Intelligent medicine research and development method and platform, computer equipment and storage medium

Abstract

The invention provides an intelligent drug research and development method and platform, computer equipment and a storage medium. The method comprises the following steps: constructing a database to store drug data information; configuring codes used for database query and different kinds of function analysis on the medicine data information; constructing an intention recognition model and carrying out question and answer training based on user intention recognition; constructing a plurality of APIs (Application Program Interface), and enabling the APIs to correspond to different types of function analysis; obtaining research and development demand information input by the user, and performing intention recognition through the intention recognition model to obtain a target intention of the user; and calling a corresponding API interface according to the target intention to obtain drug data information or a data processing result required by the user, and feeding back the drug data information or the data processing result to the user through a front-end interface. According to the scheme, accurate research and drug design and analysis of drug discovery items can be realized, and non-professionals can also participate in research and development work, so that improvement of research and development efficiency is facilitated, and research and development time and human resource cost are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of drug research and development, and particularly to an intelligent drug research and development method, platform, computer device and storage medium. Background Art

[0002] In the work of medicinal chemists, drug research and development is a crucial part, including drug data research, design, analysis, and final drug development. At present, the drug research and development process mainly relies on manual operations, including retrieval, classification, and analysis. However, these processes not only consume a large amount of time and human resources, but also are prone to missing key information, affecting the entire research and development process. In addition, the field of drug research and development is highly professional and has high requirements for workers. It is difficult for non-professional personnel to participate in the research and development work, further increasing the research and development time and human resources. Therefore, there is an urgent need for a method that can achieve rapid and accurate drug research and development and enable non-professional personnel to participate in the research and development work. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent drug research and development method, platform, computer device and storage medium, which can achieve accurate research, drug design and analysis of drug discovery projects, avoid missing key information in the research and development process, and enable non-professional personnel to participate in the research and development work, which is beneficial to improving the research and development efficiency and reducing the research and development time and human resource costs.

[0004] The technical solution provided by the present invention is as follows:

[0005] The present invention provides an intelligent drug research and development method, including the steps of:

[0006] Constructing a database to store drug data information;

[0007] Configuring codes for querying the database and performing different types of functional analysis on the drug data information;

[0008] Constructing an intent recognition model and performing question-and-answer training based on user intent recognition;

[0009] Constructing a number of API interfaces and corresponding each of the API interfaces to different types of functional analysis;

[0010] Obtaining the research and development requirement information input by the user, and performing intent recognition through the intent recognition model to obtain the user's target intent;

[0011] According to the target intent, calling the corresponding API interface to obtain the drug data information or data processing result required by the user, and feeding it back to the user through the front-end interface.

[0012] In some embodiments, the database includes a relational database and a vector database;

[0013] The drug data information stored in the relational database includes compound-target relationship data, disease-target relationship data, and compound-disease relationship data;

[0014] The vector database is used to store text data related to drug discovery and clinical information.

[0015] In some embodiments, the API interface includes:

[0016] A database query interface for querying the drug data information;

[0017] An SAR analysis interface for performing structure-activity relationship analysis on compounds;

[0018] A property prediction interface for predicting and analyzing the properties of compounds;

[0019] A drug design interface for analyzing the improvement of compounds and generating new compounds;

[0020] An infringement analysis interface for performing infringement analysis on compounds.

[0021] In some embodiments, obtaining the R & D requirement information input by the user and performing intent recognition through the intent recognition model to obtain the user's target intent specifically includes:

[0022] Obtaining the R & D requirement information input by the user through the human-computer interaction interface, and performing semantic parsing on the R & D requirement information through the intent recognition model to obtain the user requirement information;

[0023] Obtaining the user's target intent based on the user requirement information;

[0024] Generating a call instruction for calling the corresponding API interface according to the target intent.

[0025] In some embodiments, constructing the intent recognition model specifically includes:

[0026] Constructing an initial intent recognition model;

[0027] Collecting the drug data information and labeling the drug terms as the training set of the initial intent recognition model;

[0028] Training the initial intent recognition model in the way of structured dialogue to obtain the intent recognition model.

[0029] In some embodiments, feeding back to the user through the front-end interface specifically includes:

[0030] Generate any one or more of a drug research report, an SAR analysis report, a property prediction report, a compound design report, and an infringement analysis report according to the analysis result, and return it to the user through the human-computer interaction interface.

