Methods, devices and terminal equipment for analyzing the mechanisms of traditional Chinese medicine
By integrating multiple large-scale artificial intelligence models to analyze the mechanisms of traditional Chinese medicine, the biomolecular regulatory mechanisms of traditional Chinese medicine have been revealed, solving the problems that existing technologies cannot analyze and achieving a more comprehensive analysis of biomolecular regulatory mechanisms.
Patent Information
- Application Number
- CN202510637139.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-16
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies are insufficient to effectively analyze the biomolecular regulatory mechanisms in traditional Chinese medicine, and machine learning methods such as support vector machines and random forests cannot meet the analytical needs.
It integrates a strong logical reasoning AI language model, DNA sequence model, RNA sequence model, cis-regulatory element model, and protein structure prediction model, and analyzes biomolecular information related to the mechanisms of traditional Chinese medicine through a cross-modal thinking chain causal relationship discovery framework.
This study reveals the regulatory patterns of biomolecular expression in the mechanisms of traditional Chinese medicine, provides a more comprehensive analysis of biomolecular regulatory mechanisms, and supports the modernization of traditional Chinese medicine.
Smart Images

Figure CN123115080A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of traditional Chinese medicine mechanism analysis technology, and in particular relates to a method, device and terminal equipment for traditional Chinese medicine mechanism analysis. Background Technology
[0002] In recent years, artificial intelligence technology has achieved significant breakthroughs globally. The deep integration of traditional Chinese medicine (TCM) and artificial intelligence has provided new opportunities for the modernization of TCM. Over the years, a large number of studies have utilized big data and deep learning methods to analyze the mechanisms of TCM and have made progress.
[0003] Due to the high complexity of biomolecular regulatory mechanisms in traditional Chinese medicine (TCM), elucidating the scientific connotation of TCM theory and its diagnostic and treatment methods still faces enormous challenges. Machine learning methods used in related technologies, such as support vector machines and random forests, cannot analyze the biomolecular regulatory mechanisms in TCM. Therefore, there is an urgent need for an analytical method to resolve the regulatory mechanisms contained in biomolecules. Summary of the Invention
[0004] This application provides a method, apparatus, and terminal device for analyzing the mechanisms of traditional Chinese medicine, which can analyze the regulatory mechanisms contained in biomolecules in the mechanisms of traditional Chinese medicine.
[0005] In a first aspect, embodiments of this application provide a method for analyzing the mechanisms of traditional Chinese medicine, including:
[0006] Obtain the first interactive information input by the user, which is related to the first symptom and the first Chinese medicinal herb.
[0007] Based on the first interactive information, detect different types of biomolecular information related to the first disease in the first Chinese medicinal material;
[0008] Predict the regulatory mechanism of the first traditional Chinese medicine material on the first disease based on different types of biomolecular information;
[0009] The first analytical result is generated based on the regulatory mechanism;
[0010] The first analysis result is displayed to the user.
[0011] In this embodiment, relevant information input by the user regarding a specific symptom (first symptom) and a specific traditional Chinese medicine (TCM) herb (first TCM herb) is obtained. The various types of biomolecular information contained in the TCM herb are then analyzed. This multi-faceted biomolecular information allows for analysis of the input interaction information from multiple perspectives, leading to a more comprehensive and accurate understanding of the regulatory mechanism of the biomolecules in the extracted TCM herb on the first symptom, resulting in a more valuable analytical result, i.e., the first analytical result. By utilizing various types of biomolecular information, a more comprehensive understanding of the regulatory mechanism of the biomolecules in the first TCM herb on the first symptom can be achieved. Therefore, the above method can analyze the regulatory mechanisms inherent in biomolecules within the mechanisms of traditional Chinese medicine.
[0012] In one possible implementation of the first aspect, the first interactive information includes a first instruction;
[0013] Obtain the first interaction information input by the user, including:
[0014] Obtain information on the first symptom and the first Chinese medicinal herb according to the first instruction.
[0015] In one possible implementation of the first aspect, the first interactive information includes a first logical basis, which is used to provide a solution to the problem and related information;
[0016] Detecting different types of biomolecular information related to the primary disease in the primary Chinese medicinal material, including:
[0017] Based on logical reasoning, obtain the component information of the first Chinese medicinal herb and the case information corresponding to the first disease;
[0018] Based on the component information of the first Chinese medicinal material and the case information corresponding to the first disease, we detected different types of biomolecules in the first Chinese medicinal material that are related to the first disease.
