Drug molecule screening method and device based on coherent Isin machine

Abstract

The invention provides a drug molecule screening method and device, and the method comprises the steps: firstly, obtaining a chemical structure chart corresponding to a candidate molecule and a chemical structure chart corresponding to a target molecule; then, an association graph is generated, and the association graph comprises an association vertex set and an association edge set; the associated vertex set is a Cartesian product of two atom vertex sets in the two chemical structure diagrams, and a pair of associated vertexes connected with any associated edge meets the condition that chemical bond edges exist or do not exist between each pair of atom vertexes from the same chemical structure diagram in four atom vertexes corresponding to the pair of associated vertexes. Thirdly, searching a solution space represented by the association graph by using a coherent Isin machine system, and solving the ground state of the Hamiltonian H; and finally, marking the maximum similar part of the candidate molecule and the target molecule according to the ground state of the Hamiltonian H, and screening the drug molecule according to the maximum similar part. The method not only improves the accuracy of drug molecule screening, but also can improve the solving speed.

Description

Technical Field

[0001] The present invention relates to the field of drug discovery, and in particular to a drug molecule screening method and device based on a coherent Ising machine. Background Art

[0002] In the field of drug discovery, identifying the maximum common subgraph (MCS) between drug candidate molecules and target molecules is a critical task, crucial for understanding drug mechanisms of action and developing new drugs. The MCS problem aims to identify the largest similar subgraph between drug candidate and target molecules by comparing their chemical structures. The MCS problem is typically NP-hard, meaning that the computational resources and time required to find the optimal or near-optimal MCS increase dramatically with increasing molecule size.

[0003] Existing solutions to the MCS problem include: 1. Heuristic Search Algorithm: Maximum common subgraph identification methods based on heuristic search algorithms typically select node pairs based on heuristic strategies such as node degree, but these strategies are not well adapted to the complex structures of real-world molecular graphs. While traditional heuristic search algorithms can quickly find an approximate solution to MCS problems, they generally cannot guarantee a global optimal solution. They are also highly sensitive to parameter settings, easily trapped in local optimality, and have difficulty adapting to complex graph structures.

[0004] 2. Learning search model based on Graph Neural Network (GNN): Research has proposed a GNN-based learning search model based on a branch-and-bound algorithm. Instead of using traditional heuristics, it employs a GNN-based Deep Q-Network (DQN) to select node pairs. This approach uses supervision during the pre-training phase and guidance during the imitation learning phase to enhance DQN training. While this approach can select node pairs using the DQN network, it can encounter the problem of prematurely stopping the search during training, leading to suboptimal solutions. Furthermore, for large graph pairs, the search space is enormous, and the DQN may not be able to effectively explore all possible solutions. Furthermore, the training process of these models is complex, requiring supervision signals during the pre-training phase and guidance during the imitation learning phase.

[0005] 3. Heuristic Algorithm:

[0006] Heuristic algorithms quickly find approximate solutions by using empirical rules or simplifying problems. These algorithms can significantly reduce the solution time. Common heuristic algorithms include greedy algorithms, hill-climbing algorithms, simulated annealing algorithms, etc. These methods can quickly find "good enough" solutions when dealing with NP-hard problems, but they cannot guarantee finding the optimal solution.

[0007] Generally speaking, these methods have limitations when solving the MCS problem and often struggle to provide accurate MCS results within a reasonable time, especially when facing large-scale or complex molecular structures. Therefore, methods for solving the MCS problem require further research and improvement, particularly in enhancing the global search ability of the algorithm, reducing dependence on parameters, avoiding local optima, and improving the adaptability and efficiency of the algorithm. Summary of the Invention

[0008] To solve the above problems, the present invention provides a method and device for screening drug molecules based on a coherent Ising machine.

[0009] According to a first aspect, there is provided a method for screening drug molecules based on a coherent Ising machine, including the following steps: S1. Obtain the chemical structure diagrams corresponding to the candidate molecule and the target molecule respectively; each molecule includes a plurality of atoms and chemical bonds connecting the atoms; the chemical structure diagram includes an atomic vertex set and a chemical bond edge set.

