Virtual screening method, apparatus and electronic device

Abstract

The application relates to a virtual screening method, device and electronic equipment. The method comprises the following steps: performing molecular docking on ligand molecules in a molecule library and target proteins to obtain at least one molecular conformation after the ligand molecules contact the target proteins; determining a first binding free energy of the molecular conformation after the ligand molecules contact the target proteins by using a preset scoring function, and screening a molecular conformation whose first binding free energy meets a first preset condition; and determining a second binding free energy of the molecular conformation whose first binding free energy meets the first preset condition after the ligand molecules contact the target proteins based on a semi-empirical molecular orbital method, and screening a molecular conformation whose second binding free energy meets a second preset condition. The scheme provided by the application can quickly predict the binding free energy of the molecular conformation, improve the accuracy of the prediction result, and thus improve the success rate of screening active compounds.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of virtual screening, and particularly relates to a virtual screening method, device and electronic equipment. BACKGROUND

[0002] New drug design and development is a creative and exploratory research work. Drug molecule design is to transform an active compound into a safe, effective and controllable compound in the human body through step-by-step optimization with a rational strategy and scientific planning, to meet the requirements of multi-dimensional properties of drugs in the process of structure transformation and modification, and to construct a new molecular entity with expected pharmacological activity. Among them, virtual screening is an important means in the process of lead compound discovery. Virtual screening (VS) is also called computer screening, which is based on a small molecule database to screen active compounds. By using the molecular docking operation between small molecule compounds and drug targets, virtual screening can quickly select active compounds with drug properties from tens to millions of molecules, thereby greatly reducing the number of experimental screening compounds, shortening the research period, and reducing the cost of drug research and development.

[0003] In related technologies, the activity of a drug molecule as a ligand is generally manifested through the interaction between the receptor macromolecule, so the prediction of the binding free energy between the receptor and the ligand is the core problem of structure-based computer-aided drug molecule design. Among them, when virtual screening is based on a receptor, each molecule of a chemical molecule library is docked into the binding pocket of the target protein three-dimensional structure (i.e., the receptor) of the receptor in turn, and a scoring function is used to evaluate the binding "quality" of each docked protein-ligand complex. The scoring function can intuitively replace the expected strength (i.e., the binding free energy) of a given protein-ligand complex, and the scoring function is used to screen the most promising lead compounds for experimental verification. However, in virtual screening, about 100 million molecules generate about 100 million conformations through molecular docking. Although the scoring function has the advantages of fast calculation speed and high-throughput calculation, it has the problem of low prediction accuracy, which is not conducive to screening potential active compounds. SUMMARY

[0004] To solve or partially solve the problems in the related art, the present application provides a virtual screening method, device and electronic equipment, which can quickly predict the binding free energy of the molecular conformation after contacting the target protein, improve the accuracy of the prediction results, and thus improve the success rate of screening active compounds.

[0005] The first aspect of the present application provides a virtual screening method, comprising:

[0006] molecular docking is performed on ligand molecules in a molecular library and a target protein to obtain at least one molecular conformation of the ligand molecules after contacting the target protein;

[0007] A preset scoring function is used to determine a first binding free energy of the molecular conformation after contacting the target protein, and a molecular conformation satisfying a first preset condition of the first binding free energy is screened and obtained.

[0008] Based on a semi-empirical molecular orbital method, a second binding free energy of the molecular conformation satisfying the first preset condition of the first binding free energy after contacting the target protein is determined, and the molecular conformation satisfying a second preset condition of the second binding free energy is screened and obtained.

[0009] In an embodiment, the method comprises the following steps:

[0010] A preset scoring function is used to determine a first binding free energy of the molecular conformation after contacting the target protein, and a molecular conformation satisfying a first preset condition of the first binding free energy is screened and obtained.

[0011] The molecular conformations of all the ligand molecules are sorted according to the numerical values of the corresponding first binding free energies, and a molecular conformation whose sorting of the first binding free energy is within a first preset range is obtained.

[0012] In an embodiment, the method comprises the following steps:

[0013] The residues within a preset range around the cavity of the target protein are used as an empty protein, and a corresponding first single-point energy is determined.

[0014] The empty protein and the molecular conformation form a complex, and a corresponding second single-point energy is determined.

[0015] A third single-point energy of the molecular conformation is determined.

[0016] According to the difference between the second single-point energy and the first single-point energy and the third single-point energy, a second binding free energy of the molecular conformation after contacting the target protein is determined.

[0017] In an embodiment, the method comprises the following steps:

[0018] calculating, by using the MM-GBSA method or the MM-PBSA method, a third binding free energy corresponding to the molecular conformation of the second binding free energy meeting the second preset condition after the molecular conformation contacts the target protein; screening to obtain the molecular conformation of the third binding free energy meeting a third preset condition; or

[0019] calculating, by using the FEP method or the TI method, a fourth binding free energy corresponding to the molecular conformation of the second binding free energy meeting the second preset condition after the molecular conformation contacts the target protein; and screening to obtain the molecular conformation of the fourth binding free energy meeting a fourth preset condition.

[0020] In an embodiment, after the calculating, by using the MM-GBSA method or the MM-PBSA method, the third binding free energy corresponding to the molecular conformation of the second binding free energy meeting the second preset condition after the molecular conformation contacts the target protein, the method further comprises:

[0021] calculating, by using the FEP method or the TI method, a fourth binding free energy corresponding to the molecular conformation of the third binding free energy meeting the third preset condition after the molecular conformation contacts the target protein; and screening to obtain the molecular conformation of the fourth binding free energy meeting a fourth preset condition; or

[0022] after the calculating, by using the FEP method or the TI method, the fourth binding free energy corresponding to the molecular conformation of the second binding free energy meeting the second preset condition after the molecular conformation contacts the target protein, and screening to obtain the molecular conformation of the fourth binding free energy meeting a fourth preset condition, the method further comprises:

[0023] calculating, by using the MM-GBSA method or the MM-PBSA method, a third binding free energy corresponding to the molecular conformation of the fourth binding free energy meeting the fourth preset condition after the molecular conformation contacts the target protein; and screening to obtain the molecular conformation of the third binding free energy meeting a third preset condition.

[0024] In an embodiment, the method further comprises:

[0025] composing the molecular conformation of the second binding free energy meeting the second preset condition into a candidate conformation set; or composing the molecular conformation of the third binding free energy meeting the third preset condition into a candidate conformation set; or composing the molecular conformation of the fourth binding free energy meeting the fourth preset condition into a candidate conformation set;

[0026] selecting a potential active compound from the candidate conformation set.

[0027] In an embodiment, the screening to obtain the molecular conformation of the second binding free energy meeting the second preset condition comprises:

[0028] ordering each molecular conformation of the same ligand molecule according to the numerical value of the corresponding second binding free energy, and obtaining the molecular conformation whose ordering of the second binding free energy is within a second preset range;

[0029] and / or

[0030] The screening obtains the molecular conformation whose third binding free energy meets a third preset condition, including:

[0031] ordering each molecular conformation of the same ligand molecule according to the numerical value of the corresponding third binding free energy, and obtaining the molecular conformation whose ordering of the third binding free energy is within a third preset range;

[0032] and / or

[0033] The screening obtains the molecular conformation whose fourth binding free energy meets a fourth preset condition, including:

[0034] ordering each molecular conformation of the same ligand molecule according to the numerical value of the corresponding fourth binding free energy, and obtaining the molecular conformation whose ordering of the fourth binding free energy is within a fourth preset range

[0035] The second aspect of the present application provides a virtual screening method, including:

[0036] molecular docking of ligand molecules in a molecular library and target proteins, obtaining at least one molecular conformation of the ligand molecules after contacting the target proteins;

[0037] determining the first binding free energy of the molecular conformation after contacting the target protein by using a preset scoring function, and screening to obtain the molecular conformation whose first binding free energy meets a first preset condition;

[0038] calculating the third binding free energy corresponding to the molecular conformation whose first binding free energy meets the first preset condition after contacting the target protein by using the MM-GBSA method or the MM-PBSA method; and screening to obtain the molecular conformation whose third binding free energy meets a third preset condition;

[0039] determining the second binding free energy of the molecular conformation whose third binding free energy meets the third preset condition after contacting the target protein based on a semi-empirical molecular orbital method, and screening to obtain the molecular conformation whose second binding free energy meets a second preset condition.

[0040] In an embodiment, the determination of the first binding free energy of the molecular conformation after contacting the target protein by using a preset scoring function, and the screening to obtain the molecular conformation whose first binding free energy meets a first preset condition, include:

[0041] determining a first binding free energy corresponding to each of the molecular conformations of the ligand molecule after contacting with the target protein by using a preset scoring function;

[0042] ranking all the molecular conformations of the ligand molecule according to the numerical values of the first binding free energies, and obtaining the molecular conformations occupying a preset proportion in the ranking.

[0043] In an embodiment, the determining of the second binding free energy of the molecular conformation of the ligand molecule after contacting with the target protein based on the semi-empirical molecular orbital method comprises:

[0044] determining a first single-point energy of residues within a preset range around the cavity of the target protein as a dummy protein;

[0045] determining a second single-point energy of the dummy protein and the molecular conformation as a complex;

[0046] determining a third single-point energy of the molecular conformation;

[0047] determining the second binding free energy of the molecular conformation according to the difference between the second single-point energy and the first single-point energy and the third single-point energy.

