the score of the binding mode can be determined only based on the score items, but it is understandable that there may be certain inaccuracies in the score items in actual application scenarios. For example, some pharmacophore combinations do not (or rarely) appear in actual co-crystals, so when the pharmacophore combination exists in the binding mode, the score of the binding mode should be lower. Based on this, in one possible embodiment, S703 includes:

[0149] S7031: Determine the score of the combination method based on the score item and the determined penalty item.

[0150] The penalty term is a penalty term corresponding to the pharmacophore bound to the defined pharmacophore when combined with the second pharmacophore model based on a preset correspondence. This embodiment allows the score term to be modified based on the penalty term to make the determined score more accurate, thereby achieving more accurate molecular docking.

[0151] Taking the above example of S701 as an example, the penalty term corresponding to the pharmacophore combination consisting of HD and HD can be shown in Table 9:

[0152] Table 9

[0153] HDx*HDx HD1 HD2 HD3 HD1 -∞ HD2 -1.1457 -∞ HD3 -1.1457 -1.1457 -∞

[0154] The penalty term corresponding to the pharmacophore combination consisting of ha and ha can be shown in Table 10:

[0155] Table 10

[0156] hax*hax ha1 ha2 ha3 ha1 -∞ ha2 -1.3496 -∞ ha3 -1.2214 -1.2506 -∞

[0157] The penalty term corresponding to the pharmacophore combination consisting of CR and CR can be shown in Table 11:

[0158] Table 11

[0159] CRx*CRx CR1 CR2 CR3 CR1 -∞ CR2 -0.0058 -∞ CR3 -0.2718 -0.3016 -∞

[0160] Moreover, the penalty term corresponding to the pharmacophore combination formed by CR1, CR2, and CR3 is -0.2972.

[0161] It can be understood that the values of the items in Tables 9-11 are only values in one possible example. Depending on the application scenario, the values in the items in Tables 9-11 may be different, and the above examples do not impose any limitations on this.

[0162] Corresponding to the aforementioned molecular docking method, the present application also provides a molecular docking device, such as Figure 9 As shown, the device includes:

[0163] The spatial structure acquisition module 901 is used to acquire the first spatial structure of the small molecule compound and the second spatial structure of the target protein;

[0164] a pharmacophore determination module 902 for determining a first pharmacophore present in the first spatial structure and a second pharmacophore present in the second spatial structure;

[0165] a model building module 903 for establishing a first pharmacophore model for representing each of the first pharmacophores and the distance between each of the first pharmacophores, and establishing a second pharmacophore model for representing each of the second pharmacophores and the distance between each of the second pharmacophores;

[0166] The quantum approximate optimization module 904 is configured to determine, using a quantum approximate optimization algorithm, a combination mode in which the score between the first pharmacophore model and the second pharmacophore model satisfies a preset screening condition, wherein the score is related to the distance between the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are combined.

[0167] In a possible embodiment, the quantum approximate optimization module determines, by using a quantum approximate optimization algorithm, a combination mode in which the scores of the first pharmacophore model and the second pharmacophore model satisfy a preset screening condition, including:

[0168] A quantum approximate optimization algorithm is used to determine a binding mode in which the score between the first pharmacophore model and the second pharmacophore model meets a preset screening condition and the binding relationship between each of the first pharmacophore and the second pharmacophore meets a preset restriction condition.

[0169] In a possible embodiment, the preset restriction condition includes any one or more of the following conditions:

[0170] Binding frequency condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, the number of times the first target pharmacophore binds to the other pharmacophore is no greater than a preset number threshold;

[0171] Binding scoring condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, if the second target pharmacophore is bound to the third target pharmacophore, the score item corresponding to the distance between the second target pharmacophore and the third target pharmacophore is a preset score.

[0172] In a possible embodiment, the apparatus further includes a scoring module configured to determine a score for each of the combination methods in the following manner:

[0173] determining a binding distance between the first pharmacophore and the second pharmacophore in each pharmacophore pair when the first pharmacophore model and the second pharmacophore model are combined in the binding manner, wherein the pharmacophore pair is composed of the first pharmacophore and the second pharmacophore of the same type;

[0174] Determine the score item corresponding to each of the combined distances;

[0175] The score of the combination method is determined according to each of the score items.

[0176] In a possible embodiment, the scoring module is further configured to determine, based on a preset correspondence between a pharmacophore combination and a penalty term, a penalty term corresponding to the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are bound to each other in the binding manner;

[0177] The scoring module determines the score of the combination method according to each of the scoring items, including:

[0178] The score of the combination is determined according to the score item and the determined penalty item.

[0179] In a possible embodiment, the corresponding relationship is determined in advance in the following manner:

[0180] The penalty term corresponding to each pharmacophore combination is determined based on the binding mode of each pharmacophore in the cocrystal.

[0181] In a possible embodiment, the small molecule compound is a quercetin molecule, and the target protein is a human ACE2 protein.

[0182] The embodiment of the present invention further provides an electronic device, such as Figure 10 As shown, it includes a processor 1001, a communication interface 1002, a memory 1003 and a communication bus 1004, wherein the processor 1001, the communication interface 1002, and the memory 1003 communicate with each other through the communication bus 1004.