[0010] S2. Generate an association graph according to the two obtained chemical structure diagrams; the association graph includes an association vertex set and an association edge set; the association vertex set is the Cartesian product between the two atomic vertex sets corresponding to the two chemical structure diagrams; any pair of associated vertices connected by an association edge satisfies: among the four atomic vertices corresponding to this pair of associated vertices, there is a chemical bond edge between each pair of atomic vertices from the same chemical structure diagram, or there is no chemical bond edge.

[0011] S3. Use the coherent Ising machine system to search the solution space represented by the association graph and solve for the ground state of the Hamiltonian H, where the Hamiltonian H is:

[0012] where the subscripts and respectively refer to the vertices and vertex in the chemical structure diagram of the candidate molecule, the subscripts and respectively refer to the vertices and vertex in the chemical structure diagram of the target molecule, represents whether the associated vertex corresponding to the subscript is selected, is the weight of , representing whether the associated vertex corresponding to the subscript is selected, representing whether there is an associated edge between the associated vertex corresponding to the subscript and the associated vertex corresponding to the subscript . K1 is the coefficient of .

[0013] S4. According to the ground state of the Hamiltonian H, label the most similar part between the candidate molecule and the target molecule, and filter the drug molecule accordingly.

[0014] In some embodiments, using the coherent Ising machine system to search the solution space represented by the associated graph and solve the ground state of the Hamiltonian H specifically includes: Initializing the coherent Ising machine according to the Hamiltonian H and starting the evolution process of the coherent Ising machine system.

[0015] During the evolution process of the coherent Ising machine system, measure the phase and intensity of the optical field in the coherent Ising machine system multiple times, and adjust the parameters of the coherent Ising machine system in real time according to the measurement results. Finally, obtain a collective oscillation mode far higher than the threshold, and the quantum spin state in this mode corresponds to the ground state of the Hamiltonian H.

[0016] In some embodiments, the coherent Ising machine system includes: An optical parametric oscillator network for simulating the spin interaction in the Ising model.

[0017] A phase-sensitive amplifier for amplifying signals with a specific phase.

[0018] A field programmable gate array for controlling the optical parametric oscillator network.

[0019] A phase / intensity measurer for measuring the phase and intensity of the optical field in the cavity.

[0020] An optical modulator for changing the phase difference between light beams, thereby simulating the interaction intensity between different spin states.

[0021] A beam splitter for generating an interference effect and simulating the connection relationship between spins in the Ising model.

[0022] An optical fiber used as a transmission medium for optical signals.

[0023] In some embodiments, the candidate molecule comes from a molecular library; the target molecule is a biomolecule acted on by a drug, which can be a protein, DNA, or hormone.

[0024] In some more specific embodiments, the screening of drug molecules specifically includes: traversing all candidate molecules in the molecular library, marking the largest similar part of all candidate molecules and the target molecule, and accordingly screening out the drug molecules with the highest similarity to the target molecule and having drug-active functional groups in the similar part.

[0025] According to the second aspect, the present invention also provides a drug molecule screening device, including: An acquisition module, configured to acquire the chemical structure diagrams corresponding to the candidate molecule and the target molecule respectively; each molecule includes a plurality of atoms and chemical bonds connecting the atoms; the chemical structure diagram includes an atomic vertex set and a chemical bond edge set.

[0026] A graph generation module, configured to generate an association graph according to the two acquired chemical structure diagrams; the association graph includes an association vertex set and an association edge set; the association vertex set is the Cartesian product between the two atomic vertex sets corresponding to the two chemical structure diagrams; any pair of association vertices connected by an association edge satisfies: among the four atomic vertices corresponding to this pair of association vertices, there is a chemical bond edge between each pair of atomic vertices from the same chemical structure diagram, or there is no chemical bond edge.

[0027] A solution module, configured to use a coherent Ising machine system to search the solution space represented by the association graph and solve the ground state of the Hamiltonian H, where the Hamiltonian H is:

[0028] where the subscripts and respectively refer to the vertices and vertex in the chemical structure diagram of the candidate molecule, the subscripts and respectively refer to the vertices and vertex in the chemical structure diagram of the target molecule, represents whether the association vertex corresponding to the subscript is selected, is the weight of represents whether the association vertex corresponding to the subscript is selected, represents whether there is an association edge between the association vertex corresponding to the subscript and the association vertex corresponding to the subscript , and K1 is the coefficient of

[0029] A marking module, configured to mark the largest similar part of the candidate molecule and the target molecule according to the ground state of the Hamiltonian H.

[0030] A screening module configured to screen drug molecules.

