Drug molecule generation and model training methods, apparatuses, and devices

Abstract

The present disclosure provides a drug molecule generation and model training method, device and equipment, relates to the technical field of artificial intelligence, in particular to the technical field of natural language processing, deep learning, computational biology and chemistry. The drug molecule generation method comprises the following steps: generating a candidate drug molecule and a corresponding candidate synthesis route based on a target target point and a pre-constructed real synthesis route; obtaining the affinity between the candidate drug molecule and the target target point; determining a target drug molecule in the candidate drug molecule based on the candidate synthesis route and the affinity.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, specifically to the fields of natural language processing, deep learning, computational biology and chemistry, and in particular to a method, apparatus and device for drug molecule generation and model training. Background Technology

[0002] Molecular generation tasks involve iteratively generating unknown molecules at a given target site in order to obtain small drug molecules with high target affinity. Summary of the Invention

[0003] This disclosure provides a method, apparatus, and device for drug molecule generation and model training.

[0004] According to one aspect of this disclosure, a method for generating drug molecules is provided, comprising: generating candidate drug molecules and their corresponding candidate synthetic routes based on a target site and a pre-constructed real synthetic route; obtaining the affinity between the candidate drug molecules and the target site; and determining a target drug molecule among the candidate drug molecules based on the candidate synthetic route and the affinity.

[0005] According to another aspect of this disclosure, a method for training a drug molecule generation model is provided, comprising: using a drug molecule generation model to process an input target sample and a pre-constructed real synthetic route to output a predicted drug molecule and its corresponding predicted synthetic route; obtaining the predicted affinity between the predicted drug molecule and the target sample; constructing a target loss function based on the predicted synthetic route and the predicted affinity; and adjusting the model parameters of the drug molecule generation model based on the target loss function.

[0006] According to another aspect of this disclosure, a drug molecule generation apparatus is provided, comprising: a generation module for generating candidate drug molecules and their corresponding candidate synthetic routes based on a target site and a pre-constructed real synthetic route; an acquisition module for determining the affinity between the candidate drug molecules and the target site; and a determination module for determining a target drug molecule among the candidate drug molecules based on the candidate synthetic route and the affinity.

[0007] According to another aspect of this disclosure, a drug molecule generation model training apparatus is provided, comprising: a generation module for processing an input target sample and a pre-constructed real synthetic route using a drug molecule generation model to output a predicted drug molecule and its corresponding predicted synthetic route; an acquisition module for acquiring the predicted affinity between the predicted drug molecule and the target sample; a construction module for constructing a target loss function based on the predicted synthetic route and the predicted affinity; and an adjustment module for adjusting the model parameters of the drug molecule generation model based on the target loss function.

[0008] According to another aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to said at least one processor; wherein the memory stores instructions executable by said at least one processor, said instructions being executed by said at least one processor to enable said at least one processor to perform the method as described in any of the foregoing aspects.

[0009] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are configured to cause the computer to perform the method according to any of the preceding aspects.

[0010] According to another aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method according to any of the preceding aspects.

[0011] According to the technical solution disclosed herein, the syntheticability and accuracy of target drug molecules can be improved.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0013] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0014] Figure 1 This is a schematic diagram based on the first embodiment of the present disclosure;

[0015] Figure 2 This is a schematic diagram illustrating the application scenarios used to implement the embodiments of this disclosure;

[0016] Figure 3 This is a schematic diagram of a drug molecule generation model provided according to embodiments of this disclosure;

[0017] Figure 4This is a schematic diagram according to the second embodiment of the present disclosure;

[0018] Figure 5 This is a schematic diagram according to the third embodiment of the present disclosure;

[0019] Figure 6 This is a schematic diagram according to the fourth embodiment of the present disclosure;

[0020] Figure 7 This is a schematic diagram according to the fifth embodiment of the present disclosure;

[0021] Figure 8 This is a schematic diagram of an electronic device used to implement the drug molecule generation method or drug molecule generation model training method of the embodiments of this disclosure. Detailed Implementation

[0022] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0023] In related technologies, only the affinity between the generated drug molecule and the target site is usually considered, which may lead to the problem that the final target drug molecule may be impossible to synthesize.

[0024] To improve the syntheticability and accuracy of target drug molecules, the present disclosure provides the following embodiments.