[0048] In an embodiment, after the screening obtains the molecular conformation of the ligand molecule with the second binding free energy meeting the second preset condition, the method further comprises:

[0049] calculating a fourth binding free energy corresponding to the molecular conformation of the ligand molecule with the second binding free energy meeting the second preset condition after contacting with the target protein by using a FEP method or a TI method; and screening to obtain the molecular conformation of the ligand molecule with the fourth binding free energy meeting a fourth preset condition.

[0050] In an embodiment, the method further comprises:

[0051] forming a candidate conformation set by using the molecular conformation of the ligand molecule with the second binding free energy meeting the second preset condition; or forming a candidate conformation set by using the molecular conformation of the ligand molecule with the fourth binding free energy meeting the fourth preset condition;

[0052] selecting a potential active compound from the candidate conformation set.

[0053] In an embodiment, the screening to obtain the molecular conformation of the ligand molecule with the second binding free energy meeting the second preset condition comprises:

[0054] ranking all the molecular conformations of the ligand molecule according to the numerical values of the second binding free energies, and obtaining the molecular conformation with the second binding free energy within a second preset range in the ranking;

[0055] and / or

[0056] The screening obtains the molecular conformation of the third binding free energy meeting the third preset condition, and includes:

[0057] Each molecular conformation of the same ligand molecule is sorted according to the numerical value of the corresponding third binding free energy, and the molecular conformation of the third binding free energy whose sorting is within the third preset range is obtained.

[0058] And / or

[0059] The screening obtains the molecular conformation of the fourth binding free energy meeting the fourth preset condition, and includes:

[0060] Each molecular conformation of the same ligand molecule is sorted according to the numerical value of the corresponding fourth binding free energy, and the molecular conformation of the fourth binding free energy whose sorting is within the fourth preset range is obtained.

[0061] The third aspect of the present application provides a virtual screening device, including:

[0062] The molecular docking module is used for molecular docking of ligand molecules in a molecular library and target proteins, and at least one molecular conformation of the ligand molecules after contacting the target proteins is obtained.

[0063] The first screening module is used for determining the first binding free energy of the molecular conformation after contacting the target protein by using a preset scoring function, and screening obtains the molecular conformation of the first binding free energy meeting the first preset condition.

[0064] The second screening module is used for determining the second binding free energy of the molecular conformation meeting the first preset condition after contacting the target protein based on a semi-empirical molecular orbital method, and screening obtains the molecular conformation of the second binding free energy meeting the second preset condition.

[0065] In an embodiment, the device further includes a third screening module and / or a fourth screening module; wherein:

[0066] The third screening module is configured to calculate a third binding free energy corresponding to the molecular conformation that meets the second preset condition in the second screening module, by using a MM-GBSA method or a MM-PBSA method, or calculate a third binding free energy corresponding to the molecular conformation that meets the fourth preset condition in the fourth screening module, by using a MM-GBSA method or a MM-PBSA method, wherein the fourth preset condition is that the fourth binding free energy is calculated by using a FEP method or a TI method, and the fourth preset condition is that the molecular conformation that meets the third preset condition is screened in the third binding free energy calculated by using a MM-GBSA method or a MM-PBSA method in the third screening module, or the fourth binding free energy corresponding to the molecular conformation that meets the second preset condition in the second screening module is calculated by using a FEP method or a TI method; and the molecular conformation that meets the fourth preset condition in the fourth binding free energy is screened.

[0067] The fourth screening module is configured to calculate a fourth binding free energy corresponding to the molecular conformation that meets the third preset condition in the third screening module, by using a FEP method or a TI method, wherein the third preset condition is that the fourth binding free energy is calculated by using a MM-GBSA method or a MM-PBSA method in the second screening module, or the fourth binding free energy corresponding to the molecular conformation that meets the second preset condition in the second screening module is calculated by using a FEP method or a TI method; and the molecular conformation that meets the fourth preset condition in the fourth binding free energy is screened.

[0068] In an embodiment, the second screening module is configured to determine a first single-point energy by taking residues within a preset range around a cavity of the target protein as an empty protein, and determine a second single-point energy by taking the empty protein and the molecular conformation as a complex, determine a third single-point energy of the molecular conformation, and determine the second binding free energy of the molecular conformation after contacting with the target protein according to a difference between the second single-point energy and the first single-point energy and the third single-point energy.

[0069] The fourth aspect of the present application provides a virtual screening device, comprising:

[0070] A molecular docking module is configured to perform molecular docking between a ligand molecule in a molecular library and a target protein, and obtain at least one molecular conformation of the ligand molecule after contacting with the target protein.

[0071] A first screening module is configured to determine a first binding free energy of the molecular conformation after contacting with the target protein by using a preset scoring function, and screen a molecular conformation that meets a first preset condition in the first binding free energy.

[0072] The third screening module is configured to calculate, by using an MM-GBSA method or an MM-PBSA method, a third binding free energy corresponding to the molecular conformation whose first binding free energy meets the first preset condition after the molecular conformation contacts a target protein; and to screen the molecular conformation whose third binding free energy meets a third preset condition.

[0073] The second screening module is configured to determine, by using a semi-empirical molecular orbital method, a second binding free energy of the molecular conformation whose third binding free energy meets the third preset condition after the molecular conformation contacts the target protein, and to screen the molecular conformation whose second binding free energy meets a second preset condition.

[0074] In an embodiment, the device further includes a fourth screening module configured to calculate, by using a FEP method or a TI method, a fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after the molecular conformation contacts the target protein; and to screen the molecular conformation whose fourth binding free energy meets a fourth preset condition.

[0075] The fifth aspect of the present application provides an electronic device, comprising:

[0076] a processor; and

[0077] a memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method described above.

[0078] The sixth aspect of the present application provides a computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the method described above.

[0079] The technical solution provided by the present application can include the following beneficial effects:

[0080] The technical solution of the present application, by obtaining a molecular conformation through molecular docking, determining a first binding free energy of the molecular conformation by using a preset scoring function, and performing a first screening to filter out part of the molecular conformations, and then calculating a second binding free energy of the remaining molecular conformations based on a semi-empirical molecular orbital method and performing a second screening to further filter out part of the molecular conformations, ensures the rapid prediction of the binding free energy of the molecular conformation, and makes the remaining molecular conformations have strong binding ability, so as to select active compounds from more advantageous ligand molecules, improve the accuracy of the prediction result, and then improve the success rate of screening the active compounds.

[0081] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0082] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which like reference characters refer to like parts throughout the figures, and wherein:

[0083] Figure 1 is a flowchart of a virtual screening method according to an embodiment of the present application;

[0084] Figure 2 is a flowchart of another virtual screening method according to an embodiment of the present application;

[0085] Figure 3 is a schematic diagram of a molecular simulation for calculating the binding free energy according to an embodiment of the present application;

[0086] Figure 4 is a flowchart of another virtual screening method according to an embodiment of the present application;

[0087] Figure 5 is a flowchart of another virtual screening method according to an embodiment of the present application;

[0088] Figure 6 is a flowchart of another virtual screening method according to an embodiment of the present application;

[0089] Figure 7 is a flowchart of another virtual screening method according to an embodiment of the present application;

[0090] Figure 8 is a flowchart of another virtual screening method according to an embodiment of the present application;

[0091] Figure 9 is a structural diagram of a virtual screening apparatus according to an embodiment of the present application;

[0092] Figure 10 is a structural diagram of another virtual screening apparatus according to an embodiment of the present application;

[0093] Figure 11 is a structural diagram of another virtual screening apparatus according to an embodiment of the present application;

[0094] Figure 12 is a structural diagram of another virtual screening apparatus according to an embodiment of the present application;

[0095] Figure 13 is a structural diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0096] Embodiments of the present application will be described in more detail with reference to the drawings. Although embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the application to those skilled in the art.

[0097] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this application and the appended claims, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "and / or," as used herein, refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0098] It should be understood that although the terms "first", "second", "third", etc. can be used in this application to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, the first information can also be referred to as the second information without departing from the scope of the application, and similarly, the second information can also be referred to as the first information. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0099] In the related art, when a scoring function is used to predict the binding free energy of a candidate drug molecule and a target protein in virtual screening, the prediction result has the problem of low accuracy, which is not conducive to screening the desired active compound.

[0100] To solve the above problems, the embodiments of the present application provide a virtual screening method, device and electronic equipment, which can quickly predict the binding free energy of molecular conformation and improve the accuracy of the prediction result, thereby improving the success rate of screening the active compound.

[0101] The technical solutions of the embodiments of the present application will be described in detail below with reference to the drawings.

[0102] Embodiment one

[0103] Figure 1 is a flowchart of the virtual screening method provided by the embodiments of the present application.

[0104] Referring to Figure 1 The virtual screening method provided by an embodiment of the present application comprises:

[0105] S110, performing molecular docking of the ligand molecules in the molecule library with the target protein to obtain at least one molecular conformation of the ligand molecules after contacting with the target protein.

[0106] The molecule library can be a small molecule database, such as known databases such as ZINC, Specs, ChemBridge, etc., without limitation. At least one molecule in the molecule library is used as a ligand molecule to perform molecular docking with the target protein. It can be understood that the target protein is designed and selected according to the specific need of the drug efficacy function. Further, each ligand molecule is docked into the protein pocket of the target protein, so that each ligand molecule generates one or more molecular conformations in contact with the target protein.

[0107] S120, determining the first binding free energy of the molecular conformation after contacting with the target protein by using a preset scoring function, and screening to obtain the molecular conformation whose first binding free energy meets a first preset condition.

[0108] The known scoring function can be used as the preset scoring function, or the self-designed scoring function can be used as the preset scoring function. Each molecular conformation is evaluated by the preset scoring function, i.e. the binding free energy of the molecular conformation after docking of the ligand molecule with the target protein is predicted, and the first binding free energy of each molecular conformation with the target protein is calculated.