[0031] In some embodiments, the solving module is configured to: Initialize a coherent Ising machine according to the Hamiltonian H and start the system evolution process of the coherent Ising machine; During the system evolution process of the coherent Ising machine, measure the phases and intensities of the optical fields in the coherent Ising machine system multiple times, and adjust the parameters of the coherent Ising machine system in real time according to the measurement results, and finally obtain a collective oscillation mode far higher than the threshold, and the quantum spin state in this mode corresponds to the ground state of the Hamiltonian H.

[0032] In some embodiments, the coherent Ising machine system includes: An optical parametric oscillator network for simulating spin interactions in the Ising model.

[0033] A phase-sensitive amplifier for amplifying signals of a specific phase.

[0034] A field programmable gate array for controlling the optical parametric oscillator network.

[0035] A phase / intensity measurer for measuring the phases and intensities of the optical fields in the cavity.

[0036] An optical modulator for changing the phase difference between light beams, thereby simulating the interaction strength between different spin states.

[0037] A beam splitter for generating an interference effect and simulating the connection relationship between spins in the Ising model.

[0038] An optical fiber used as a transmission medium for optical signals.

[0039] In some embodiments, the candidate molecules are from a molecular library; the target molecule is a biomolecule to which the drug acts.

[0040] In some more specific embodiments, the screening module is specifically configured to: traverse all candidate molecules in the molecular library, label the largest similar part of all candidate molecules and the target molecule, and accordingly screen out drug molecules with the highest similarity to the target molecule and the similar part having drug-active functional groups.

[0041] The drug molecule screening method provided by the present invention can systematically search for all possible matching solutions to ensure that the found solution is globally optimal. In addition, the drug molecule screening method provided by the present invention is based on a coherent Ising machine, and compared with the traditional drug molecule MCS algorithm, it can explore the solution space more quickly and improve the solution speed. This method not only improves the accuracy of drug molecule screening, but also greatly enhances the ability to quickly identify drug molecules with expected biological activity from massive data. By focusing on the identification of MCS, the challenges of drug screening can be transformed into a more specific and computationally tractable problem, thus opening up an efficient and innovative path for the discovery of new drugs. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0043] Figure 1 Shows the flow of a drug molecule screening method based on a coherent Ising machine provided by the present invention; Figure 2 Shows the system structure of a coherent Ising machine provided by an embodiment of the present invention; Figure 3 Shows a protein drug molecule screening device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will describe the technical solutions in the embodiments of the present application in conjunction with the drawings.

[0045] A coherent Ising machine (CIM) is a computational platform based on quantum optics to simulate the Ising model. Its advantages lie in fast calculation speed and strong scalability, and it is suitable for dealing with various large-scale combinatorial optimization problems. The unique parallelism of the coherent Ising machine enables it to explore multiple possible solutions simultaneously during coherent evolution, providing potential advantages for solving NP-hard problems. When dealing with these problems, CIM can ensure that the found solution is globally optimal while maintaining the computational efficiency and solution accuracy.

[0046] Therefore, the drug screening method based on the Coherent Ising Machine (CIM) not only promises to significantly improve the computational efficiency of MCS search, but also can handle larger-scale molecular datasets while maintaining high accuracy. The present invention aims to utilize the computational advantages of CIM to provide an efficient solution to the MCS problem, with the expectation of achieving breakthrough progress in the field of drug discovery.

[0047] The following combines with the attached drawings to introduce the drug screening method provided by the present invention. The flowchart of this method is as Figure 1 shown, including the following steps:

[0048] S1. Obtain the chemical structure diagrams corresponding to the candidate molecule and the target molecule respectively; each molecule includes a plurality of atoms and chemical bonds connecting the atoms; the chemical structure diagram includes an atomic vertex set and a chemical bond edge set.

[0049] In some embodiments, the candidate molecules are from a molecular library; the target molecule is a biomolecule to which the drug acts.

[0050] Among them, the candidate molecule refers to a molecule considered for further development as a drug, from a molecular library, and can be a small molecule compound, natural product, biological agent, etc. The target molecule refers to the object to which the drug acts, usually a biomolecule involved in the disease process, and can be a protein, DNA, hormone, etc.