[0025] Figure 1 This is a schematic diagram based on the first embodiment of the present disclosure. This embodiment provides a method for generating drug molecules, the method comprising:

[0026] 101. Based on the target and pre-constructed real synthetic routes, generate candidate drug molecules and their corresponding candidate synthetic routes.

[0027] 102. Obtain the affinity between the candidate drug molecule and the target site.

[0028] 103. Based on the candidate synthetic route and the affinity, identify the target drug molecule among the candidate drug molecules.

[0029] A drug target is a protein that a drug directly binds to, such as an enzyme, ion channel, receptor, or other biomolecule. Most drug targets are proteins, where chemicals used to treat or diagnose diseases selectively interact with the target protein, causing changes in its biological pathways or functions.

[0030] The target site refers to the target point that is to interact with the target drug molecule, and it is a known quantity.

[0031] Synthetic routes are used to illustrate the synthetic process of drug molecules. Drug synthesis is the process of gradually transforming raw material compounds into target drug molecules through chemical synthesis methods.

[0032] A true synthetic route refers to a synthetic route obtained based on wet experiments.

[0033] Dry experiments refer to experiments conducted using computers and bioinformatics technologies. Wet experiments refer to experiments involving the handling of liquids and biological samples in a laboratory setting.

[0034] Since wet experiments are conducted in the laboratory and are verified in the laboratory, they have practical syntheticability.

[0035] After obtaining the target and the actual synthetic route, these are input into the generative model, which outputs candidate drug molecules and their corresponding candidate synthetic routes. The generative model is a pre-trained deep learning model, such as the Transformer model.

[0036] There is at least one true synthetic route, and each true synthetic route can generate at least one candidate drug molecule and its corresponding candidate synthetic route.

[0037] Typically, each real synthetic route can generate multiple candidate drug molecules and their corresponding candidate synthetic routes. For example, based on real synthetic route 'a', a first candidate drug molecule and its corresponding first candidate synthetic route, a second candidate drug molecule and its corresponding second candidate synthetic route, and so on, can be generated.

[0038] For each candidate drug molecule, the affinity between the candidate drug molecule and the target site can be calculated.

[0039] Affinity refers to the affinity between a drug and its target, reflecting the ability of the drug to bind to the target. The greater the affinity, the tighter the binding between the two, and the stronger the drug activity.

[0040] After obtaining candidate synthesis routes, the syntheticity of the corresponding candidate drug molecules can be analyzed based on the candidate synthesis routes. For example, the candidate synthesis routes can be input into a pre-trained analysis model, and the output is the syntheticity result, which includes whether the drug is synthetic or not.

[0041] For each candidate drug molecule, its corresponding affinity can be obtained, and based on the candidate synthetic route of the candidate drug molecule, it can be analyzed whether the candidate drug molecule is synthetic.

[0042] Subsequently, the target drug molecule can be identified from the candidate drug molecules based on their syntheticity and affinity for the target.

[0043] Specifically, synthetic drug candidates can be obtained based on their syntheticity, and then the candidate with the highest affinity can be selected as the target drug molecule.

[0044] In this embodiment, by obtaining candidate synthetic routes for candidate drug molecules, the synthetic feasibility of candidate drug molecules can be analyzed based on the candidate synthetic routes, thereby improving the synthetic feasibility of target drug molecules. In addition, generating candidate drug molecules based on real synthetic routes can improve the accuracy of candidate drug molecules, and thus improve the accuracy of target drug molecules determined based on candidate drug molecules.

[0045] To better understand the embodiments of this disclosure, the application scenarios to which the embodiments of this disclosure may be applied are described.

[0046] Figure 2 This is a schematic diagram illustrating the application scenarios used to implement the embodiments of this disclosure.

[0047] like Figure 2 As shown, an application (APP) for drug molecule generation can be installed on the user terminal 201. This APP interacts with a server, which is deployed on server 202. The server can be a local server or a cloud server, a single server or a server cluster. The user terminal and the server can connect via a wired network and / or a wireless network.

[0048] Users can send target information, such as target name, identifier, or representation sequence, to the server through the APP; the server can obtain the target representation sequence by querying or directly based on the information sent by the user terminal.

[0049] The server can also obtain pre-built real synthesis routes.

[0050] A real synthetic route can be constructed based on single-step reaction data, which is known data, such as data obtained from an open-source single-step reaction database or a purchased commercial database.