[0183] Memory 1003, used for storing computer programs;

[0184] The processor 1001 is configured to execute the program stored in the memory 1003 by performing the following steps:

[0185] Obtain the first spatial structure of the small molecule compound and the second spatial structure of the target protein;

[0186] determining a first pharmacophore present in the first spatial structure and a second pharmacophore present in the second spatial structure;

[0187] establishing a first pharmacophore model for representing each of the first pharmacophores and the distance between each of the first pharmacophores, and establishing a second pharmacophore model for representing each of the second pharmacophores and the distance between each of the second pharmacophores;

[0188] A quantum approximate optimization algorithm is used to determine a combination mode in which the score between the first pharmacophore model and the second pharmacophore model meets a preset screening condition, wherein the score is related to the distance between the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are combined.

[0189] The communication bus mentioned in the electronic device mentioned above may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0190] The communication interface is used for communication between the above electronic device and other devices.

[0191] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0192] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0193] In another embodiment provided by the present invention, a computer-readable storage medium is further provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of any of the above-mentioned molecular docking methods are implemented.

[0194] In another embodiment provided by the present invention, a computer program product comprising instructions is also provided, which, when executed on a computer, enables the computer to execute any one of the molecular docking methods in the above embodiments.

[0195] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0196] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0197] Each embodiment in this specification is described in a related manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the embodiments of the apparatus, electronic device, computer-readable storage medium, and computer program product are generally similar to the method embodiments, so their descriptions are relatively simple. For related portions, reference can be made to the descriptions of the method embodiments.

[0198] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention are included in the scope of protection of the present invention.

Claims

1. A molecular docking method, characterized in that: The method comprises: Obtain the first spatial structure of the small molecule compound and the second spatial structure of the target protein; determining a first pharmacophore present in the first spatial structure and a second pharmacophore present in the second spatial structure; establishing a first pharmacophore model for representing each of the first pharmacophores and the distance between each of the first pharmacophores, and establishing a second pharmacophore model for representing each of the second pharmacophores and the distance between each of the second pharmacophores; A quantum approximate optimization algorithm is used to determine a combination mode in which the score between the first pharmacophore model and the second pharmacophore model meets a preset screening condition, wherein the score is related to the distance between the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are combined.

2. The method according to claim 1, characterized in that The method of determining the combination of the first pharmacophore model and the second pharmacophore model by using a quantum approximate optimization algorithm, wherein the scores of the first pharmacophore model and the second pharmacophore model satisfy a preset screening condition, includes: A quantum approximate optimization algorithm is used to determine a binding mode in which the score between the first pharmacophore model and the second pharmacophore model meets a preset screening condition and the binding relationship between each of the first pharmacophore and the second pharmacophore meets a preset restriction condition.

3. The method according to claim 2, characterized in that The preset restriction conditions include any one or more of the following conditions: Binding frequency condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, the number of times the first target pharmacophore binds to the other pharmacophore is no greater than a preset number threshold; Binding scoring condition: when the first pharmacophore and the second pharmacophore are bound in the binding manner, if the second target pharmacophore is bound to the third target pharmacophore, the score item corresponding to the distance between the second target pharmacophore and the third target pharmacophore is a preset score.

4. The method according to claim 1, wherein The score of each combination is determined by the following method: determining a binding distance between the first pharmacophore and the second pharmacophore in each pharmacophore pair when the first pharmacophore model and the second pharmacophore model are combined in the binding manner, wherein the pharmacophore pair is composed of the first pharmacophore and the second pharmacophore of the same type; Determine the score item corresponding to each of the combined distances; The score of the combination method is determined according to each of the score items.

5. The method according to claim 4, characterized in that The method further comprises: Determining, based on a preset correspondence between a pharmacophore combination and a penalty term, a penalty term when the first pharmacophore model and the second pharmacophore model are combined in the combination manner; Determining the score of the combination according to each of the score items includes: The score of the combination is determined according to the score item and the determined penalty item.

6. The method according to claim 5, characterized in that The corresponding relationship is determined in advance by the following method: The penalty term corresponding to each pharmacophore combination is determined based on the binding mode of each pharmacophore in the cocrystal.

7. The method according to claim 1, characterized in that The small molecule compound is a quercetin molecule, and the target protein is a human ACE2 protein.

8. A molecular docking device, characterized in that: The device comprises: A spatial structure acquisition module is used to obtain the first spatial structure of the small molecule compound and the second spatial structure of the target protein; a pharmacophore determination module, configured to determine a first pharmacophore present in the first spatial structure and a second pharmacophore present in the second spatial structure; a model building module, configured to establish a first pharmacophore model for representing each of the first pharmacophores and the distance between each of the first pharmacophores, and to establish a second pharmacophore model for representing each of the second pharmacophores and the distance between each of the second pharmacophores; A quantum approximate optimization module is used to determine, through a quantum approximate optimization algorithm, a combination mode between the first pharmacophore model and the second pharmacophore model in which the score satisfies a preset screening condition, wherein the score is related to the distance between the pharmacophores bound to each other when the first pharmacophore model and the second pharmacophore model are combined.

9. The device according to claim 8, characterized in that The quantum approximate optimization module determines, by using a quantum approximate optimization algorithm, a combination method in which the scores of the first pharmacophore model and the second pharmacophore model meet a preset screening condition, including: A quantum approximate optimization algorithm is used to determine a binding mode in which the score between the first pharmacophore model and the second pharmacophore model meets a preset screening condition and the binding relationship between each of the first pharmacophore and the second pharmacophore meets a preset restriction condition.

10. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory for storing computer programs; A processor, configured to implement the method steps described in any one of claims 1 to 7 when executing a program stored in a memory.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method steps according to any one of claims 1 to 7 are implemented.

Images