[0109] Further, in an embodiment, the molecular conformations of all ligand molecules are sorted according to the numerical value of the first binding free energy, and the molecular conformation whose sorting is within a first preset range is obtained. That is, part or all of the molecular conformations corresponding to the part of the ligand molecules whose sorting is not within the first preset range are screened out. In an embodiment, the molecular conformations are sorted in order from small to large according to the numerical value of the first binding free energy, and a preset proportion of the molecular conformations in the top ranking are obtained. That is, the molecular conformations in the first preset range are the first preset proportion of the molecular conformations in the top ranking after sorting according to the first binding free energy. The first preset proportion can be 0.1% to 100%. It can be understood that if the sorting is in order from large to small, the molecular conformations in the last ranking with a proportion of 0.1% to 100% are obtained. By screening the molecular conformations whose first binding free energy meets the first preset condition, the ligand molecules and their corresponding molecular conformations that do not meet the condition are filtered out, which simplifies the calculation amount of the subsequent steps and reduces the interference of redundant molecular conformations on the selection of active compounds. It should be understood that the numerical value of the calculated binding free energy is generally negative, and the smaller the numerical value of the binding free energy, i.e. the larger the absolute value of the numerical value, the stronger the binding ability of the ligand molecule with the target protein, otherwise the opposite. In this embodiment, the molecular conformations with smaller numerical values and in the top ranking of the binding ability are screened out, so that the molecular conformations with stronger binding ability are screened out.

[0110] S130, determining, based on a semi-empirical molecular orbital method, second binding free energies of the molecular conformations of the ligand molecules after contacting the target protein, and screening to obtain the molecular conformations of the ligand molecules with the second binding free energies meeting a second preset condition.

[0111] After a part of the molecular conformations are filtered out in the above steps, the second binding free energies of the remaining molecular conformations of each ligand molecule are calculated based on the molecular orbital theory. Specifically, the semi-empirical molecular orbital method used in the present application is a quantum mechanical method for solving the Schrödinger equation. The molecular orbitals and the corresponding energy levels, i.e., energy values, are solved by the semi-empirical molecular orbital method. It can be understood that the calculation methods based on the second binding free energy and the first binding free energy are different, and the calculated values of the binding free energies for the same molecular conformation of the same ligand molecule can be the same or different.

[0112] After obtaining the values of the second binding free energies of the molecular conformations, the molecular conformations with the second binding free energies meeting the second preset condition are screened. In an embodiment, the molecular conformations of the same ligand molecule are sorted according to the values of the corresponding second binding free energies, and the molecular conformations with the sorted second binding free energies within the second preset range are obtained. In an embodiment, the molecular conformations of the same ligand molecule are sorted according to the values of the second binding free energies from small to large, and the molecular conformations with the second preset proportion in the top ranking are obtained. That is, the molecular conformations within the second preset range can be the molecular conformations with the second preset proportion in the top ranking after sorting the second binding free energies from small to large. The second preset proportion can be 0.1% to 100%. It can be understood that if the sorting is from large to small, the molecular conformations with the proportion of 0.1% to 100% in the last ranking are obtained. Further, in an embodiment, the molecular conformation with the smallest second binding free energy value in the same ligand molecule is obtained. That is, only the molecular conformation with the smallest second binding free energy value in each ligand molecule is screened, and other molecular conformations of the same ligand molecule are filtered out. It can be understood that the smaller the value of the binding free energy, the stronger the binding ability of the ligand molecule to the target protein, otherwise the opposite.

[0113] As can be seen from the example, the virtual screening method of the application obtains the molecular conformation through molecular docking, determines the first binding free energy of the molecular conformation by using a preset scoring function, and performs a first screening to filter out part of the molecular conformations; then, the second binding free energy of the remaining molecular conformations is calculated based on a semi-empirical molecular orbital method, and a second screening is performed to further filter out part of the molecular conformations. Such a design ensures the rapid prediction of the binding free energy of the molecular conformation, while making the remaining molecular conformations have strong binding capacity, so as to facilitate the selection of active compounds from more advantageous ligand molecules, improve the accuracy of the prediction result, and then improve the success rate of screening to obtain active compounds.

[0114] Embodiment Two

[0115] Figure 2 is another flowchart of the virtual screening method shown in the embodiments of the application.

[0116] Referring to Figure 2 The virtual screening method of the application comprises the following steps:

[0117] S210, molecular docking is performed on a plurality of ligand molecules in a molecular library and a target protein to obtain at least one molecular conformation after the ligand molecules contact the target protein.

[0118] In this step, the molecular docking can be performed by using the known molecular docking software such as MOE, Glide, LeDock, AutoDock, etc. After each ligand molecule is docked with the target protein, one or more molecular conformations can be generated, and the molecular conformations generated after the molecular docking of the plurality of ligand molecules can be collected in a conformation library.

[0119] S220, a preset scoring function is used to determine the first binding free energy corresponding to each molecular conformation after the molecular conformation contacts the target protein; and a molecular conformation whose first binding free energy meets a first preset condition is screened and obtained.

[0120] In this embodiment, the preset scoring function can use the following formula (1), and in other embodiments, other scoring functions can also be used, which are not limited herein.

[0121]

[0122] In the formula, ΔGbind represents the first binding free energy of a certain molecular conformation, E vdw represents the van der Waals interaction, E hb is the hydrogen bond interaction, Θ is the Heaviside step function, E co is the cutoff value of flexible docking. q i is the charge of the i-th atom on the target protein, q jr is the distance between the i-th and j-th atoms of the ligand molecule. ij r is the distance between the i-th and j-th atoms of the ligand molecule. strain is the strain energy of the molecular conformation of the ligand molecule, and lig refers to a certain molecular conformation of a certain ligand molecule, pro refers to the empty protein, and a, b, and g are constants.

[0123] Further, in an embodiment, the molecular conformations of all ligand molecules are sorted according to the numerical values of the corresponding first binding free energies, and the molecular conformations occupying a preset proportion in the ranking are obtained. According to the property that the smaller the numerical value of the binding free energy is, the stronger the binding ability of the molecular conformation is, the numerical values of the first binding free energies are arranged in order from small to large, the smaller the numerical value is, the higher the ranking is, and the molecular conformations in the preset proportion in the ranking are obtained. For example, the molecular conformations in the preset proportion of 0.1% to 100% in the ranking are obtained, so as to screen the molecular conformations with strong binding ability and filter out the molecular conformations that do not meet the conditions. For the convenience of understanding, for example, when the molecular conformations of all ligand molecules are ten thousand, the molecular conformations are sorted according to the numerical values of the first binding free energies from small to large, and each molecular conformation is ranked according to the sequence number of 1 to 10,000 according to the corresponding first binding free energy. If the preset proportion is 0.1%, the molecular conformations ranked from 1 to 10 are obtained to enter the next step.

[0124] In S230, the second binding free energy of the molecular conformation whose first binding free energy meets the first preset condition after contacting the target protein is determined based on the semi-empirical molecular orbital method, and the molecular conformation whose second binding free energy meets the second preset condition is screened and obtained.

[0125] In this step, the semi-empirical molecular orbital method can use calculation models such as AM1, PM3, PM6, or PM7, and of course other calculation models can also be used, which are not limited herein.

[0126] Further, in an embodiment, the residues within a preset range around the cavity of the target protein are taken as the empty protein, and the corresponding first single-point energy E apo is determined. The empty protein and the molecular conformation form a complex, and the corresponding second single-point energy E comp is determined. The third single-point energy E lig of the molecular conformation is determined. The second binding free energy ΔE bind of the molecular conformation after contacting the target protein is determined according to the difference between the second single-point energy and the first single-point energy and the third single-point energy. That is, the second binding free energy ΔE bind = E apo -E comp -E ligThe second binding free energy can be calculated by using relevant software, such as Gaussian software or MOPAC software. In comparison with calculating the binding free energy of the entire target protein and the specific conformation of the ligand molecule, the design of the present embodiment can reduce the demand for computing resources and improve processing efficiency by using the binding free energy of the residues within the preset range around the cavity of the target protein and the specific conformation of the ligand molecule.

[0127] Specifically, the residues within 3-5 angstroms around the cavity of the target protein can be used as the empty protein. The empty protein is a free protein that does not bind any ligand molecule. It can be understood that the residues around the same cavity of the same target protein, i.e., the same binding site, are used as the empty protein for each molecular conformation. Further, the first single-point energy E apo Then, the empty protein and each molecular conformation are combined to form a corresponding complex, and the second single-point energy E comp Then, the third single-point energy E lig Finally, the second binding free energy of each molecular conformation is the difference between the corresponding second single-point energy, the first single-point energy, and the third single-point energy.

[0128] After obtaining the second binding free energy of each molecular conformation, for the same ligand molecule, the molecular conformation with the smallest second binding free energy value in the same ligand molecule is screened and obtained. That is, each ligand molecule retains only one molecular conformation, i.e., the molecular conformation with the smallest second binding free energy value. Such a design can filter out most of the molecular conformations, thereby improving the accuracy and success rate of obtaining active compounds.

[0129] S240, calculating the third binding free energy of the molecular conformation with the second binding free energy meeting the second preset condition by using the MM-GBSA method or the MM-PBSA method; and screening and obtaining the molecular conformation with the third binding free energy meeting the third preset condition.