[0031] In some embodiments, it further includes:

[0032] Pre-store several analysis report templates and correspond them to each analysis type;

[0033] Select the analysis report template according to the target intention, and organize the analysis report template according to the analysis result to obtain the target analysis report.

[0034] In a second aspect, the present application provides an intelligent drug R & D platform, including: a database, a human-computer interaction interface, and a processor, and the processor executes a computer program to implement the steps of the intelligent drug R & D method described in the first aspect.

[0035] In a third aspect, the present application provides a computer device, including a memory, a processor, and a computer program stored on the memory, and the processor executes the computer program to implement the steps of the intelligent drug R & D method described in the first aspect.

[0036] In a fourth aspect, the present application provides a computer storage medium, on which a computer program or instruction is stored, and when the computer program or instruction is executed by a processor, the steps of the intelligent drug R & D method described in the first aspect are implemented.

[0037] According to an intelligent drug R & D method, platform, computer device, and storage medium provided by the present invention, accurate R & D of drugs can be achieved, key information can be avoided from being omitted during R & D, and non-professionals can also participate in the R & D work, which is beneficial to improving the R & D efficiency and reducing the R & D time and human resource costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The above characteristics, technical features, advantages, and their implementation manners of the present solution will be further described below in a clear and easy-to-understand manner in combination with the drawings in the preferred embodiments.

[0039] Figure 1 is the overall flow schematic diagram of an embodiment of the present invention;

[0040] Figure 2 is the flow schematic diagram of an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will describe the specific embodiments of the present invention with reference to the accompanying drawings. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings and other embodiments can be obtained.

[0042] To make the drawings concise, only the parts related to the present invention are schematically shown in each drawing, and they do not represent the actual structure of the product. In addition, to make the drawings concise and easy to understand, in some drawings, for components with the same structure or function, only one of them is schematically shown, or only one of them is labeled. In this article, "one" not only means "only this one", but also can mean "more than one" situation.

[0043] In the work of medicinal chemists, drug research and development is a crucial part, including drug data research, design, analysis, and final drug development. Currently, the drug research and development process mainly relies on manual operations, including retrieval, classification, and analysis. However, these processes not only consume a large amount of time and human resources, but also are prone to missing key information, affecting the entire research and development process. In addition, the field of drug research and development is highly professional and has high requirements for workers. It is difficult for non-professionals to participate in the research and development work, further increasing the research and development time and human resources. Therefore, there is an urgent need for a method that can achieve rapid and accurate drug research and development and enable professionals to also participate in the research and development work.

[0044] This solution realizes an intelligent drug research and development platform through the following steps: constructing a database to classify and store drug data information; implementing various functions through code, including database query, data analysis, SAR analysis, drug design, compound property prediction, patent infringement analysis, etc.; fine-tuning a large model with biomedical data and training an intent recognition model through questions and answers; constructing several API interfaces and corresponding each API interface to different functions (such as literature research, data analysis, SAR analysis, compound property prediction, patent infringement analysis, etc.). This platform can obtain the required information input by the user, understand the user's needs through the intent recognition model, call the corresponding API interface according to the target intent, obtain the information or data processing result required by the user, and feedback the result to the user through the front-end interface. This platform combines a large model fine-tuned with professional data and an intent recognition model, and seamlessly docks with traditional medicinal chemistry tools, enabling medicinal chemists to conveniently carry out professional work such as patent data analysis, SAR analysis, patent infringement analysis, and drug design, thereby improving work efficiency and outputting professional results. The following will describe this solution in detail with reference to the accompanying drawings:

[0045] In one embodiment, refer to the accompanying drawings of the specificationFigure 1 , the present invention provides an intelligent drug R & D method, including the steps of:

[0046] S100. Construct a database to store drug data information;

[0047] S200. Configure codes for querying the database and performing different types of functional analysis on the drug data information;

[0048] S300. Construct an intention recognition model and perform question - answering training based on user intention recognition;

[0049] S400. Construct a number of API interfaces and correspond each API interface to different types of functional analysis respectively;

[0050] S500. Obtain the R & D requirement information input by the user, perform intention recognition through the intention recognition model, and obtain the user's target intention;

[0051] S600. Call the corresponding API interface according to the target intention to obtain the drug data information or data processing result required by the user, and feedback it to the user through the front - end interface.