[0051] After obtaining the single-step reaction data, it can be pieced together according to chemical reaction rules to obtain the actual synthetic route. Generally speaking, a large number of actual synthetic routes can be obtained, which in turn can generate a large number of candidate drug molecules.

[0052] After obtaining the target and the actual synthetic route, the characterization sequence of the target and the actual synthetic route are input into the drug molecule synthesis model, and the output is the candidate drug molecule and its corresponding candidate synthetic route.

[0053] After obtaining the candidate drug molecule, the affinity between the candidate drug molecule and the target site is calculated. For example, a preset algorithm is used to calculate the docking score between the candidate drug molecule and the target site. The larger the docking score, the stronger the affinity.

[0054] After obtaining candidate synthetic routes, the syntheticity of candidate drug molecules is calculated. For example, the candidate synthetic routes are input into a preset analysis model, and the output is the syntheticity result. The syntheticity result is used to indicate whether the candidate drug molecule corresponding to the candidate synthetic route is synthetic or not synthetic.

[0055] Once the syntheticity and affinity of the candidate drug molecule are obtained, the synthetic candidate drug molecule with the highest affinity can be used as the target drug molecule.

[0056] The candidate drug molecules output by the drug molecule synthesis model can be specific molecular structure diagrams, and the corresponding target drug molecules can also be specific molecular structure diagrams. A molecular structure diagram refers to a graphical structure composed of atoms and chemical bonds. The server can then feed back the molecular structure diagram of the target drug molecule to the app on the user's terminal, where the app displays it to the user. This embodiment uses server-side processing as an example; it can be understood that if the user terminal has the corresponding capabilities, it can also execute the relevant processes locally on the user terminal.

[0057] Figure 3 This is a schematic diagram of a drug molecule generation model provided according to embodiments of this disclosure.

[0058] like Figure 3 As shown, the drug molecule generation model includes: a first embedding layer, a second embedding layer, an encoding layer, and a decoding layer.

[0059] The first embedding layer is used to convert the target points into embedding vectors. Specifically, after the representation sequence of the target points is input into the first embedding layer, the output is the first embedding vector.

[0060] The second embedding layer is used to convert the real synthesis route into an embedding vector. Specifically, after the representation sequence of the real synthesis route is input into the second embedding layer, the output is the second embedding vector.

[0061] Generally speaking, there are multiple real synthesis routes. In this embodiment, the real synthesis routes include real synthesis route a and real synthesis route b as an example.

[0062] For each real synthesis route, a second embedding layer is used to convert it into a corresponding embedding vector. Different real synthesis routes can share the same second embedding layer.

[0063] After obtaining the first embedding vector and the second embedding vector, the first embedding vector is added to the second embedding vector corresponding to each real synthesis route to obtain the fusion vector corresponding to each real synthesis route. For example, for real synthesis route a, the first embedding vector is added to the second embedding vector corresponding to real synthesis route a to obtain the fusion vector corresponding to real synthesis route a.

[0064] After obtaining the fusion vector corresponding to each real synthesis route, the fusion vector of the coding layer is used for encoding to obtain the hidden layer vector. Different real synthesis routes can share the same coding layer. For example, for real synthesis route 'a', the fusion vector corresponding to real synthesis route 'a' is encoded using the coding layer to obtain the hidden layer vector corresponding to real synthesis route 'a'.

[0065] After obtaining the hidden layer vector corresponding to each real synthesis route, a decoding layer is used to decode the hidden layer vector to obtain the candidate drug molecule and its corresponding candidate synthesis route.

[0066] For each real synthetic route, multiple candidate drug molecules and their corresponding candidate synthetic routes can be generated. For example, for real synthetic route 'a', different decoding layers are used to obtain different candidate drug molecules and their corresponding candidate synthetic routes. For instance, decoding layer-1 is used to decode the hidden layer vector corresponding to real synthetic route 'a' to obtain candidate drug molecule ad1 and its corresponding candidate synthetic route ar1, ..., and decoding layer-n is used to decode the hidden layer vector corresponding to real synthetic route 'a' to obtain candidate drug molecule adn and its corresponding candidate synthetic route arn.

[0067] Different real synthetic routes may or may not share the decoding layer. Taking sharing as an example, for real synthetic route b, if decoding layer-1 is used to decode the hidden layer vector corresponding to real synthetic route b, candidate drug molecule bd1 and its corresponding candidate synthetic route br1, ... are obtained. If decoding layer-n is used to decode the hidden layer vector corresponding to real synthetic route b, candidate drug molecule bdn and its corresponding candidate synthetic route brn are obtained.