[0130] To further narrow down the range of active compounds selected, the calculation method in this step is used to calculate the binding free energy of each molecular conformation after contacting with the target protein based on the molecular conformation obtained in the previous step S230, and further screen the molecular conformation. Among them, MM / PBSA (molecular mechanics with Poisson-Boltzmann surface area solvation method), MM / GBSA (molecular mechanics with generalized Born surface-area solvation method) is a free energy calculation method based on molecular force field. After calculating the third binding free energy corresponding to the molecular conformation by using the MM-GBSA method or the MM-PBSA method, in an embodiment, each molecular conformation of the same ligand molecule is sorted according to the numerical value of the corresponding third binding free energy, and the molecular conformation whose third binding free energy is within the third preset range is obtained. In an embodiment, each molecular conformation of the same ligand molecule is sorted according to the numerical value of the third binding free energy from small to large, and the third preset proportion of the molecular conformation in the front rank is obtained. That is, the molecular conformation within the third preset range can be the third preset proportion of the molecular conformation in the front rank sorted according to the third binding free energy from small to large. The third preset proportion can be 0.1% to 100%. It can be understood that if the sorting is according to the numerical value from large to small, the molecular conformation with a proportion of 0.1% to 100% in the rear rank is obtained. Further, the molecular conformation with the smallest third binding free energy value in the same ligand molecule is obtained, that is, only one molecular conformation is retained for each ligand molecule, so as to obtain the molecular conformation with the strongest binding ability in this calculation result, and filter out the molecular conformation that does not meet the third preset condition. It can be understood that based on different calculation methods, the third binding free energy calculated for the same molecular conformation and the second binding free energy and the first binding free energy calculated in the above steps may be the same or different.

[0131] The calculation principle and method of calculating the third binding free energy corresponding to the molecular conformation contacting with the target protein by using the MM-GBSA method or the MM-PBSA method in this step will be introduced in detail below.

[0132] In the reaction solution, the receptor target protein receptor and the ligand molecule ligand are subjected to molecular docking to form a corresponding complex complex:

[0133]

[0134] The binding free energy ΔG can be directly calculated using the following formula (2). bind This allows for the calculation of the energy ΔG of the receptor target protein in solution. receptor The energy ΔG of the ligand molecule in solution ligand And the energy ΔG of their complexes in solution. complex Then calculate the difference to obtain the binding free energy ΔG. bind .

[0135]

[0136] However, in actual calculations, the fluctuation range of the total energy in the solution is much greater than that of the binding free energy. Therefore, the following formula (3) is constructed through thermodynamic cycles:

[0137]

[0138] Formula (3) above decomposes the total binding free energy in the solvent into two parts: molecular mechanics (binding free energy in vacuum) and solvation energy, which are calculated separately.

[0139] In formula (3) above, the molecular mechanics term is still calculated according to formula (2), that is:

[0140]

[0141] According to the definition of binding free energy, for each receptor target protein molecule in the above formula, ΔG vacuum Calculate according to the following formula (4):

[0142]

[0143] in, It includes bond, bond angle, and dihedral angle energies. For non-bonded van der Waals energy, For nonbonded electrostatic energy; TΔS 0 The contribution to entropy can be obtained using normal mode analysis, but its contribution to structure is very small, and can be ignored in this application.

[0144] The solvation energy ΔG in the above formula (3) solv It can be divided into two parts, namely the polar solvation energy ΔG solv,polar Nonpolar solvation energy ΔG solv,nonpolar As shown in formula (5) below.

[0145]

[0146] There are two main methods for calculating the polar solvation energy: Generalized Born (GB) and Poisson-Boltzmann (PB) model equations.

[0147] When the GB model is used to calculate the polar solvation energy, it can be calculated according to the following formula (6).

[0148]

[0149] where q i is the charge of the i-th atom on the acceptor target protein, q j is the charge of the j-th atom on the ligand molecule; R i is the atomic radius of the i-th atom on the acceptor target protein, R j is the atomic radius of the j-th atom on the acceptor ligand molecule, r ij is the distance between the i-th and j-th atoms, ε is the dielectric constant of the solvent, and κ is the Debye-Huckel limiting parameter. Where f GB is calculated using the following formula (7).

[0150]

[0151] When the PB model is used to calculate the polar solvation energy, it can be calculated according to the following formula (8). It can be understood that in the PB model, the solute is represented by the atomic detail model in the molecular mechanics force field, while the solvent molecules and any dissolved electrolytes are treated as a structureless continuum. The continuum treatment represents the solute as a dielectric body, the shape of which is defined by the atomic coordinates and atomic radii. The solute contains a set of point charges at the center of the atoms, generating an electrostatic field in the solute region and the solvent region. The electrostatic field in this system, including the solvent reaction field and the Coulomb field, can be calculated by solving the PB equation:

[0152]

[0153] where ε(r) is the dielectric constant, φ(r) is the electrostatic potential, ρ(r) is the solute charge, λ(r) is the Stern layer masking function, z i is the charge of ion type i, c i is the volume number density of ion type i far from the solute, k B is the Boltzmann constant, and T is the temperature.

[0154] For the non-polar solvation energy in formula (5), since the non-polar solvation energy is proportional to the solvent-accessible surface area (SASA), which reflects the hydrophobic effect, it can be calculated using the following formula (9).

[0155]

[0156] wherein γ and β are constants. In using the GB model, the default value of γ is 0.0072 and the default value of β is 0.0. In using the PB model, γ is a constant with the default value of 0.0378 and the default value of β is -0.5692.

[0157] In this embodiment, the MM-PBSA method or the MM-GBSA method can be completed by a molecular dynamics simulation software, and the molecular dynamics simulation software package is but not limited to AMBER, Gromacs, NAMD, etc. Wherein, the complete receptor, i.e. the target protein, and the single molecule conformation can be composed into a complex, and the NPT (constant-pressure, constant-temperature, isothermal-isobaric ensemble) ensemble molecular dynamics simulation is performed, i.e. the third binding free energy can be calculated by the MM-PBSA method or the MM-GBSA method.

[0158] S250, using the FEP method or the TI method to calculate the fourth binding free energy corresponding to the third binding free energy conformations that meet the third preset conditions after contacting with the target protein; and screening to obtain the molecular conformations with the fourth binding free energy meeting the fourth preset conditions.

[0159] FEP (Free Energy Perturbation) is a high-precision free energy calculation method based on a molecular force field, and TI (Thermodynamic integration) is a high-precision free energy calculation method based on a molecular force field. In this step, the fourth binding free energy corresponding to the molecular conformation screened in step S230 or step S240 is calculated by the FEP method or the TI method. After the fourth binding free energy corresponding to the molecular conformation is calculated by the FEP method or the TI method, in an embodiment, the molecular conformations of the same ligand molecule are sorted according to the numerical values of the corresponding fourth binding free energy, and the molecular conformations with the fourth binding free energy in the fourth preset range are obtained. In an embodiment, the molecular conformations of the same ligand molecule are sorted according to the numerical values of the fourth binding free energy from small to large, and the molecular conformations with the fourth preset proportion in the front rank are obtained. That is, the molecular conformations in the fourth preset range can be the molecular conformations with the fourth preset proportion in the front rank sorted according to the fourth binding free energy from small to large. The fourth preset proportion can be 0.1% to 100%. It can be understood that if the sorting is according to the numerical values from large to small, the molecular conformations with the proportion of 0.1% to 100% in the rear rank are obtained. Further, the molecular conformation with the smallest fourth binding free energy value in the same ligand molecule is obtained, that is, only one molecular conformation is retained for each ligand molecule, so that the molecular conformation with the strongest binding ability determined by the calculation result is obtained, the molecular conformations not meeting the fourth preset condition are filtered out, the redundant molecular conformations are further reduced, and the accuracy of selecting active compounds is improved.

[0160] As shown in the following, the calculation principle and method of calculating the fourth binding free energy corresponding to the molecular conformation by the FEP method or the TI method in this step will be specifically introduced. Figure 3

[0161] The FEP method and the TI method are high-precision methods for evaluating the binding strength of a drug small molecule and a target protein. This method is mainly based on the principle of thermodynamic cycle (Thermodynamic Cycle), and can calculate the binding free energy of the ligand through a series of gradual changes of atomic-level van der Waals parameters and charge parameters.

[0162] In the FEP method, the complete target protein receptor and the single molecular conformation form a complex, and the binding free energy from the binding state A to the binding state B of the complex is calculated by the Zwanzig equation, that is, the following formula (10), that is, the fourth binding free energy is calculated.

[0163]

[0164] ​where T is temperature, k B is the Boltzmann constant, and the term in triangular brackets <> represents the average over the state A simulation.

[0165] In the TI method, the free energy difference is calculated by defining a thermodynamic path between the states and integrating the ensemble-averaged enthalpy change along the path. Thus, the fourth binding free energy obtained by using the TI method can be calculated according to the following formula (11).

[0166]

[0167] where F(A) and F(B) are the lambda coupling potential functions corresponding to λ = 0 and λ = 1, respectively.

[0168] The above-mentioned FEP method or TI method in the present application can be used, but is not limited to, the XFEP package, the FEP+ package to complete the above-mentioned calculation. The basic principles and processes of calculation in the XFEP package are as follows: the complete target protein receptor and the single molecule conformation are combined into a complex, and the molecular dynamics simulation of the constant temperature and constant pressure (NPT) ensemble is first performed. After the simulation is completed, the corresponding binding free energy is calculated by the FEP method or the TI method. In order to facilitate understanding, for example, six atoms (three atoms l1, l2 and l3 from the ligand, and three atoms p1, p2 and p3 from the protein) are selected before the simulation to perform Boresch restraint. The atom l1 is the ligand atom closest to the geometric center of the ligand. The atom l2 is the ligand atom farthest from l1. The atom l3 is the ligand atom farthest from l1 under the condition that the angle l1-l2-l3 is between 60° and 120°. The atoms p1, p2 and p3 are selected from the CA, CB and N atoms within 5 angstroms of the non-glycine residue of the ligand.