[0052] This solution can collect a large amount of compound - target relationship data (activity data), the corresponding relationship between diseases and targets, and the corresponding relationship between compounds and diseases from sources such as patents, literature, and open - source databases, and store them in the database. The database can select MySQL database, other relational database management systems (such as PostgreSQL, Oracle), etc. Considering that the collected drug data information may contain both structured data and a lot of unstructured data from literature, news, etc. that need to be accurately searched but are difficult to structure. This solution's database includes a relational database and a vector database; the drug data information stored in the relational database includes compound - target relationship data, disease - target relationship data, and compound - disease relationship data. The vector database can select databases such as FAISS, Annoy, etc., and the vector database is used to store text data related to drug discovery and clinical information. This solution adopts a hybrid architecture combining MySQL database and vector database, uses the MySQL database to store and manage structured data such as compound - target relationship data and disease - target relationships, and uses the vector database to store unstructured data such as clinical information and literature, so as to support efficient, comprehensive, and accurate information retrieval. And by integrating information from multiple data sources in the field of medicinal chemistry, as well as various tools commonly used by medicinal chemists, it can provide comprehensive and accurate analysis results and reduce information omission.

[0053] This solution uses large language models for drug research and development. Currently, large language models have achieved remarkable success in many general scenarios. However, the application scenarios of medicinal chemists are highly specialized, which makes general large models unable to meet their needs. For example, ordinary large models cannot help medicinal chemists professionally analyze the patent layouts of pharmaceutical companies such as Pfizer, nor can they perform structure-activity relationship (SAR) analysis on compounds of a certain target, or conduct MARKUSH infringement analysis on compounds of interest to medicinal chemists, etc. Medicinal chemists usually rely on traditional tools to complete these complex tasks, which require specific software and skills. Since the skills required for each task are different, these activities are usually carried out separately, resulting in low efficiency. This application combines the advantages of large models with traditional medicinal chemistry tools, improves the analysis quality through the semantic parsing ability and large-scale data processing ability of large models, enabling medicinal chemists to conveniently perform professional tasks such as patent data analysis, SAR analysis, patent infringement analysis, and drug design, thereby improving work efficiency and outputting professional results. This application does not limit the specific architecture of the underlying fine-tuned model. For example, the underlying large model of this solution can be obtained by fine-tuning the open-source large model LLaMA38B or LLaMA370B. Another example is that other open-source or commercial large models (such as GPT-4, BERT, etc.) can also be used for fine-tuning, as long as they can understand the professional terms in drug research and development.

[0054] In a specific implementation, an intent recognition model is constructed, which specifically includes: constructing an initial intent recognition model; collecting drug data information and marking drug terms as the training set of the initial intent recognition model; training the initial intent recognition model in a structured dialogue manner to obtain the intent recognition model.

[0055] Based on the open-source model, this solution is fine-tuned with biomedical data to enhance the model's understanding ability in the patent field. Specifically, it includes: collecting and annotating a large amount of corpus data in the patent field; training and optimizing the model, and regularly updating and maintaining the model to ensure its understanding ability and accuracy. During training, drug data information is collected and drug terms are marked as the training set of the initial intent recognition model, and the initial intent recognition model is trained in a structured dialogue manner. The accuracy of the training results can be judged according to the marked drug terms.

[0056] This solution supports a variety of analysis and research tasks, including querying information related to diseases, targets, compounds, etc., patent data analysis, SAR analysis, physicochemical property calculation, ADMET property calculation, structure improvement, and MARKUSH infringement analysis. To ensure the accuracy of each analysis task and enable independent execution, this solution constructs several API interfaces and corresponds each API interface to different types of functional analysis. Specifically, the API interfaces include: a database query interface for querying the drug data information, including information related to diseases, targets, compounds, etc.; an SAR analysis interface for performing structure-activity relationship analysis on compounds; a property prediction interface for predicting and analyzing the properties of compounds; a drug design interface for improving compounds and analyzing the generation of new compounds; and an infringement analysis interface for analyzing whether a compound infringes existing MARKUSH patents. Structure-Activity Relationship (SAR) refers to the relationship between the chemical structure of a drug or other physiologically active substance and its physiological activity, which is an important research content in medicinal chemistry and related fields. SAR analysis can conduct a detailed analysis of the relationship between the structure of a compound and its biological activity. Property analysis includes physicochemical property analysis and ADMET property analysis. Physicochemical analysis is used to calculate the physicochemical properties of compounds, such as solubility, stability, etc.; ADMET properties refer to the characteristics of a drug's absorption, distribution, metabolism, excretion, and toxicity in the body, which are important indicators for evaluating the drug-likeness and safety of compounds in drug research and development. ADMET property analysis is used to analyze the absorption, distribution, metabolism, excretion, and toxicity of compounds. Derivative analysis is used to improve compounds and analyze the generation of new compounds. Infringement analysis is used to analyze whether a compound infringes existing MARKUSH patents. The API interfaces for SAR analysis, physicochemical property calculation, ADMET property calculation, structure improvement, and MARKUSH infringement analysis can be implemented in different programming languages and frameworks, or existing commercial API services can be used to replace the self-developed interfaces. The API interfaces are designed flexibly and can be efficiently called according to user needs. Moreover, an integrated API interface platform that integrates SAR analysis, physicochemical property calculation, ADMET property calculation, structure improvement, new compound generation, and MARKUSH infringement analysis supports one-stop processing of a variety of complex tasks.