[0068] After obtaining multiple candidate drug molecules and their corresponding candidate synthetic routes, the synthetic feasibility of the candidate drug molecules is determined by the candidate synthetic routes, and the affinity between the candidate drug molecules and the target site is calculated. Then, the synthetic candidate drug molecule with the highest affinity can be selected as the target drug molecule.

[0069] In conjunction with the above application scenarios, this disclosure also provides the following embodiments.

[0070] Figure 4This is a schematic diagram according to a second embodiment of the present disclosure. This embodiment provides a method for generating drug molecules, the method comprising:

[0071] 401. Based on the pre-acquired single-step reaction data, construct the actual synthetic route.

[0072] One-step reaction data can be obtained from open-source single-step reaction databases or purchased commercial databases.

[0073] After obtaining the single-step reaction data, it can be pieced together according to chemical reaction rules to obtain the actual synthetic route. Generally speaking, a large number of actual synthetic routes can be obtained, which in turn can generate a large number of candidate drug molecules.

[0074] In this embodiment, a real synthetic route is obtained based on single-step reaction data. Since the dimensionality of variation in the real synthetic route is much greater than that in the single-step reaction, a larger searchable molecular space can be generated based on the real synthetic route, improving the accuracy of candidate drug molecules and thus improving the accuracy of target drug molecules. The single-step reaction data has been experimentally verified to be feasible, and the splicing rules are also feasible. Therefore, the real synthetic route obtained based on the splicing rules and single-step reaction data has high laboratory feasibility, thereby improving the syntheticability of target drug molecules.

[0075] 402. A pre-trained drug molecule generation model is used to process the input target and the actual synthetic route to output candidate drug molecules and the corresponding candidate synthetic routes.

[0076] Among them, the drug molecule generation model is a deep learning model. Using this model to generate drug molecules can leverage the excellent performance of deep learning models to improve the accuracy and efficiency of candidate drug molecules and their candidate synthetic routes.

[0077] The training process for the drug molecule generation module can be found in the description of subsequent related embodiments.

[0078] In some embodiments, the drug molecule generation model includes: a first embedding layer, a second embedding layer, an encoding layer, and a decoding layer; the step of using a pre-trained drug molecule generation model to process the input target site and the actual synthetic route to output the candidate drug molecule and its corresponding candidate synthetic route includes:

[0079] The first embedding layer is used to embed the target site to obtain a first embedding vector; the second embedding layer is used to embed the real synthetic route to obtain a second embedding vector; the first embedding vector and the second embedding vector are added together to obtain a fusion vector; the encoding layer is used to encode the fusion vector to obtain a hidden layer vector; the decoding layer is used to decode the hidden layer vector to obtain the candidate drug molecule and the candidate synthetic route.

[0080] In this embodiment, based on the processing of the above layers, candidate drug molecules and their corresponding candidate synthetic routes can be obtained simply and efficiently.

[0081] 403. Obtain the affinity between the candidate drug molecule and the target site.

[0082] For example, candidate drug molecules and target sites are input into a docking score calculation model, and the output is the docking score. Affinity is determined based on the docking score; that is, the higher the docking score, the greater the affinity.

[0083] 404. Based on the candidate synthetic route, determine the synthetic candidate drug molecules; and, among the synthetic candidate drug molecules, select the candidate drug molecule with the highest affinity as the target drug molecule.

[0084] After obtaining the candidate synthesis routes, the candidate synthesis routes can be input into the preset analysis model. The analysis model can be a binary classification model, which represents whether the routes can be synthesized or not.

[0085] For a given candidate synthetic route, if the analysis result indicates that it is synthetic, then the candidate drug molecule corresponding to that route is synthetic; conversely, if the analysis result indicates that it is not synthetic, then the candidate drug molecule corresponding to that route is not synthetic.

[0086] After obtaining synthetic drug candidates based on the candidate synthetic route, the candidate drug molecule with the highest affinity to the target can be selected as the target drug molecule. For example, if both the first and second candidate drugs are synthetic, and the first candidate drug molecule has a greater affinity to the target than the second candidate drug molecule, then the first candidate drug molecule will be selected as the target drug molecule.