[0019] In one possible implementation of the first aspect, different types of biomolecules related to the first disease in the first traditional Chinese medicine are detected based on the component information of the first medicinal material and the case information corresponding to the first disease, including:
[0020] Based on the component information of the first Chinese medicinal material and the case information corresponding to the first disease, the first type of biomolecule information related to the first disease in the first Chinese medicinal material is detected to obtain the first result;
[0021] The second interactive information is generated based on the first result; wherein the second interactive information includes a second instruction, a second logical basis, and a second example;
[0022] Based on the second interactive information, the second type of biomolecular information related to the first disease in the first Chinese medicinal material is detected to obtain the second result; wherein the first type is different from the second type.
[0023] In one possible implementation of the first aspect, the first interactive information includes a first example, which indicates the format of the input and output of the first interactive information;
[0024] The first analytical result is generated based on the regulatory mechanism, including:
[0025] The first parsing result is generated based on the regulatory mechanism and conforms to the format of the first example.
[0026] In one possible implementation of the first aspect, the method further includes:
[0027] The first analysis result is evaluated;
[0028] If the evaluation result of the first analysis result is unqualified, the first interaction information is readjusted to obtain the adjusted third interaction information;
[0029] Predict the regulatory mechanism of the first Chinese medicinal material on the first symptom based on the third interactive information;
[0030] A second analytical result is generated based on the regulatory mechanism, and the second analytical result is re-evaluated.
[0031] In one possible implementation of the first aspect, the method further includes:
[0032] Obtain a preset model; wherein the preset model is used to detect different types of biomolecular information related to the first disease in the first Chinese medicine based on the first interactive information, predict the regulatory mechanism of the first Chinese medicine on the first disease based on the different types of biomolecular information, and generate the first analysis result based on the regulatory mechanism.
[0033] Input the first interactive information into the preset model and output the first analysis result;
[0034] The first analysis result is displayed to the user.
[0035] Secondly, embodiments of this application provide an apparatus for analyzing the mechanisms of traditional Chinese medicine, comprising:
[0036] The interactive information acquisition module is used to acquire the first interactive information input by the user, which is related to the first symptom and the first Chinese medicinal material.
[0037] The molecular information detection module is used to detect different types of biomolecular information related to the first disease in the first Chinese medicinal material based on the first interactive information.
[0038] The regulatory mechanism prediction module is used to predict the regulatory mechanism of the first traditional Chinese medicine on the first disease based on different types of biomolecular information;
[0039] The analysis result generation module is used to generate the first analysis result based on the regulation mechanism;
[0040] The parsing result display module is used to display the first parsing result to the user.
[0041] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the traditional Chinese medicine mechanism analysis method as described in any of the first aspects above.
[0042] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the traditional Chinese medicine mechanism analysis method as described in any of the first aspects above.
[0043] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the traditional Chinese medicine mechanism analysis method described in any of the first aspects above. Attached Figure Description
[0044] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0045] Figure 1 This is a flowchart illustrating the method for analyzing the mechanism of traditional Chinese medicine provided in the embodiments of this application;
[0046] Figure 2 This is a schematic flowchart of the detection of biomolecular information provided in the embodiments of this application. Figure 1 ;
[0047] Figure 3 This is a schematic flowchart of the detection of biomolecular information provided in the embodiments of this application. Figure 2 ;
[0048] Figure 4 This is a schematic diagram of the process for evaluating the parsing results provided in an embodiment of this application;
[0049] Figure 5 This is a schematic diagram of the process for analyzing the mechanism of traditional Chinese medicine using the model provided in the embodiments of this application;
[0050] Figure 6 This is a schematic diagram of the overall structure for analyzing the mechanisms of traditional Chinese medicine provided in the embodiments of this application;
[0051] Figure 7 This is a structural block diagram of the device for analyzing the mechanism of traditional Chinese medicine provided in the embodiments of this application;
[0052] Figure 8 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation
[0053] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.
[0054] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.
[0055] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.
[0056] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."
[0057] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0058] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.
[0059] In recent years, artificial intelligence technology has achieved significant breakthroughs globally. The deep integration of traditional Chinese medicine (TCM) and artificial intelligence has provided new opportunities for the modernization of TCM. Over the years, a large number of studies have utilized big data and deep learning methods to analyze the mechanisms of TCM and have made progress.
[0060] Due to the high complexity of biomolecular regulatory mechanisms in traditional Chinese medicine (TCM), elucidating the scientific connotation of TCM theory and its diagnostic and treatment methods still faces enormous challenges. Machine learning methods used in related technologies, such as support vector machines and random forests, cannot analyze the biomolecular regulatory mechanisms in TCM. Therefore, there is an urgent need for an analytical method to resolve the regulatory mechanisms contained in biomolecules.