[0169] The binding free energy simulation is carried out in two stages. In the first stage, six constraints are set, including one bond l1-p1, two angles l2-l1-p1 and l1-p1-p2, and three dihedral angles l3-l2-l1-p1, l2-l1-p1-p2 and l1-p1-p2-p3 are opened, while the partial charges on all ligand atoms are closed, and the lambda window is 0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9 and 1.0, etc. In the second stage, the LJ interaction between the ligand and the environment is closed while the constraints are maintained, and the lambda window is 0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.55, 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9, 0.95 and 1.0, etc. The simulation time for each lambda window is 2 ns. After the simulation is completed, the binding free energy can be calculated by the above-mentioned calculation formula.

[0170] In other embodiments, according to the acceptable accuracy of the prediction results of the virtual screening method of the present application, the steps S240 and S250 can be selectively executed, i.e. the steps S240 and S250 can not be executed, so that the subsequent step S260 is directly executed; or the steps S240 and S250 are selectively executed, i.e. only the step S240 or the step S250 is executed. When the steps S240 and / or S250 are not executed, the screening of the molecular conformations is reduced by one or two times, the amount of data calculation is simplified, the calculation resources are saved, and the calculation efficiency is improved. In other embodiments, when the virtual screening method of the present application determines to execute the steps S240 and S250, the two steps can be executed in reverse order, for example, the step S250 is executed first, and then the step S240 is executed, or the step S240 is executed first, and then the step S250 is executed. It can be understood that the step executed later will further screen the molecular conformations screened in the step executed earlier. In related embodiments, reference can be made to Embodiment Three to Embodiment Five below.

[0171] S260, the screened molecular conformations are grouped into a candidate conformation set, and active compounds are selected from the candidate conformation set.

[0172] In an embodiment, the molecular conformations with the second binding free energy meeting the second preset condition are grouped into the candidate conformation set, or the molecular conformations with the third binding free energy meeting the third preset condition are grouped into the candidate conformation set, or the molecular conformations with the fourth binding free energy meeting the fourth preset condition are grouped into the candidate conformation set.

[0173] That is, the molecular conformations in the candidate conformation set are related to whether the steps S240 and S250 are executed and the execution order of the steps S240 and S250. In the present embodiment, all the steps S210 to S250 are executed in sequence, and the candidate conformation set is composed of the molecular conformations with the fourth binding free energy meeting the fourth preset condition screened by the step S250. The molecular conformations screened by the last executed step in the above steps form the candidate conformation set, so that the molecular conformations in the candidate conformation set are molecular conformations with better binding ability after multiple screenings, and the corresponding ligand molecules are the most reactive molecules with the target protein in the molecular library, so that 0.1% to 100% of the compounds in the molecular conformations of these ligand molecules can be selected as potential active compounds for testing, the success rate of obtaining the required active compounds is improved, the cost of drug research and development is reduced, and the research and development efficiency is improved.

[0174] From the example, the virtual screening method of the present application can optionally use at least two methods such as MM-GBSA method or MM-PBSA method and FEP method or TI method to screen molecular conformations layer by layer on the basis of scoring function and semi-empirical molecular orbital method, to obtain a set of candidate conformations with better binding ability. Such design can quickly predict the binding ability of molecular conformations, and multiple screening can improve the accuracy of the prediction results of the binding ability, i.e., the molecular conformations with better binding ability can be screened, thereby effectively improving the success rate of discovering active compounds by virtual screening. At the same time, the computing resources consumed by different methods are different, and the execution or order can be flexibly selected or changed to meet the computing needs of users.

[0175] For ease of understanding, as shown in the flowchart, Figures 4 to 8 the following will further clarify the different method combinations used to obtain different sets of candidate conformations according to the above embodiments.

[0176] Example Three

[0177] Referring to Figure 4 In an embodiment, the virtual screening method of the present application comprises:

[0178] S310, molecular docking is performed on a plurality of ligand molecules in a molecular library and a target protein to obtain at least one molecular conformation of the ligand molecules after contacting the target protein.

[0179] S320, a preset scoring function is used to determine a first binding free energy corresponding to each molecular conformation after contacting the target protein; all molecular conformations of the ligand molecules are sorted according to the numerical values of the corresponding first binding free energies to obtain molecular conformations occupying a preset proportion in the ranking.

[0180] S330, a semi-empirical molecular orbital method is used to determine a second binding free energy of the molecular conformation whose first binding free energy meets a first preset condition after contacting the target protein, and to screen and obtain a molecular conformation whose second binding free energy meets a second preset condition.

[0181] S340, a FEP method or a TI method is used to calculate a fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after contacting the target protein; and a molecular conformation whose fourth binding free energy meets a fourth preset condition is screened and obtained.

[0182] S350, an MM-GBSA method or an MM-PBSA method is used to calculate a third binding free energy corresponding to the molecular conformation whose fourth binding free energy meets the fourth preset condition after contacting the target protein; and a molecular conformation whose third binding free energy meets a third preset condition is screened and obtained.

[0183] S360, the third binding free energy of the molecule conformation meeting the third preset condition is selected to form a candidate conformation set, and an active compound is selected from the candidate conformation set.

[0184] The difference between the embodiment and the embodiment two is that the order of the execution contents of the steps S340 and S350 is different from the order of the execution contents of the steps S240 and S250, and therefore, the step S340 further screens the second binding free energy meeting the second preset condition based on the second binding free energy obtained by the step S330, and the step S350 further screens the fourth binding free energy meeting the fourth preset condition based on the fourth binding free energy obtained by the step S340. The candidate conformation set in the step S360 is composed of the third binding free energy meeting the third preset condition obtained by the step S350.

[0185] Embodiment four

[0186] Referring to Figure 5 In an embodiment, the virtual screening method of the present application comprises the following steps:

[0187] S410, a plurality of ligand molecules in a molecule library are subjected to molecular docking with a target protein to obtain at least one molecule conformation after the ligand molecules contact the target protein.

[0188] S420, a preset scoring function is used to determine a first binding free energy corresponding to each molecule conformation after the molecule conformation contacts the target protein; all molecule conformations of the ligand molecules are sorted according to the numerical value of the corresponding first binding free energy to obtain molecule conformations occupying a preset proportion in ranking.

[0189] S430, a semi-empirical molecular orbital method is used to determine a second binding free energy of the molecule conformation meeting the first preset condition after the molecule conformation contacts the target protein, and a molecule conformation meeting a second preset condition of the second binding free energy is screened and obtained.

[0190] S440, a FEP method or a TI method is used to calculate a fourth binding free energy corresponding to the molecule conformation meeting the second preset condition of the second binding free energy after the molecule conformation contacts the target protein; and a molecule conformation meeting a fourth preset condition of the fourth binding free energy is screened and obtained.

[0191] S450, the fourth binding free energy of the molecule conformation meeting the fourth preset condition is selected to form a candidate conformation set, and an active compound is selected from the candidate conformation set.

[0192] The difference between the present embodiment and Embodiment Two is that the present embodiment cancels the calculation of the binding free energy corresponding to the molecular conformation after the target protein is contacted by the MM-GBSA method or the MM-PBSA method, and one step of screening the molecular conformation is omitted compared with Embodiment Two. Therefore, the to-be-selected conformation set obtained in the present step S450 is composed of the molecular conformations whose fourth binding free energy meets the fourth preset condition and are obtained by the fourth screening in S440.

[0193] Embodiment Five

[0194] Referring to Figure 6 In an embodiment, the virtual screening method of the present application comprises:

[0195] S510, performing molecular docking on the plurality of ligand molecules in the molecular library and the target protein to obtain at least one molecular conformation after the ligand molecules are contacted with the target protein.

[0196] S520, determining the first binding free energy corresponding to each molecular conformation after the target protein is contacted by using a preset scoring function; and sorting the molecular conformations of all ligand molecules according to the numerical value of the corresponding first binding free energy to obtain the molecular conformations occupying a preset proportion in ranking.

[0197] S530, determining the second binding free energy after the target protein is contacted by the molecular conformation whose first binding free energy meets the first preset condition based on a semi-empirical molecular orbital method, and screening to obtain the molecular conformation whose second binding free energy meets the second preset condition.

[0198] S540, calculating the third binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after the target protein is contacted by using the MM-GBSA method or the MM-PBSA method; and screening to obtain the molecular conformation whose third binding free energy meets the third preset condition.

[0199] S550, composing the to-be-selected conformation set by the molecular conformation whose third binding free energy meets the third preset condition, and selecting the active compound in the to-be-selected conformation set.

[0200] The difference between the present embodiment and Embodiment Two is that the present embodiment cancels the calculation of the binding free energy corresponding to the molecular conformation after the target protein is contacted by the MM-GBSA method or the MM-PBSA method, and one step of screening the molecular conformation is omitted compared with Embodiment Two. Therefore, the to-be-selected conformation set obtained in the present step S550 is composed of the molecular conformations whose third binding free energy meets the third preset condition and are obtained by the third screening in S540.

[0201] Embodiment Six

[0202] Referring to Figure 7In an embodiment, the virtual screening method of the present application comprises:

[0203] S610, performing molecular docking of the plurality of ligand molecules in the molecule library with the target protein to obtain at least one molecular conformation of the ligand molecules after contacting with the target protein.