[0087] In this embodiment, among the synthesizable candidate drug molecules, the candidate drug molecule with the highest affinity is selected as the target drug molecule, thereby obtaining a target drug molecule with high affinity and synthesizability, and improving the performance of the target drug molecule.

[0088] Figure 5Based on the schematic diagram of the third embodiment of this disclosure, this embodiment provides a method for training a drug molecule generation model, the training method including:

[0089] 501. A drug molecule generation model is used to process the input target sample and the pre-constructed real synthetic route to output a predicted drug molecule and its corresponding predicted synthetic route.

[0090] 502. Obtain the predicted affinity between the predicted drug molecule and the target sample.

[0091] 503. Based on the predicted synthesis route and the predicted affinity, construct the target loss function.

[0092] 504. Based on the target loss function, adjust the model parameters of the drug molecule generation model.

[0093] In this process, the target sample is predetermined. The actual synthetic route is pre-constructed, for example, based on pre-obtained single-step reaction data.

[0094] Using a similar process to the reasoning described above, during the training phase, the output of the drug molecule generation model is called the predicted drug molecule and the predicted synthetic route, and the predicted affinity between the drug molecule and the target sample is called the predicted affinity.

[0095] After obtaining the predicted synthesis route and predicted affinity, an objective function can be constructed, and then the model parameters can be adjusted based on the objective loss function, such as adjusting the model's weight coefficient w and bias coefficient b, until the preset termination condition is reached. The model that reaches the preset termination condition is taken as the final model.

[0096] The final generated drug molecule generation model can be used in the reasoning stage to obtain candidate drug molecules and their corresponding candidate synthetic routes.

[0097] In this embodiment, by obtaining the predicted synthetic route of the predicted drug molecule, constructing a target loss function based on the predicted synthetic route and predicted affinity, and adjusting the model parameters based on the target loss function, a model with high syntheticity and accuracy can be obtained. Therefore, when using this model for drug generation, the syntheticity and accuracy of the generated drug molecule can be improved.

[0098] In some embodiments, the predicted drug molecules can be divided into positive and negative samples based on the predicted synthetic route; a contrastive loss function can be constructed based on the predicted affinity of the positive samples and the predicted affinity of the negative samples, serving as the target loss function.

[0099] For example, after obtaining the predicted synthetic route, it can be input into an analysis model, and the output is the syntheticity result. Based on the syntheticity result, the predicted drug molecules can be divided into positive and negative samples. For instance, non-synthetic predicted drug molecules can be treated as negative samples, and synthetic predicted drug molecules as positive samples.

[0100] Then, a contrastive learning approach can be adopted, using the predictive affinity of positive and negative samples to construct a contrastive loss function, which can then be used as the target loss function.

[0101] In this embodiment, positive and negative samples are distinguished based on the predicted synthesis route, and then a target loss function is constructed based on the predicted affinity of positive and negative samples. The target loss function can be obtained by contrastive learning, thereby improving model performance.

[0102] In some embodiments, the drug molecule generation model includes: a first embedding layer, a second embedding layer, an encoding layer, and a decoding layer; the process of using the drug molecule generation model to process the input target sample and a pre-constructed real synthetic route to output a predicted drug molecule and its corresponding predicted synthetic route includes:

[0103] The first embedding layer is used to embed the target sample to obtain a first embedding vector; the second embedding layer is used to embed the real synthesis route to obtain a second embedding vector; the first embedding vector and the second embedding vector are added together to obtain a fusion vector; the encoding layer is used to encode the fusion vector to obtain a hidden layer vector; the decoding layer is used to decode the hidden layer vector to obtain the predicted drug molecule and the predicted synthesis route.

[0104] In this embodiment, based on the processing of the above layers, the predicted drug molecule and its corresponding predicted synthesis route can be obtained simply and efficiently.

[0105] Figure 6 This is a schematic diagram according to the fourth embodiment of the present disclosure. This embodiment provides a drug molecule generation apparatus 600, which includes: a generation module 601, an acquisition module 602, and a determination module 603.

[0106] The generation module 601 is used to generate candidate drug molecules and their corresponding candidate synthetic routes based on the target site and a pre-constructed real synthetic route; the acquisition module 602 is used to determine the affinity between the candidate drug molecules and the target site; and the determination module 603 is used to determine the target drug molecule among the candidate drug molecules based on the candidate synthetic route and the affinity.