[0057] This application uses natural language interaction. Users can input questions in natural language without having to master complex professional skills and tools, greatly reducing the usage threshold. Users input questions related to drug research and development in natural language, such as asking what the targets under research of Pfizer are this year, etc. In a specific implementation manner, in order to enable the system to accurately determine the user's research and development needs, the research and development requirement information input by the user is obtained and intention recognition is performed through an intention recognition model to obtain the target intention related to drug terms, specifically including:

[0058] The research and development requirement information input by the user is obtained through the human-computer interaction interface, and semantic analysis is performed on the research and development requirement information through the intention recognition model to obtain the user requirement information; the user's target intention is obtained through the user requirement information; a call instruction for calling the corresponding API interface is generated according to the target intention. Thus, after obtaining the research and development requirement information input by the user, intention recognition can be performed through the intention recognition model to obtain the target intention of the user's requirement, and it is converted into an instruction that can be understood by the system, and then the corresponding API interface is called according to the target intention to obtain the drug data information or data processing result of the user's requirement. In addition to using a trained natural language processing model to identify the user's intention, a rule-based system, a machine learning model or a hybrid method can also be used for intention recognition and function call, and this application does not make any restrictions. Through multi-level semantic analysis and intention recognition, the system can accurately distinguish the intention input by the user, intelligently call different function modules or API interfaces, and achieve fast and accurate responses.

[0059] The design and interaction method of the human-computer interaction interface can be implemented through different technologies (such as Web applications, mobile applications, desktop software, etc.) to ensure that users can conveniently input questions and view results.

[0060] After obtaining the drug data information or data processing result of the user's requirement, it is fed back to the user through the front-end interface, specifically including: generating any one or several of a drug research report, an SAR analysis report, a property prediction report, a compound design report, and an infringement analysis report according to the analysis result, and returning it to the user through the front-end human-computer interaction interface.

[0061] According to the analysis and processing results, this solution can automatically generate a detailed analysis report and return the results to the user through the user interface. The user can choose to view specific analysis content, such as: Patent data analysis report: including information such as patent layout analysis and the key compound structures in patents; SAR analysis results: showing detailed charts and conclusions on the relationship between compound structures and activities; Compound design report, providing strategies for compound improvement and the generation of new compounds; Patent infringement analysis report: providing an assessment of the patent infringement risks of compounds. This solution automatically generates a detailed analysis report in the form of natural language dialogue, covering various professional contents such as patent data analysis, SAR analysis, and patent infringement risk assessment, and presents it to the user through an intuitive user interface.

[0062] To improve the generality, accuracy, and speed of the analysis report, the system can pre-store several analysis report templates and correspond them to each analysis type; after obtaining the target intention, it can select the analysis report template according to the target intention and organize the analysis report template based on the analysis results to obtain the target analysis report. There can be multiple options for the generation and presentation of the analysis report. In addition to the detailed text report, charts, visualization tools, etc. can also be used to make the results more intuitive.

[0063] In a specific implementation manner, the intelligent drug R & D method provided by this application includes:

[0064] S110. Build a database to classify and store drug data information. This step involves collecting drug data information and classifying and storing this data for subsequent query and analysis.

[0065] S210. Implement multiple functions by writing code to implement functions such as database query, data analysis, SAR analysis, drug design, prediction of compound properties, and patent infringement analysis. These functions can help medicinal chemists carry out comprehensive R & D work.

[0066] S310. Fine-tune the large model using biomedical data to improve the accuracy of the model in related fields. Through the fine-tuned large model, the professional needs in the drug R & D process can be better met.