[0204] S620, determining a first binding free energy corresponding to each molecular conformation after contacting with the target protein by using a preset scoring function; and screening to obtain a molecular conformation whose first binding free energy meets a first preset condition.

[0205] The screening to obtain a molecular conformation whose first binding free energy meets a first preset condition comprises: sorting all molecular conformations of the ligand molecules according to the numerical values of the corresponding first binding free energies, and obtaining molecular conformations occupying a preset proportion in the ranking.

[0206] S630, calculating a third binding free energy corresponding to the molecular conformation whose first binding free energy meets the first preset condition after contacting with the target protein by using an MM-GBSA method or an MM-PBSA method; and screening to obtain a molecular conformation whose third binding free energy meets a third preset condition.

[0207] S640, determining a second binding free energy of the molecular conformation whose third binding free energy meets the third preset condition after contacting with the target protein based on a semi-empirical molecular orbital method, and screening to obtain a molecular conformation whose second binding free energy meets a second preset condition.

[0208] S650, grouping the molecular conformation whose second binding free energy meets the second preset condition into a candidate conformation set, and selecting an active compound in the candidate conformation set.

[0209] The difference between the present embodiment and Embodiment Five is that the execution order of the contents of steps S630 and S640 is opposite to that of steps S530 and S540. Therefore, the candidate conformation set obtained in the present step S650 is composed of the molecular conformation whose second binding free energy meets the second preset condition obtained by screening in S640 after three screenings in S620 to S640.

[0210] Embodiment Seven

[0211] Referring to Figure 8 In an embodiment, the virtual screening method of the present application comprises:

[0212] S710, performing molecular docking of the plurality of ligand molecules in the molecule library with the target protein to obtain at least one molecular conformation of the ligand molecules after contacting with the target protein.

[0213] S720, determining, by using a preset scoring function, a first binding free energy corresponding to each molecular conformation after the molecular conformation contacts the target protein; and screening to obtain a molecular conformation whose first binding free energy meets a first preset condition.

[0214] The screening to obtain the molecular conformation whose first binding free energy meets the first preset condition includes: sorting all molecular conformations of the ligand molecules according to the numerical values of the corresponding first binding free energies, and obtaining molecular conformations occupying a preset proportion in ranking.

[0215] S730, calculating, by using an MM-GBSA method or an MM-PBSA method, a third binding free energy corresponding to the molecular conformation whose first binding free energy meets the first preset condition after the molecular conformation contacts the target protein; and screening to obtain a molecular conformation whose third binding free energy meets a third preset condition.

[0216] S740, determining, by using a semi-empirical molecular orbital method, a second binding free energy of the molecular conformation whose third binding free energy meets the third preset condition after the molecular conformation contacts the target protein, and screening to obtain a molecular conformation whose second binding free energy meets a second preset condition.

[0217] S750, calculating, by using a FEP method or a TI method, a fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after the molecular conformation contacts the target protein; and screening to obtain a molecular conformation whose fourth binding free energy meets a fourth preset condition.

[0218] S760, grouping the molecular conformation whose fourth binding free energy meets the fourth preset condition into a candidate conformation set, and selecting an active compound in the candidate conformation set.

[0219] The embodiment of the present application differs from the embodiment seven in that the embodiment of the present application adds step S750 to further screen the molecular conformation obtained in step S740. Therefore, the candidate conformation set obtained in step S760 is composed of the molecular conformation whose fourth binding free energy meets the fourth preset condition obtained by screening in step S750 after the molecular conformation is screened four times in steps S720 to S750.

[0220] Corresponding to the foregoing application function implementation method embodiment, the present application further provides a virtual screening device, an electronic device, and corresponding embodiments.

[0221] Figure 9 FIG. 1 is a structural schematic diagram of a virtual screening device according to an embodiment of the present application.

[0222] Referring to FIG. 1, Figure 9 The virtual screening device 30 in an embodiment of the present application includes a molecular docking module 310, a first screening module 320, and a second screening module 330. Wherein:

[0223] The molecule docking module 310 is configured to dock each ligand molecule in the molecule library with the target protein to obtain at least one molecule conformation of the ligand molecule after contacting with the target protein.

[0224] The first screening module 320 is configured to determine a first binding free energy of the molecule conformation after contacting with the target protein by using a preset scoring function, and screen a molecule conformation satisfying a first preset condition in terms of the first binding free energy.

[0225] The second screening module 330 is configured to determine a second binding free energy of the molecule conformation satisfying the first preset condition in terms of the first binding free energy after contacting with the target protein based on a semi-empirical molecular orbital method, and screen a molecule conformation satisfying a second preset condition in terms of the second binding free energy.

[0226] Further, the molecule docking module 310 is configured to dock each ligand molecule with the target protein to obtain one or more molecule conformations of each ligand molecule in the target protein.

[0227] The first screening module 320 is configured to determine a first binding free energy of each molecule conformation after contacting with the target protein by using a preset scoring function, and sort all molecule conformations of the ligand molecules according to the numerical values of the corresponding first binding free energies to obtain molecule conformations occupying a preset proportion in ranking.

[0228] The second screening module 330 is configured to determine a first one-point energy of residues in a preset range around a cavity of the target protein as a dummy protein, determine a second one-point energy of the dummy protein and the molecule conformation as a complex, determine a third one-point energy of the molecule conformation, and determine a second binding free energy of the molecule conformation after contacting with the target protein according to a difference between the second one-point energy and the first one-point energy and the third one-point energy.

[0229] In this embodiment, after obtaining the molecule conformations by the molecule docking module, the first screening module and the second screening module are sequentially used for twice screening to obtain a set of molecule conformations satisfying the requirements, so that the molecule conformations remaining after the screening have strong binding ability, thereby facilitating selection of active compounds from more advantageous ligand molecules, improving the accuracy of the prediction result, and then improving the success rate of screening the active compounds.

[0230] Referring to Figure 10 , Figure 10 is another structural schematic diagram of the virtual screening device according to an embodiment of the present application.

[0231] As Figure 10 shown, the virtual screening device 30 in an embodiment of the present application includes a molecule docking module 310, a first screening module 320, a second screening module 330, and a third screening module 340, which are sequentially executed. Wherein:

[0232] The molecular docking module 310, the first screening module 320, and the second screening module 330 are described above and will not be repeated here.

[0233] The third screening module 340 is configured to calculate, by using the MM-GBSA method or the MM-PBSA method, third binding free energies corresponding to the molecular conformations that meet the second preset condition in the second binding free energies, and screen to obtain the molecular conformations that meet a third preset condition in the third binding free energies. Further, the third screening module 340 is configured to, after calculating the third binding free energies corresponding to the molecular conformations that meet the second preset condition in the second binding free energies by using the MM-GBSA method or the MM-PBSA method, sort the molecular conformations of the same ligand molecule according to the numerical values of the corresponding third binding free energies, and obtain the molecular conformations whose third binding free energies are within a third preset range; for example, the molecular conformation with the minimum third binding free energy in the same ligand molecule is obtained.

[0234] Further, as shown in FIG. 3, the virtual screening device of the embodiment can further include a fourth screening module 350. The fourth screening module 350 is configured to calculate, by using the FEP method or the TI method, fourth binding free energies corresponding to the molecular conformations that meet the third preset condition in the third binding free energies, and screen to obtain the molecular conformations that meet a fourth preset condition in the fourth binding free energies. Further, the fourth screening module 350 is configured to, after calculating the fourth binding free energies corresponding to the molecular conformations that meet the third preset condition in the third binding free energies by using the FEP method or the TI method, sort the molecular conformations of the same ligand molecule according to the numerical values of the corresponding fourth binding free energies, and obtain the molecular conformations whose fourth binding free energies are within a fourth preset range; for example, the molecular conformation with the minimum fourth binding free energy in the same ligand molecule is obtained. Figure 10 Referring to FIG. 3, in another embodiment, the virtual screening device 30 of the embodiment includes the molecular docking module 310, the first screening module 320, the second screening module 330, and the fourth screening module 350 that are sequentially executed.

[0235] Figure 11 The molecular docking module 310, the first screening module 320, and the second screening module 330 are described above and will not be repeated here.

[0236] The fourth screening module 350 is configured to calculate, by using the FEP method or the TI method, fourth binding free energies corresponding to the molecular conformations that meet the second preset condition in the second binding free energies, and screen to obtain the molecular conformations that meet a fourth preset condition in the fourth binding free energies.

[0237] Further, as shown in FIG. 3, the virtual screening device of the embodiment can further include a fourth screening module 350. The fourth screening module 350 is configured to calculate, by using the FEP method or the TI method, fourth binding free energies corresponding to the molecular conformations that meet the third preset condition in the third binding free energies, and screen to obtain the molecular conformations that meet a fourth preset condition in the fourth binding free energies. Further, the fourth screening module 350 is configured to, after calculating the fourth binding free energies corresponding to the molecular conformations that meet the third preset condition in the third binding free energies by using the FEP method or the TI method, sort the molecular conformations of the same ligand molecule according to the numerical values of the corresponding fourth binding free energies, and obtain the molecular conformations whose fourth binding free energies are within a fourth preset range; for example, the molecular conformation with the minimum fourth binding free energy in the same ligand molecule is obtained.