[0107] In this embodiment, by obtaining candidate synthetic routes for candidate drug molecules, the synthetic feasibility of candidate drug molecules can be analyzed based on the candidate synthetic routes, thereby improving the synthetic feasibility of target drug molecules. In addition, generating candidate drug molecules based on real synthetic routes can improve the accuracy of candidate drug molecules, and thus improve the accuracy of target drug molecules determined based on candidate drug molecules.

[0108] In some embodiments, the determining module 603 is further configured to:

[0109] Based on the aforementioned candidate synthetic routes, synthetic candidate drug molecules were identified;

[0110] Among the synthetic candidate drug molecules, the candidate drug molecule with the highest affinity is selected as the target drug molecule.

[0111] In this embodiment, among the synthesizable candidate drug molecules, the candidate drug molecule with the highest affinity is selected as the target drug molecule, thereby obtaining a target drug molecule with high affinity and synthesizability, and improving the performance of the target drug molecule.

[0112] In some embodiments, the device 600 further includes:

[0113] A construction module is used to construct the actual synthetic route based on pre-acquired single-step reaction data.

[0114] In this embodiment, a real synthetic route is obtained based on single-step reaction data. Since the dimensionality of variation in the real synthetic route is much greater than that in the single-step reaction, a larger searchable molecular space can be generated based on the real synthetic route, improving the accuracy of candidate drug molecules and thus improving the accuracy of target drug molecules. The single-step reaction data has been experimentally verified to be feasible, and the splicing rules are also feasible. Therefore, the real synthetic route obtained based on the splicing rules and single-step reaction data has high laboratory feasibility, thereby improving the syntheticability of target drug molecules.

[0115] In some embodiments, the generation module 601 is further configured to:

[0116] A pre-trained drug molecule generation model is used to process the input target and the actual synthetic route to output the candidate drug molecule and the candidate synthetic route.

[0117] Among them, the drug molecule generation model is a deep learning model. Using this model to generate drug molecules can leverage the excellent performance of deep learning models to improve the accuracy and efficiency of candidate drug molecules and their candidate synthetic routes.

[0118] In some embodiments, the drug molecule generation model includes: a first embedding layer, a second embedding layer, an encoding layer, and a decoding layer; the generation module 601 is further used for:

[0119] The first embedding layer is used to embed the input target point to obtain a first embedding vector; the second embedding layer is used to embed the real synthetic route to obtain a second embedding vector; the first embedding vector and the second embedding vector are added together to obtain a fusion vector; the encoding layer is used to encode the fusion vector to obtain a hidden layer vector; the decoding layer is used to decode the hidden layer vector to output the candidate drug molecule and the candidate synthetic route.

[0120] In this embodiment, based on the processing of the above layers, candidate drug molecules and their corresponding candidate synthetic routes can be obtained simply and efficiently.

[0121] Figure 7 This is a schematic diagram according to the fifth embodiment of the present disclosure. This embodiment provides a drug molecule generation model training device, the device 700 including: a generation module 701, an acquisition module 702, a construction module 703, and an adjustment module 704.

[0122] The generation module 701 is used to process the input target sample and the pre-constructed real synthetic route using a drug molecule generation model to output a predicted drug molecule and its corresponding predicted synthetic route; the acquisition module 702 is used to acquire the predicted affinity between the predicted drug molecule and the target sample; the construction module 703 is used to construct a target loss function based on the predicted synthetic route and the predicted affinity; and the adjustment module 704 is used to adjust the model parameters of the drug molecule generation model based on the target loss function.

[0123] In this embodiment, by obtaining the predicted synthetic route of the predicted drug molecule, constructing a target loss function based on the predicted synthetic route and predicted affinity, and adjusting the model parameters based on the target loss function, a model with high syntheticity and accuracy can be obtained. Therefore, when using this model for drug generation, the syntheticity and accuracy of the generated drug molecule can be improved.

[0124] In some embodiments, the building module 703 is further configured to:

[0125] Based on the predicted synthesis route, the predicted drug molecules are divided into positive samples and negative samples; based on the predicted affinity corresponding to the positive samples and the predicted affinity corresponding to the negative samples, a contrastive loss function is constructed as the target loss function.