[0051] The chemical structure diagram provides detailed information about the spatial structure of the two molecules and the interaction of their internal atoms in the form of a graph. After converting the complex molecular structure into a chemical structure diagram, graph theory can be used to identify and compare the similarities between molecules. In the chemical structure diagram, the atomic vertex set includes vertices representing atoms, and the chemical bond edge set includes edges representing chemical bonds.

[0052] S2. Generate an association graph according to the two obtained chemical structure diagrams; the association graph includes an association vertex set and an association edge set; the association vertex set is the Cartesian product between the two atomic vertex sets corresponding to the two chemical structure diagrams; any pair of associated vertices connected by an associated edge meets the following condition: among the four atomic vertices corresponding to this pair of associated vertices, there are chemical bond edges between each pair of atomic vertices from the same chemical structure diagram, or there are no chemical bond edges.

[0053] The following combines with an example to explain the generation process of the association graph G:

[0054] Denote the two input chemical structure diagrams as G1 = (V1, E1) and G2 = (V2, E2), where V1 and V2 represent atomic vertex sets, and E1 and E2 represent chemical bond sets. The association graph G = (V, E) is constructed based on G1 and G2.

[0055] The associated vertex set V of the associated graph G is the Cartesian product V1 × V2 of V1 and V2. That is, each associated vertex in the associated vertex set V is a combination of atomic vertices in V1 and V2, and the associated vertex set V is all possible combinations of atomic vertices in V1 and V2.

[0056] In the associated graph, there are two associated vertices (u1, v1) and (u2, v2). If one of the following conditions is satisfied, an associated edge is added between these two associated vertices:

[0057] Condition 1: u1 and u2 are adjacent in G1 (i.e., (u1, u2) ∈ E1), and v1 and v2 are also adjacent in G2 (i.e., (v1, v2) ∈ E2).

[0058] Condition 2: u1 and u2 are not adjacent in G1 (i.e., ), and v1 and v2 are not adjacent in G2 (i.e., ).

[0059] In this way, the construction of the associated graph is completed, and the MCS problem is converted into a graph theory problem. The maximum clique of the associated graph represents the largest similar part between the two molecular structures. Next, by finding the maximum clique in the associated graph, the maximum common subgraph between the two molecular structures can be determined. This can be achieved by constructing the Hamiltonian H of the associated graph G and using CIM to solve for the ground state of the Hamiltonian H. The specific steps are as follows:

[0060] S3. Use the coherent Ising machine system to search the solution space represented by the associated graph and solve for the ground state of the Hamiltonian H. The Hamiltonian H is:

[0061] (1)

[0062] The meanings and possible values of each parameter are as follows:

[0063] The subscripts and respectively refer to the atomic vertices and atomic vertex in the chemical structure diagram of the candidate molecule, and the subscripts and respectively refer to the atomic vertices and atomic vertex in the chemical structure diagram of the target molecule.

[0064] represents whether the associated vertex corresponding to the subscript is selected. Similarly, represents whether the associated vertex corresponding to the subscript Whether the corresponding associated vertex is selected corresponds to the node spin state in the coherent Ising machine. For example, it can be set that when the spin is +1, it means this associated vertex is selected.

[0065] is the weight of, which is a positive number. For example, it can be taken as 1.

[0066] represents the association between the associated vertex corresponding to the subscript and the associated vertex corresponding to the subscript Whether there is an associated edge. For example, when there is an associated edge, it takes 1, and when there is no associated edge, it takes 0.

[0067] K1 is the coefficient of, which can be set to any positive number, such as 1.

[0068] In some embodiments, using the coherent Ising machine system, search the solution space represented by the associated graph to solve the ground state of the Hamiltonian H, specifically including:

[0069] (1) Initialize the coherent Ising machine according to the Hamiltonian H and start the evolution process of the coherent Ising machine system.

[0070] (2) During the evolution process of the coherent Ising machine system, measure the phase and intensity of the optical field in the coherent Ising machine system multiple times, and adjust the parameters of the coherent Ising machine system in real time according to the measurement results. Finally, a collective oscillation mode far higher than the threshold is obtained, and the quantum spin state in this mode corresponds to the ground state of the Hamiltonian H.

[0071] From the above, the ground state of the Hamiltonian H can be obtained using the coherent Ising machine system.

[0072] S4. According to the ground state of the Hamiltonian H, mark the most similar part between the candidate molecule and the target molecule.