[0061] To address the problems in the aforementioned related technologies, this application provides a method for analyzing the mechanisms of traditional Chinese medicine (TCM). This method integrates a strong logical reasoning AI large language model (DeepSeek R1), a DNA sequence model (Evo 2), a large RNA sequence model (CRAFTS), a large cis-regulatory element model (seq2PRINT), and a protein structure prediction model (AlphaFold3). These models are adept at processing different modalities of data, such as natural language, DNA sequences, RNA sequences, and protein sequences, and collaborate to demonstrate knowledge emergence capabilities. Based on this large model cluster, through reinforcement learning and suggestive engineering design, a cross-modal thought chain causal relationship discovery architecture is established to mine biomolecular information related to the mechanisms of TCM and reveal the regulatory laws of biomolecular expression.
[0062] The above-mentioned integrated artificial intelligence model is used to analyze the mechanisms of traditional Chinese medicine. (See [link to relevant documentation]). Figure 1 This is a flowchart illustrating the method for analyzing the mechanism of traditional Chinese medicine provided in this application embodiment. It is intended as an example and not a limitation. The method may include the following steps:
[0063] S101, Obtain the first interactive information input by the user, which is related to the first symptom and the first Chinese medicinal material.
[0064] In this embodiment, the first interactive information is the prompting engineering information proposed in this application. Prompt engineering refers to the technology of designing, constructing, and optimizing prompts to enable large-scale artificial intelligence models to produce specific and useful outputs. A prompt is a text instruction input to the model, containing key elements such as task description, background information, examples, or constraints. In complex tasks, prompting engineering helps the model understand ambiguous or abstract problems, transforming them into actionable steps, and thus generating content that meets user needs. In traditional Chinese medicine research, it can guide the model to accurately extract information related to specific diseases and Chinese medicines from a large amount of literature, avoiding the output of irrelevant or erroneous content.
[0065] In Traditional Chinese Medicine (TCM) research, the first interactive information refers to the information related to the medical scenario that the user initially inputs, typically presented in the form of "symptom (first symptom) + Chinese medicinal material (first Chinese medicinal material)". The first symptom is the described health problem, such as insomnia or heart failure, while the Chinese medicinal material is the herb used to treat the mentioned first symptom, such as "Ziziphus jujuba" or "Astragalus membranaceus". By designing specific prompting engineering (first interactive information), the large model is guided to generate more accurate and logically coherent output. In this scenario, prompting engineering helps the model understand user input and extract key information.
[0066] In one embodiment, the first interactive information includes a first instruction, and step S101 includes:
[0067] Obtain information on the first symptom and the first Chinese medicinal herb according to the first instruction.
[0068] In this embodiment, the prompting process includes a first instruction, which serves as the basis for obtaining information. This instruction is a clear directive issued by the user, system, or program to guide the direction of information retrieval. In the intelligent analysis system for traditional Chinese medicine, the prompting process input by the user (e.g., "Analyze the mechanism of action of Astragalus membranaceus in treating heart failure") is the first instruction. Instructions can take various forms, such as text input boxes or voice commands, and must clearly express the intent to obtain information.
[0069] For example, initially, the first command a user inputs to the DeepSeek R1, a large-scale AI model integrating strong logical reasoning, could be: "Extract all information related to Astragalus membranaceus in the treatment of heart failure from authoritative TCM academic literature and modern medical research literature, including the components of Astragalus membranaceus, the pathological characteristics of heart failure, known related targets and pathways, arranged in descending order of information relevance, and listing the literature sources in detail." Based on the first command, information on the first disease to be extracted, such as heart failure, and information on the first TCM herb, such as Astragalus membranaceus, can be obtained.
[0070] S102, based on the first interactive information, detect different types of biomolecular information related to the first disease in the first traditional Chinese medicine material.
[0071] In the embodiments of this application, each traditional Chinese medicine contains multiple chemical components, which may exert their effects in vivo through different biomolecular mechanisms. For example, Astragalus contains multiple chemical components, which are potential bioactive molecules. These include astragaloside and astragalus polysaccharides, which may be key to its therapeutic effect on heart failure. The DeepSeek R1 AI model, with its strong logical reasoning capabilities, can accurately detect biomolecular information related to heart failure in Astragalus, including different types of biomolecular information at the gene, RNA, and protein levels.