[0126] In this embodiment, positive and negative samples are distinguished based on the predicted synthesis route, and then a target loss function is constructed based on the predicted affinity of positive and negative samples. The target loss function can be obtained by contrastive learning, thereby improving model performance.

[0127] In some embodiments, the drug molecule generation model includes: a first embedding layer, a second embedding layer, an encoding layer, and a decoding layer; the generation module 701 is further used for:

[0128] The first embedding layer is used to embed the input target sample to obtain a first embedding vector; the second embedding layer is used to embed the real synthesis route to obtain a second embedding vector; the first embedding vector and the second embedding vector are added together to obtain a fusion vector; the encoding layer is used to encode the fusion vector to obtain a hidden layer vector; the decoding layer is used to decode the hidden layer vector to output the predicted drug molecule and the predicted synthesis route.

[0129] In this embodiment, based on the processing of the above layers, the predicted drug molecule and its corresponding predicted synthesis route can be obtained simply and efficiently.

[0130] It is understood that the same or similar content in different embodiments of this disclosure can be referred to each other.

[0131] It is understood that the terms "first" and "second" in the embodiments of this disclosure are only used for distinction and do not indicate the degree of importance or the order of events.

[0132] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0133] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0134] Figure 8 A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present disclosure is shown. Electronic device 800 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 800 may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0135] like Figure 8 As shown, the electronic device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. The RAM 803 may also store various programs and data required for the operation of the electronic device 800. The computing unit 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0136] Multiple components in electronic device 800 are connected to I / O interface 805, including: input unit 806, such as keyboard, mouse, etc.; output unit 807, such as various types of displays, speakers, etc.; storage unit 808, such as disk, optical disk, etc.; and communication unit 809, such as network card, modem, wireless transceiver, etc. Communication unit 809 allows electronic device 800 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0137] The computing unit 801 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as drug molecule generation methods or drug molecule generation model training methods. For example, in some embodiments, the drug molecule generation method or drug molecule generation model training method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 800 via ROM 802 and / or communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the drug molecule generation method or drug molecule generation model training method described above can be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to perform a drug molecule generation method or a drug molecule generation model training method by any other suitable means (e.g., by means of firmware).

[0138] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0139] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0140] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0141] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0142] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0143] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem, addressing the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0144] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0145] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for generating drug molecules, comprising: generating a candidate drug molecule and a corresponding candidate synthesis route of the candidate drug molecule based on a target target point and a pre-constructed real synthesis route, including: using a pre-trained drug molecule generation model to process the input target target point and the real synthesis route to output the candidate drug molecule and the candidate synthesis route; obtaining an affinity between the candidate drug molecule and the target target point; determining a target drug molecule from the candidate drug molecule based on the candidate synthesis route and the affinity, including: inputting the candidate synthesis route into a pre-trained analysis model to output a synthesizability result of the corresponding candidate drug molecule, the synthesizability result including synthesizable or unsynthesizable; determining the target drug molecule from the candidate drug molecule based on the synthesizability result of the candidate drug molecule and the affinity between the candidate drug molecule and the target target point; the drug molecule generation model comprises: a first embedding layer, a second embedding layer, an encoding layer and a decoding layer; the using a pre-trained drug molecule generation model to process the input target target point and the real synthesis route to output the candidate drug molecule and the corresponding candidate synthesis route of the candidate drug molecule, including: using the first embedding layer to embed the input target target point to obtain a first embedding vector; using the second embedding layer to embed the real synthesis route to obtain a second embedding vector; adding the first embedding vector and the second embedding vector to obtain a fusion vector; using the encoding layer to encode the fusion vector to obtain a hidden layer vector; using the decoding layer to decode the hidden layer vector to output the candidate drug molecule and the candidate synthesis route.

2. The method of claim 1, wherein, the determining a target drug molecule from the candidate drug molecule based on the candidate synthesis route and the affinity, including: determining a synthesizable candidate drug molecule based on the candidate synthesis route; in the synthesizable candidate drug molecule, the candidate drug molecule with the maximum affinity is determined as the target drug molecule.

3. The method of claim 1, further comprising: constructing the real synthesis route based on pre-acquired single-step reaction data.