[0073] Specifically, it is to inverse-encode the ground state of the Hamiltonian H, or the nodes in the maximum clique, into the corresponding atomic vertices and chemical bond edges in the chemical structure diagrams of the two molecules, which is the most similar part of the two molecules and serves as the final result of the MCS problem.

[0074] S5. Screen drug molecules according to the most similar part.

[0075] In some embodiments, this step can be implemented as: screen out the drug molecules with the highest similarity to the target molecule and having drug-active functional groups in the most similar part according to all candidate molecules in the molecular library and the most similar part to the target molecule.

[0076] The goal of drug molecule screening is to identify potential candidate molecules with similar structures to the target molecule from a large molecular library. In the previous steps, the maximum common subgraph between the candidate molecule and the target molecule has been identified. Finally, by traversing the candidate molecules in the molecular library and labeling their most similar parts to the target molecule one by one, the potential candidate molecule with the most similar structure to the target molecule can be found, completing the drug molecule screening.

[0077] The above is the method for predicting the protein side-chain structure based on a coherent Ising machine provided by the present invention. Next, the coherent Ising machine system used in this method will be briefly introduced.

[0078] This coherent Ising machine system uses a Doubly Resonant Optical Parametric Oscillator (DOPO) to implement artificial spins, and enhances the optical signal of a specific phase by placing a Phase Sensitive Amplifier (PSA) in the optical cavity. The PSA is an optical amplifier based on optical parametric amplification, which can effectively amplify the optical components with phases of 0 and π relative to the pump phase. Therefore, the DOPO only adopts phases of 0 or π higher than the oscillation threshold; thus, discrete phase states can be used to represent the Ising spin states. The interaction between DOPO pulses is realized by using measurement feedback technology. By repeatedly measuring and feedback modulating the process in the cavity multiple times while increasing the pump amplitude from 0, the "strongest" collective oscillation mode far above the threshold will finally be obtained, which corresponds to the best solution to a given Ising problem.

[0079] The schematic diagram of this coherent Ising machine system is as Figure 2 shown, including:

[0080] (1) An optical parametric oscillator network for simulating the spin interaction in the Ising model.

[0081] (2) A phase sensitive amplifier for amplifying the signal of a specific phase.

[0082] (3) A Field-Programmable Gate Array (FPGA) for controlling the optical parametric oscillator network.

[0083] The FPGA can configure its internal circuit through programming, and can be used to adjust the parameters of the coherent Ising machine system in real time according to the measurement results during the evolution process of the coherent Ising machine system.

[0084] (4) A phase / intensity measurer for measuring the phase and intensity of the optical field in the cavity.

[0085] (5) An optical modulator for changing the phase difference between light beams, thereby simulating the interaction strength between different spin states.

[0086] (6) A beam splitter for generating an interference effect and simulating the connection relationship between spins in the Ising model.

[0087] (7) An optical fiber used as a transmission medium for optical signals.

[0088] As can be seen from the above, compared with the prior art, the drug molecule screening method provided by the present invention has the following beneficial effects:

[0089] 1. Guarantee of the global optimal solution: This method can systematically search all possible matching schemes to ensure that the solution found is globally optimal. This is particularly important in the NP-hard MCS problem because although many traditional heuristic algorithms can quickly find approximate solutions, they cannot guarantee the optimality of the solutions.

[0090] 2. Fast and accurate solution: This method is implemented based on a coherent Ising machine system. Due to the use of coherent effects, the coherent Ising machine can explore the solution space more quickly, improving the solution speed. It is particularly suitable for the clique problem of large graphs. The acceleration advantage of the Ising machine goes beyond the field of complexity theory, providing a polynomial improvement in scaling or a constant pre-factor advantage, which has a great impact on the running time of large problems. That is, as the system scale increases, the advantage of this method becomes more and more obvious, which is crucial for drug screening and molecular structure analysis.

[0091] 3. Efficient and innovative path: Solving the MCS problem by the coherent Ising machine in the present invention not only improves the accuracy of the screening process but also greatly enhances our ability to quickly identify molecules with expected biological activities from a large amount of data. By focusing on the identification of MCS, the challenges of drug screening are transformed into a more specific and computationally tractable problem, thus opening up an efficient and innovative path for discovering new drugs.