[0067] S410. Train the intention recognition model by training the model with questions and answers so that it can understand user needs. This step enables the model to accurately identify the user's intention and provide corresponding solutions.

[0068] S510. Build several API interfaces and correspond each interface to different functions (such as literature research, data analysis, SAR analysis, prediction of compound properties, patent infringement analysis, etc.). These interfaces can help different functional modules conduct effective communication and data exchange.

[0069] S610. Obtain the required information input by the user and understand the user's needs through the intent recognition model. Call the corresponding API interface according to the target intent to obtain the information or data processing results required by the user, and feedback the results to the user through the front-end interface. This step ensures that the user can quickly obtain the required drug R & D information.

[0070] By combining the intelligent advantages of the large model with traditional drug chemistry tools, the drug R & D method provided by this application has at least the following technical effects:

[0071] (1) Improve work efficiency:

[0072] Automated retrieval and analysis: The system can automatically retrieve and analyze patent data, literature data, compound and disease target relationship data, greatly shortening the time of manual operation and improving the overall work efficiency;

[0073] Fast function call: Through intent recognition technology, the system can quickly determine the user's needs and call the corresponding function module or API, reducing the user's operation steps and improving the response speed.

[0074] (2) Output professional results:

[0075] Accurate semantic parsing: The fine-tuned large model can accurately understand the user's professional needs and provide professional answers and analysis results to ensure the accuracy and professionalism of the results;

[0076] Comprehensive analysis ability: The system supports a variety of analysis tasks, including patent data analysis, SAR analysis, physicochemical property calculation, ADMET property calculation, structure improvement and MARKUSH infringement analysis, and can provide comprehensive professional solutions.

[0077] (3) Reduce learning costs:

[0078] Natural language interaction: Users can input questions through natural language without mastering complex professional skills and tools, greatly reducing the usage threshold;

[0079] Integrate multiple tools: The system integrates different drug chemistry tools on one platform, and users do not need to learn and use multiple tools separately, simplifying the operation process.

[0080] (4) Reduce information omission:

[0081] Accurate data processing: Through accurate semantic parsing and structured data processing, the system can ensure that relevant information will not be omitted and provide comprehensive and accurate analysis results;

[0082] Professional database support: The system integrates multiple data sources to ensure the comprehensiveness and accuracy of data, reducing the risk of information omission.

[0083] (5) Convenience and ease of use:

[0084] One-stop service: Users can complete multiple professional analysis tasks through a single platform. The operation is simple and convenient, enhancing the user experience.

[0085] Flexible function call: The system automatically calls the corresponding API according to user needs, providing highly flexible and convenient services.

[0086] (6) Comprehensiveness:

[0087] Multi-dimensional problem-solving: The system can answer a variety of questions that pharmaceutical chemists are concerned about, covering multiple aspects such as disease targets, drug R & D, SAR analysis, prediction and evaluation, patent analysis, and market trends.

[0088] Comprehensive data coverage: The system integrates multiple data sources to ensure that the information and analysis results provided are comprehensive and accurate.

[0089] In one embodiment, the present application provides an intelligent drug R & D platform, including: a database, a human-computer interaction interface, and a processor. The processor executes a computer program to implement the steps of the intelligent drug R & D method in the above embodiment.

[0090] In one embodiment, the present application further provides a computer device, including a memory, a processor, and a computer program stored on the memory. The processor executes the computer program to implement the steps of the intelligent drug R & D in the above embodiment.

[0091] In one embodiment, the present application further provides a computer storage medium, on which a computer program or instruction is stored. When the computer program or instruction is executed by the processor, the steps of the intelligent drug R & D in the above embodiment are implemented.

[0092] In one embodiment, the present application further provides a computer program product, including a computer program or instruction. When the computer program or instruction is executed by the processor, the steps of the intelligent drug R & D in the above embodiment are implemented.

[0093] In the foregoing embodiments, the processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0094] The memory may be an internal storage unit of the simulation system, such as: the hard disk or memory of the intelligent device. The memory may also be an external storage device of the intelligent device, such as: a plug-in hard disk equipped on the intelligent device, a SmartMedia Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory is used to store computer programs and other programs and data required for the verification method of the FPGA. The memory may also be used to temporarily store data that has been output or is to be output.