4. A method for training a drug molecule generation model, comprising: using a drug molecule generation model to process input target target point samples and pre-constructed real synthesis routes to output predicted drug molecules and corresponding predicted synthesis routes of the predicted drug molecules; obtaining a predicted affinity between the predicted drug molecules and the target target point samples; constructing a target loss function based on the predicted synthesis routes and the predicted affinity, including: inputting the predicted synthesis routes into an analysis model to output a synthesizability result; using a contrast learning method to construct a contrast loss function as the target loss function by using the predicted affinity of positive and negative samples, wherein the predicted drug molecule with the unsynthesizable synthesizability result is used as a negative sample, and the predicted drug molecule with the synthesizable synthesizability result is used as a positive sample; adjusting model parameters of the drug molecule generation model based on the target loss function. The drug molecule generation model comprises a first embedding layer, a second embedding layer, an encoding layer and a decoding layer. The drug molecule generation model is used for processing input target sample and pre-constructed real synthesis route to output predicted drug molecule and corresponding predicted synthesis route, and the method comprises the following steps of: The first embedding layer is used for embedding the input target sample to obtain a first embedding vector; The second embedding layer is used for embedding the real synthesis route to obtain a second embedding vector; The first embedding vector and the second embedding vector are added to obtain a fusion vector; The encoding layer is used for encoding the fusion vector to obtain a hidden layer vector; The decoding layer is used for decoding the hidden layer vector to output the predicted drug molecule and the predicted synthesis route.

5. A drug molecule generation device, comprising: A generation module is configured to generate a candidate drug molecule and a corresponding candidate synthesis route based on a target target and a pre-constructed real synthesis route, comprising: using a pre-trained drug molecule generation model to process the input target and the real synthesis route to output the candidate drug molecule and the candidate synthesis route; An acquisition module is configured to determine the affinity between the candidate drug molecule and the target target; A determination module is configured to determine a target drug molecule in the candidate drug molecule based on the candidate synthesis route and the affinity, comprising: inputting the candidate synthesis route into a pre-trained analysis model to output a corresponding candidate drug molecule synthesis result, the synthesis result comprises synthesis or non-synthesis; based on the synthesis result of the candidate drug molecule and the affinity between the target target, determine the target drug molecule in the candidate drug molecule; The drug molecule generation model comprises a first embedding layer, a second embedding layer, an encoding layer and a decoding layer; The generation module is further configured to: The first embedding layer is used for embedding the input target to obtain a first embedding vector; The second embedding layer is used for embedding the real synthesis route to obtain a second embedding vector; The first embedding vector and the second embedding vector are added to obtain a fusion vector; The encoding layer is used for encoding the fusion vector to obtain a hidden layer vector; The decoding layer is used for decoding the hidden layer vector to output the candidate drug molecule and the candidate synthesis route.

6. The apparatus of claim 5, wherein, The determination module is further configured to: Determine the synthesisable candidate drug molecule based on the candidate synthesis route; In the synthesisable candidate drug molecule, the candidate drug molecule with the largest affinity is taken as the target drug molecule.

7. The device of claim 5, further comprising: A construction module is configured to construct the real synthesis route based on pre-acquired single-step reaction data.

8. A drug molecule generation model training device, comprising: The generating module is configured to process the input target sample and the pre-constructed real synthetic route by using a drug molecule generation model to output a predicted drug molecule and a corresponding predicted synthetic route; The obtaining module is configured to obtain a predicted affinity between the predicted drug molecule and the target sample; The constructing module is configured to construct a target loss function based on the predicted synthetic route and the predicted affinity, including: inputting the predicted synthetic route into an analysis model to output a synthesizability result; taking the predicted drug molecule with the synthesizability result being unsynthesizable as a negative sample, taking the predicted drug molecule with the synthesizability result being synthesizable as a positive sample, constructing a contrast loss function by using the predicted affinity of the positive and negative samples in a contrast learning manner as the target loss function; The adjusting module is configured to adjust model parameters of the drug molecule generation model based on the target loss function; The drug molecule generation model includes a first embedding layer, a second embedding layer, an encoding layer, and a decoding layer; The generating module is further configured to: The first embedding layer is used to embed the input target sample to obtain a first embedding vector; The second embedding layer is used to embed the real synthetic route to obtain a second embedding vector; The first embedding vector and the second embedding vector are added to obtain a fusion vector; The encoding layer is used to encode the fusion vector to obtain a hidden layer vector; The decoding layer is used to decode the hidden layer vector to output the predicted drug molecule and the predicted synthetic route.

9. An electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-4.

10. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method of any one of claims 1-4.

11. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-4.

Images