[0092] The present invention also provides a drug molecule screening device 300, the schematic diagram of which is as Figure 3 shown, including:

[0093] An acquisition module 301 for acquiring the chemical structure diagrams corresponding to candidate molecules and target molecules respectively; each molecule includes a plurality of atoms and chemical bonds connecting the atoms; the chemical structure diagram includes an atomic vertex set and a chemical bond edge set;

[0094] A graph generation module 302, configured to generate an association graph according to two obtained chemical structure graphs; the association graph includes an association vertex set and an association edge set; the association vertex set is the Cartesian product between two atom vertex sets corresponding to the two chemical structure graphs; any pair of associated vertices connected by an associated edge satisfies: among the four atom vertices corresponding to the pair of associated vertices, there is a chemical bond edge between each pair of atom vertices from the same chemical structure graph, or there is no chemical bond edge;

[0095] A solution module 303, configured to use a coherent Ising machine system to search the solution space represented by the association graph and solve the ground state of the Hamiltonian H, where the Hamiltonian H is:

[0096]

[0097] where the subscripts and respectively refer to the vertices and vertices in the chemical structure graph of the candidate molecule, the subscripts and respectively refer to the vertices and vertices in the chemical structure graph of the target molecule, represents whether the associated vertex corresponding to the subscript is selected, is the weight of, represents whether the associated vertex corresponding to the subscript is selected, represents whether there is an associated edge between the associated vertex corresponding to the subscript and the associated vertex corresponding to the subscript , and K1 is the coefficient of;

[0098] A labeling module 304, configured to label the most similar part between the candidate molecule and the target molecule according to the ground state of the Hamiltonian H.

[0099] A screening module 305, configured to screen drug molecules.

[0100] It should be noted that the above device can execute the foregoing drug molecule screening method. For the functions of each module, reference can also be made to the foregoing introduction to the method, which will not be elaborated here.

[0101] In the description of the embodiments of the present application, words such as "exemplary", "for example" or "for illustration" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "exemplary", "for example" or "for illustration" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary", "for example" or "for illustration" is intended to present the relevant concepts in a specific manner.

[0102] In the description of the embodiments of the present application, the term "and / or" is merely a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, B exists alone, and both A and B exist simultaneously. In addition, unless otherwise specified, the meaning of the term "plural" refers to two or more.

[0103] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly indicating the technical features indicated. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The terms "comprise", "include", "have" and their variants all mean "including but not limited to", unless otherwise particularly emphasized in other ways.

[0104] The specific embodiments described above further elaborate on the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for screening drug molecules, characterized in that, It includes the following steps: Obtain the chemical structure diagrams corresponding to the candidate molecule and the target molecule respectively; each molecule includes a plurality of atoms and chemical bonds connecting the atoms; the chemical structure diagram includes an atomic vertex set and a chemical bond edge set; Generate an association graph according to the two obtained chemical structure diagrams; the association graph includes an association vertex set and an association edge set; the association vertex set is the Cartesian product between the two atomic vertex sets corresponding to the two chemical structure diagrams; For any pair of associated vertices connected by an associated edge, it satisfies that among the four atomic vertices corresponding to this pair of associated vertices, there is a chemical bond edge between each pair of atomic vertices from the same chemical structure diagram, or there is no chemical bond edge; Use a coherent Ising machine system to search the solution space represented by the association graph and solve the ground state of the Hamiltonian H, where the Hamiltonian H is: Among them, the subscripts and respectively refer to the vertices and vertex in the chemical structure diagram of the candidate molecule. The subscripts and respectively refer to the vertices and vertex in the chemical structure diagram of the target molecule. represents whether the associated vertex corresponding to the subscript is selected. is the weight of . . represents whether the associated vertex corresponding to the subscript is selected. . represents whether there is an associated edge between the associated vertex corresponding to the subscript and the associated vertex corresponding to the subscripts . K1 is the coefficient of . . . According to the ground state of the Hamiltonian H, label the most similar part between the candidate molecule and the target molecule, and accordingly screen drug molecules.

2. The method according to claim 1, characterized in that, The step of using a coherent Ising machine system to search the solution space represented by the association graph and solve the ground state of the Hamiltonian H specifically includes: Initialize the coherent Ising machine according to the Hamiltonian H and start the evolution process of the coherent Ising machine system; During the evolution process of the coherent Ising machine system, measure the phase and intensity of the optical field in the coherent Ising machine system multiple times, and adjust the parameters of the coherent Ising machine system in real time according to the measurement results. Finally, obtain a collective oscillation mode far higher than the threshold, and the quantum spin state in this mode corresponds to the ground state of the Hamiltonian H.