[0238] Further, as shown in FIG. 3, the virtual screening device of the embodiment can further include a fourth screening module 350. The fourth screening module 350 is configured to calculate, by using the FEP method or the TI method, fourth binding free energies corresponding to the molecular conformations that meet the third preset condition in the third binding free energies, and screen to obtain the molecular conformations that meet a fourth preset condition in the fourth binding free energies. Further, the fourth screening module 350 is configured to, after calculating the fourth binding free energies corresponding to the molecular conformations that meet the third preset condition in the third binding free energies by using the FEP method or the TI method, sort the molecular conformations of the same ligand molecule according to the numerical values of the corresponding fourth binding free energies, and obtain the molecular conformations whose fourth binding free energies are within a fourth preset range; for example, the molecular conformation with the minimum fourth binding free energy in the same ligand molecule is obtained. Figure 11 ​As shown, the virtual screening device of the embodiment can further include a third screening module 340. The third screening module 340 is configured to calculate, by using the MM-GBSA method or the MM-PBSA method, a third binding free energy corresponding to a molecular conformation that meets a fourth preset condition for the fourth binding free energy after the molecular conformation contacts the target protein; and screen the molecular conformation that meets a third preset condition for the third binding free energy.

[0239] Referring back to Figure 9 , Figure 10 and Figure 11 , the virtual screening device can further include a conformation collection module 360. The conformation collection module 360 is configured to group the molecular conformation that meets the second preset condition for the second binding free energy into a candidate conformation set; or group the molecular conformation that meets the third preset condition for the third binding free energy into the candidate conformation set; or group the molecular conformation that meets the fourth preset condition for the fourth binding free energy into the candidate conformation set; and select a potential active compound in the candidate conformation set. Specifically, by determining one of the second screening module 330, the third screening module 340, and the fourth screening module 350 as the last screening module, the screening result corresponding to the screening module is selected to form the candidate conformation set.

[0240] In the virtual screening devices in the above two embodiments, after the molecular conformation is obtained by the molecular docking module, the first screening module, the second screening module, and the third screening module are sequentially used for three times of screening, or the first screening module, the second screening module, and the fourth screening module are sequentially used for three times of screening, or the first screening module, the second screening module, the third screening module, and the fourth screening module are sequentially used for four times of screening, or the first screening module, the second screening module, the fourth screening module, and the third screening module are sequentially used for four times of screening, to obtain a corresponding set of molecular conformations that meet the preset conditions. The molecular conformations remaining after the screening have strong binding ability, so that the active compound can be easily selected from the more advantageous ligand molecules, the accuracy of the prediction result is improved, and the success rate of screening the active compound is improved.

[0241] Referring back to Figure 12 In another embodiment, the virtual screening device 30 in the embodiment of the present application includes the molecular docking module 310, the first screening module 320, the third screening module 340, and the second screening module 330 that are sequentially executed. Wherein:

[0242] The molecular docking module 310 and the first screening module 320 are described above and will not be repeated here.

[0243] The third screening module 340 is configured to calculate, by using the MM-GBSA method or the MM-PBSA method, third binding free energy corresponding to the molecular conformation whose first binding free energy meets the first preset condition after the molecular conformation contacts the target protein; and screen to obtain the molecular conformation whose third binding free energy meets a third preset condition.

[0244] The second screening module 330 is configured to determine, by using the semi-empirical molecular orbital method, second binding free energy corresponding to the molecular conformation whose third binding free energy meets the third preset condition after the molecular conformation contacts the target protein, and screen to obtain the molecular conformation whose second binding free energy meets a second preset condition.

[0245] Further, referring to Figure 12 The virtual screening device of the embodiment can further include a fourth screening module 350. The fourth screening module 350 is configured to calculate, by using the FEP method or the TI method, fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after the molecular conformation contacts the target protein; and screen to obtain the molecular conformation whose fourth binding free energy meets a fourth preset condition.

[0246] Further, referring to Figure 12 The virtual screening device can further include a conformation set module 360. The conformation set module 360 is configured to group the molecular conformation whose second binding free energy meets the second preset condition into a candidate conformation set, or group the molecular conformation whose fourth binding free energy meets the fourth preset condition into the candidate conformation set, and select a potential active compound in the candidate conformation set. Specifically, by determining that one of the second screening module 330 and the fourth screening module 350 is the last screening module, the screening result corresponding to the screening module is selected to form the candidate conformation set.

[0247] The virtual screening device in the embodiment, after obtaining the molecular conformation by the molecular docking module, performs three screenings by the first screening module, the third screening module and the second screening module in sequence, or performs four screenings by the first screening module, the third screening module, the second screening module and the fourth screening module in sequence, to obtain a corresponding molecular conformation set meeting the preset condition, so that the molecular conformation remaining after the screening has a strong binding ability, thereby facilitating selection of an active compound in a more advantageous ligand molecule, improving the accuracy of the prediction result, and then improving the success rate of screening to obtain the active compound.

[0248] As to the device in the above-mentioned embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be described in detail here.

[0249] Figure 13 FIG. 1 is a structural schematic diagram of an electronic device according to an embodiment of the present application.

[0250] Referring toFigure 13 The electronic device 1000 includes a memory 1010 and a processor 1020.

[0251] The processor 1020 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gates or transistor logic, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can be any conventional processor.

[0252] The memory 1010 can include various types of storage units, such as a system memory, a read-only memory (ROM), and a permanent storage device. Among them, the ROM can store static data or instructions required by the processor 1020 or other modules of the computer. The permanent storage device can be a read and write storage device. The permanent storage device can be a non-volatile storage device that does not lose stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a mass storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In some other embodiments, the permanent storage device can be a removable storage device (such as a floppy disk, an optical drive). The system memory can be a read and write storage device or a volatile read and write storage device, such as a dynamic random access memory. The system memory can store some or all of the instructions and data required by the processor during runtime. In addition, the memory 1010 can include a combination of any computer readable storage media, including various types of semiconductor storage chips (such as DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), magnetic disks and / or optical disks. In some embodiments, the memory 1010 can include a read and / or write removable storage device, such as a compact disc (CD), a read-only digital versatile disc (such as DVD-ROM, double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (such as an SD card, a min SD card, a Micro-SD card, etc.), a magnetic floppy disk, etc. The computer readable storage medium does not include a carrier wave and a transient electronic signal transmitted through a wireless or wired transmission.

[0253] The memory 1010 stores executable code, which, when processed by the processor 1020, can cause the processor 1020 to perform part or all of the above-mentioned methods.

[0254] Furthermore, the method according to the present application can also be implemented as a computer program or a computer program product, which comprises computer program code instructions for executing some or all of the steps of the above-mentioned method according to the present application.

[0255] Alternatively, the present application can also be implemented as a computer readable storage medium (or a non-transitory machine readable storage medium or a machine readable storage medium) having stored thereon executable codes (or computer programs or computer instruction codes) which, when executed by a processor of an electronic device (or a server, etc.), cause the processor to perform some or all of the steps of the above-mentioned method according to the present application.

[0256] The above has described the embodiments of the present application, the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, practical application, or improvement to the technology in the market, or to enable other ordinary skilled in the art to understand the embodiments disclosed herein.

Claims

1. A virtual screening method, characterized in that: Molecular docking is performed between ligand molecules in the molecular library and target proteins to obtain at least one molecular conformation of the ligand molecules after contact with the target proteins. A preset scoring function is used to determine the first binding free energy after the molecular conformation comes into contact with the target protein, and molecular conformations whose first binding free energy meets the first preset condition are screened. Based on a semi-empirical molecular orbital method, the second binding free energy of a molecular conformation whose first binding free energy meets a first preset condition is determined upon contact with a target protein. Molecular conformations whose second binding free energy meets a second preset condition are then screened for acquisition. The semi-empirical molecular orbital method includes: treating residues within a preset range around the cavity of the target protein as an empty protein and determining the corresponding first single-point energy; forming a complex with the empty protein and the molecular conformation and determining the corresponding second single-point energy; determining the third single-point energy of the molecular conformation; and determining the second binding free energy of the molecular conformation upon contact with the target protein based on the difference between the second single-point energy and the first and third single-point energies.

2. The method according to claim 1, characterized in that, The step of determining the first binding free energy after the molecular conformation comes into contact with the target protein using a preset scoring function, and screening out molecular conformations whose first binding free energy meets a first preset condition, includes: A preset scoring function is used to determine the first binding free energy corresponding to each molecular conformation after it comes into contact with the target protein. The molecular conformations of all the ligand molecules are sorted according to the numerical value of the corresponding first binding free energy, and the molecular conformations whose first binding free energy is sorted within a first preset range are obtained.

3. The method according to claim 1, characterized in that, The target protein cavity is surrounded by a preset range of 3 to 5 angstroms.

4. The method according to claim 1, characterized in that, After determining the second binding free energy of the molecular conformation whose first binding free energy meets the first preset condition and then contacting the target protein using the semi-empirical molecular orbital method, and screening to obtain the molecular conformation whose second binding free energy meets the second preset condition, the method further includes: The third binding free energy corresponding to the contact between the molecular conformation whose second binding free energy meets the second preset condition and the target protein is calculated using the MM-GBSA or MM-PBSA method; molecular conformations whose third binding free energy meets the third preset condition are then screened for acquisition; or The fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition is calculated using the FEP method or the TI method after contact with the target protein; molecular conformations whose fourth binding free energy meets the fourth preset condition are screened and obtained.