[0095] A communication bus is a circuit that connects the described elements and enables transmission between these elements. For example, a central processing unit receives commands from other elements via the communication bus, decrypts the received commands, and performs calculations or data processing based on the decrypted commands. The memory may include program modules, such as a kernel, middleware, an Application Programming Interface (API), and applications. These program modules can be composed of software, firmware, hardware, or at least two of them. The input / output interface forwards commands or data input by the user through the input / output interface (such as sensors, keyboards, touchscreens). The communication interface connects the rotational speed measurement device of this artificial heart to other network devices, user devices, and networks. For example, the communication interface can be connected to a network through a wired or wireless connection to connect to other external network devices or user devices. Wireless communication may include at least one of the following sparse matrix solution methods, computer devices, storage media, and program products: Wi-Fi (Wireless Fidelity), Bluetooth (BT), Near Field Communication (NFC), Global Positioning System (GPS), and cellular communication, etc. Wired communication may include at least one of the following sparse matrix solution methods, computer devices, storage media, and program products: Universal Serial Bus (USB), High-Definition Multimedia Interface (HDMI), Asynchronous Transmission Standard Interface (RS-232), etc. The network can be a telecommunications network and a communication network. The communication network can be a computer network, the Internet, the Internet of Things, or a telephone network. The verification device of the FPGA can be connected to the network through the communication interface, and the protocol used for communication between the rotational speed measurement device of the artificial heart and other network devices can be supported by at least one of the application, Application Programming Interface (API), middleware, kernel, and communication interface.

[0096] The intelligent drug research and development of this application can be implemented with program codes executable by a computing device. Thus, they can be stored in a storage device for execution by the computing device, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. In this way, the present invention is not limited to any specific combination of hardware and software.

[0097] It should be noted that the above-mentioned embodiments can be freely combined as needed. The above is only the preferred embodiment of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can still be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A smart drug development method, characterized in that: Includes steps: Build a database to store drug data information; Codes configured for querying the database and performing different types of functional analysis on the drug data information; Build an intent recognition model and conduct question-answering training based on user intent recognition; Constructing several API interfaces, and corresponding each of the API interfaces to different types of functional analysis; Acquire the R&D demand information input by the user, and perform intent recognition through the intent recognition model to obtain the user's target intent; According to the target intention, the corresponding API interface is called to obtain the drug data information or data processing results required by the user, and feedback is given to the user through the front-end interface.

2. The intelligent drug development method according to claim 1, characterized in that: The database includes a relational database and a vector database; The drug data information stored in the relational database includes compound-target relationship data, disease-target relationship data and compound-disease relationship data; The vector database is used to store text data related to drug discovery and clinical information.

3. The intelligent drug development method according to claim 1, characterized in that: The API interface includes: A database query interface, used to query the drug data information; SAR analysis interface, used for structure-activity relationship analysis of compounds; Property prediction interface, used to predict and analyze the properties of compounds; Drug design interface for analysis of compound improvements and generation of new compounds; Infringement analysis interface, used to perform infringement analysis on compounds.

4. The intelligent drug development method according to claim 1, characterized in that: The obtaining of the R&D demand information input by the user and performing intent recognition through the intent recognition model to obtain the user's target intent specifically includes: Acquire the R&D demand information input by the user through the human-computer interaction interface, and perform semantic analysis on the R&D demand information through the intention recognition model to obtain user demand information; Obtaining the user's target intention through the user demand information; A calling instruction for calling the corresponding API interface is generated according to the target intent.

5. The intelligent drug development method according to claim 1, characterized in that: The construction of the intention recognition model specifically includes: Build an initial intent recognition model; Collecting the drug data information and marking drug terms as a training set for the initial intent recognition model; The initial intent recognition model is trained in a structured dialogue manner to obtain the intent recognition model.

6. The intelligent drug development method according to claim 1, characterized in that: The feedback to the user through the front-end interface specifically includes: Any one or more of a drug research report, a SAR analysis report, a property prediction report, a compound design report, and an infringement analysis report are generated based on the analysis results and returned to the user through a human-computer interaction interface.

7. The intelligent drug development method according to claim 6, characterized in that: Also includes: Pre-store several analysis report templates and correspond to each analysis type; The analysis report template is selected according to the target intention, and the analysis report template is sorted according to the analysis result to obtain a target analysis report.

8. An intelligent drug development platform, characterized in that: include: A database, a human-computer interaction interface and a processor, wherein the processor executes a computer program to implement the steps of the intelligent drug development method according to any one of claims 1 to 7.

9. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the smart drug development method according to any one of claims 1 to 7.

10. A computer storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the intelligent drug development method according to any one of claims 1 to 7 are implemented.

Images