3. The method according to claim 1, characterized in that, The coherent Ising machine system includes: An optical parametric oscillator network for simulating the spin interaction in the Ising model; A phase-sensitive amplifier for amplifying signals with specific phases; A field programmable gate array for controlling the optical parametric oscillator network; A phase / intensity measurer for measuring the phase and intensity of the intracavity optical field; An optical modulator for changing the phase difference between light beams, thereby simulating the interaction strength between different spin states; A beam splitter for generating an interference effect and simulating the connection relationship between spins in the Ising model; An optical fiber used as a transmission medium for optical signals.

4. The method according to claim 1, characterized in that, The candidate molecule comes from a molecular library; the target molecule is a biomolecule acted on by a drug.

5. The method according to claim 4, wherein the screening of the drug molecules specifically comprises: Traverse all candidate molecules in the molecular library, label the most similar part between all candidate molecules and the target molecule, and accordingly screen out drug molecules with the highest similarity to the target molecule and the similar part having drug-active functional groups.

6. A drug molecule screening device, characterized in that, It includes: An acquisition module for obtaining the chemical structure diagrams corresponding to the candidate molecule and the target molecule respectively; Each molecule includes a plurality of atoms and chemical bonds connecting the atoms; the chemical structure diagram includes an atomic vertex set and a chemical bond edge set; A graph generation module for generating an association graph according to the two obtained chemical structure diagrams; the association graph includes an association vertex set and an association edge set; the association vertex set is the Cartesian product between the two atomic vertex sets corresponding to the two chemical structure diagrams; For any pair of associated vertices connected by an associated edge, the following condition is satisfied: among the four atomic vertices corresponding to the pair of associated vertices, there is a chemical bond edge between each pair of atomic vertices from the same chemical structure diagram, or there is no chemical bond edge between them; A solving module, configured to search the solution space represented by the associated graph by using a coherent Ising machine system to solve the ground state of the Hamiltonian H, where the Hamiltonian H is: Among them, the subscripts and respectively refer to the vertices and vertex in the chemical structure diagram of the candidate molecule. The subscripts and respectively refer to the vertices and vertex in the chemical structure diagram of the target molecule. represents whether the associated vertex corresponding to the subscript , is selected. is the weight. represents whether the associated vertex corresponding to the subscript , is selected. represents whether there is an associated edge between the associated vertex corresponding to the subscript , and the associated vertex corresponding to the subscripts , . K1 is the coefficient. A labeling module, configured to label the most similar part between the candidate molecule and the target molecule according to the ground state of the Hamiltonian H; A screening module, configured to screen drug molecules.

7. The device according to claim 6, wherein The solving module is specifically configured as follows: Initialize a coherent Ising machine according to the Hamiltonian H and start the evolution process of the coherent Ising machine system; During the evolution process of the coherent Ising machine system, measure the phase and intensity of the optical field in the coherent Ising machine system multiple times, and adjust the parameters of the coherent Ising machine system in real time according to the measurement results. Finally, a collective oscillation mode much higher than the threshold is obtained, and the quantum spin state in this mode corresponds to the ground state of the Hamiltonian H.

8. The device according to claim 6, characterized in that The coherent Ising machine system includes: An optical parametric oscillator network, configured to simulate the spin interaction in the Ising model; A phase-sensitive amplifier, configured to amplify a signal with a specific phase; A field programmable gate array, configured to control the optical parametric oscillator network; A phase / intensity measurer, configured to measure the phase and intensity of the intracavity optical field; An optical modulator, configured to change the phase difference between light beams, thereby simulating the interaction strength between different spin states; A beam splitter, configured to generate an interference effect and simulate the connection relationship between spins in the Ising model; An optical fiber, used as a transmission medium for optical signals.

9. The device according to claim 6, characterized in that The candidate molecule is from a molecular library; the target molecule is a biomolecule acted on by a drug.

10. The device according to claim 9, characterized in that, The screening module is specifically configured as follows: traverse all candidate molecules in the molecular library, label the most similar part between all candidate molecules and the target molecule, and accordingly screen out drug molecules with the highest similarity to the target molecule and with a drug-active functional group in the similar part.

Images