5. The method according to claim 4, characterized in that: After calculating the third binding free energy corresponding to the molecular conformation that meets the second preset condition after contacting the target protein using the MM-GBSA or MM-PBSA method, the method further includes: The fourth binding free energy corresponding to the contact between the molecular conformation whose third binding free energy meets the third preset condition and the target protein is calculated using the FEP method or the TI method; molecular conformations whose fourth binding free energy meets the fourth preset condition are screened for acquisition; or The step of calculating the fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after contacting the target protein using the FEP method or TI method; and after screening and obtaining the molecular conformation whose fourth binding free energy meets the fourth preset condition, the step further includes: The third binding free energy corresponding to the molecular conformation that meets the fourth preset condition and comes into contact with the target protein is calculated using the MM-GBSA or MM-PBSA method; molecular conformations that meet the third preset condition and have the third binding free energy are then screened for.

6. The method according to claim 4 or 5, characterized in that, The method further includes: The molecular conformations whose second binding free energy meets the second preset condition are used to form a set of candidate conformations; or the molecular conformations whose third binding free energy meets the third preset condition are used to form a set of candidate conformations; or the molecular conformations whose fourth binding free energy meets the fourth preset condition are used to form a set of candidate conformations. Potentially active compounds are selected from the set of candidate conformations.

7. The method according to claim 4 or 5, characterized in that: The process of screening to obtain the molecular conformation whose second binding free energy meets the second preset condition includes: The molecular conformations of the same ligand molecule are sorted according to the numerical value of the corresponding second binding free energy to obtain the molecular conformations whose second binding free energy is sorted within a second preset range. and / or The process of screening to obtain the molecular conformation whose third binding free energy meets the third preset condition includes: The molecular conformations of the same ligand molecule are sorted according to the numerical value of the corresponding third binding free energy to obtain the molecular conformations whose third binding free energy is within a third preset range. and / or The process of screening to obtain the molecular conformation whose fourth binding free energy meets the fourth preset condition includes: The molecular conformations of the same ligand molecule are sorted according to the numerical value of the corresponding fourth binding free energy to obtain the molecular conformations whose fourth binding free energy is within a fourth preset range.

8. A virtual screening method, characterized in that: Molecular docking is performed between ligand molecules in the molecular library and target proteins to obtain at least one molecular conformation of the ligand molecules after contact with the target proteins. A preset scoring function is used to determine the first binding free energy after the molecular conformation comes into contact with the target protein, and molecular conformations whose first binding free energy meets the first preset condition are screened. The third binding free energy corresponding to the molecular conformation whose first binding free energy meets the first preset condition is calculated using the MM-GBSA method or the MM-PBSA method after contact with the target protein; the molecular conformation whose third binding free energy meets the third preset condition is screened to obtain the molecular conformation. Based on a semi-empirical molecular orbital method, the second binding free energy of a molecular conformation whose third binding free energy meets the third preset condition is determined after contact with the target protein, and molecular conformations whose second binding free energy meets the second preset condition are screened for acquisition; wherein, the semi-empirical molecular orbital method includes: taking the residues within a preset range around the cavity of the target protein as an empty protein and determining the corresponding first single-point energy; forming a complex between the empty protein and the molecular conformation and determining the corresponding second single-point energy; determining the third single-point energy of the molecular conformation; and determining the second binding free energy of the molecular conformation after contact with the target protein based on the difference between the second single-point energy and the first and third single-point energies.

9. The method according to claim 8, characterized in that, The step of determining the first binding free energy after the molecular conformation comes into contact with the target protein using a preset scoring function, and screening out molecular conformations whose first binding free energy meets a first preset condition, includes: A preset scoring function is used to determine the first binding free energy corresponding to each molecular conformation after it comes into contact with the target protein. The molecular conformations of all the ligand molecules are sorted according to the numerical value of the corresponding first binding free energy, and the molecular conformations that occupy a preset proportion of the ranking are obtained.

10. The method according to claim 8, characterized in that, The target protein cavity is surrounded by a preset range of 3 to 5 angstroms.

11. The method according to claim 8, characterized in that, After obtaining the molecular conformation whose second binding free energy meets the second preset condition through screening, the process further includes: The fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition is calculated using the FEP method or the TI method after contact with the target protein; molecular conformations whose fourth binding free energy meets the fourth preset condition are screened and obtained.

12. The method according to claim 11, characterized in that, The method further includes: The molecular conformations whose second binding free energy meets the second preset condition are used to form a set of candidate conformations; or the molecular conformations whose fourth binding free energy meets the fourth preset condition are used to form a set of candidate conformations. Potentially active compounds are selected from the set of candidate conformations.

13. The method according to claim 11, characterized in that: The process of screening to obtain the molecular conformation whose second binding free energy meets the second preset condition includes: The molecular conformations of the same ligand molecule are sorted according to the numerical value of the corresponding second binding free energy to obtain the molecular conformations whose second binding free energy is sorted within a second preset range. and / or The process of screening to obtain the molecular conformation whose third binding free energy meets the third preset condition includes: The molecular conformations of the same ligand molecule are sorted according to the numerical value of the corresponding third binding free energy to obtain the molecular conformations whose third binding free energy is within a third preset range. and / or The process of screening to obtain the molecular conformation whose fourth binding free energy meets the fourth preset condition includes: The molecular conformations of the same ligand molecule are sorted according to the numerical value of the corresponding fourth binding free energy to obtain the molecular conformations whose fourth binding free energy is within a fourth preset range.

14. A virtual screening device, characterized in that, include: The molecular docking module is used to perform molecular docking between ligand molecules in the molecular library and target proteins to obtain at least one molecular conformation of the ligand molecule after contact with the target protein. The first screening module is used to determine the first binding free energy after the molecular conformation comes into contact with the target protein using a preset scoring function, and to screen and obtain molecular conformations whose first binding free energy meets the first preset condition. The second screening module is used to determine the second binding free energy of a molecular conformation whose first binding free energy meets the first preset condition after contact with the target protein based on a semi-empirical molecular orbital method, and to screen and obtain the molecular conformation whose second binding free energy meets the second preset condition; wherein, the semi-empirical molecular orbital method includes: taking the residues within a preset range around the cavity of the target protein as an empty protein and determining the corresponding first single-point energy; forming a complex between the empty protein and the molecular conformation and determining the corresponding second single-point energy; determining the third single-point energy of the molecular conformation; and determining the second binding free energy of the molecular conformation after contact with the target protein based on the difference between the second single-point energy and the first and third single-point energies.

15. The apparatus according to claim 14, characterized in that, The device further includes a third screening module and / or a fourth screening module; wherein: The third screening module is used to calculate the third binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after contact with the target protein using the MM-GBSA or MM-PBSA method; or, to calculate the third binding free energy corresponding to the molecular conformation whose fourth binding free energy meets the fourth preset condition after contact with the target protein using the MM-GBSA or MM-PBSA method. The molecular conformation meeting the fourth preset condition is obtained by screening from the fourth binding free energies corresponding to the molecular conformations whose second binding free energy meets the second preset condition after contact with the target protein calculated by the fourth screening module using the FEP or TI method; the molecular conformation whose third binding free energy meets the third preset condition is then screened for. The fourth screening module is used to calculate the fourth binding free energy corresponding to the molecular conformation whose third binding free energy meets the third preset condition after contact with the target protein using the FEP method or the TI method. The molecular conformation that meets the third preset condition is obtained by screening from the third binding free energies corresponding to the molecular conformations whose second binding free energy meets the second preset condition after contact with the target protein calculated by the third screening module using the MM-GBSA method or the MM-PBSA method; or by calculating the fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after contact with the target protein using the FEP method or the TI method; and by screening to obtain the molecular conformation whose fourth binding free energy meets the fourth preset condition.

16. The apparatus according to claim 14, characterized in that: The target protein cavity is surrounded by a preset range of 3 to 5 angstroms.

17. A virtual screening device, characterized in that, include: The molecular docking module is used to perform molecular docking between ligand molecules in the molecular library and target proteins to obtain at least one molecular conformation of the ligand molecule after contact with the target protein. The first screening module is used to determine the first binding free energy after the molecular conformation comes into contact with the target protein using a preset scoring function, and to screen and obtain molecular conformations whose first binding free energy meets the first preset condition. The third screening module is used to calculate the third binding free energy corresponding to the molecular conformation whose first binding free energy meets the first preset condition after contacting the target protein using the MM-GBSA method or the MM-PBSA method; and to screen and obtain the molecular conformation whose third binding free energy meets the third preset condition. The second screening module is used to determine the second binding free energy of a molecular conformation whose third binding free energy meets the third preset condition after contact with the target protein based on a semi-empirical molecular orbital method, and to screen and obtain the molecular conformation whose second binding free energy meets the second preset condition; wherein, the semi-empirical molecular orbital method includes: taking the residues within a preset range around the cavity of the target protein as an empty protein and determining the corresponding first single-point energy; forming a complex between the empty protein and the molecular conformation and determining the corresponding second single-point energy; determining the third single-point energy of the molecular conformation; and determining the second binding free energy of the molecular conformation after contact with the target protein based on the difference between the second single-point energy and the first and third single-point energies.

18. The apparatus according to claim 17, characterized in that: The device further includes a fourth screening module, which is used to calculate the fourth binding free energy corresponding to the molecular conformation whose second binding free energy meets the second preset condition after contact with the target protein using the FEP method or the TI method; and to screen and obtain the molecular conformation whose fourth binding free energy meets the fourth preset condition.

19. The apparatus according to claim 17, characterized in that: The target protein cavity is surrounded by a preset range of 3 to 5 angstroms.

20. An electronic device, characterized in that, include: processor; as well as A memory having executable code stored thereon, which, when executed by the processor, causes the processor to perform the method as described in any one of claims 1-13.

21. A computer-readable storage medium having executable code stored thereon, which, when executed by a processor of an electronic device, causes the processor to perform the method as described in any one of claims 1-13.

Images