[0072] In one embodiment, the first interactive information includes a first logical basis, which is used to provide a solution to the problem and related information; see also Figure 2 This is a schematic diagram of the process for detecting biomolecular information provided in the embodiments of this application. Figure 1 ,like Figure 2 As shown, step S102 includes:
[0073] S201, based on logical basis, obtain the component information of the first Chinese medicinal material and the case information corresponding to the first disease.
[0074] In this embodiment, the prompting process includes a logical basis (first logical basis). This logical basis provides the rationale for the prompt content, guides the model's reasoning direction, and ensures the output meets expectations. It contains background knowledge, principles, and rules, and is a key element of prompt design, playing a crucial role in guiding the model to generate effective content. The prompting process serves a specific task objective (first instruction), and the task requirements determine the direction of the logical basis. If the task is to analyze the molecular mechanisms of traditional Chinese medicine in treating diseases, then the logical basis will focus on principles of biomolecular regulation, the relationship between gene expression and disease, and other related content.
[0075] For example, when using models to analyze the treatment of heart failure with Astragalus membranaceus, the goal is to elucidate the molecular mechanisms. Logical support should be sought at the biomolecular level, including genes, RNA, and proteins, to guide the model's output of information related to biomolecular regulation, including target genes, regulatory elements, and signaling pathways, thus meeting the task's requirements for depth and expertise. Based on this logical support, the model could be designed as follows: "Astragalus membranaceus contains various active ingredients, such as astragaloside and astragalus polysaccharides, which have been found to have regulatory effects on the cardiovascular system in previous studies. Heart failure is a complex cardiovascular disease involving multiple pathological processes such as myocardial remodeling and neuroendocrine activation. Related studies have shown that traditional Chinese medicine components can improve the disease state by acting on specific targets and influencing cellular pathways."
[0076] Based on the first logical basis, one can obtain information such as the composition of the first Chinese medicinal material (e.g., Astragalus membranaceus) and the case process of the first disease (e.g., heart failure).
[0077] S202, based on the component information of the first Chinese medicinal material and the case information corresponding to the first disease, detect different types of biomolecule information in the first Chinese medicinal material that are related to the first disease.
[0078] In the embodiments of this application, the components of Chinese medicinal materials are diverse. These components are the material basis for the therapeutic effects of Chinese medicinal materials. Different components may have different biological activities and mechanisms of action. The case information of the first case includes clinical symptoms, pathological examination results, gene and protein expression data, etc.
[0079] For example, patients with heart failure may exhibit pathological features such as increased cardiomyocyte apoptosis and enhanced oxidative stress, as well as abnormal changes in the expression of certain genes and proteins. This information provides a starting point for studying the mechanism of Astragalus membranaceus in treating heart failure. Based on this information, it can be predicted that components of Astragalus membranaceus can act on biomolecular information related to heart failure, thereby improving the disease state.
[0080] In one embodiment, see Figure 3 This is a schematic diagram of the process for detecting biomolecular information provided in the embodiments of this application. Figure 2 ,like Figure 3 As shown, step S202 includes:
[0081] S301, based on the component information of the first traditional Chinese medicine and the case information corresponding to the first disease, detect the first type of biomolecule information related to the first disease in the first traditional Chinese medicine and obtain the first result.
[0082] In this application, DeepSeek R1 can be used to predict the gene expression information of traditional Chinese medicine herbs such as Astragalus membranaceus on the primary disease, such as heart failure. Heart failure often involves pathological processes such as myocardial remodeling and neuroendocrine activation, and abnormal expression of certain genes plays a key role in these processes. The application detects whether Astragalus membranaceus components can regulate the expression of these genes and outputs the first result according to the example format. For example, the first result output by DeepSeek R1 indicates that Astragalus membranaceus components may affect the expression levels of target genes such as Akt and eNOS (first-type biomolecular information), thereby improving the pathological state of heart failure.
[0083] The above example, which should be noted, is the first example contained in the first interactive information. It is a basic format for demonstrating input-output pairs through a small number of samples. Each example includes a question, reasoning process, and answer, and is designed to be concise and clear, focusing on the core of the question.
[0084] The first example could be: "Instruction: Analyze the mechanism of action of ginseng in treating coronary heart disease. Reasoning process: First, find the main components of ginseng in the literature, such as ginsenosides; then study the pathological process of coronary heart disease, including coronary atherosclerosis and myocardial ischemia; then identify the relevant targets of ginseng components, such as vascular endothelial growth factor receptors; finally, analyze the cellular pathways affected by these targets, such as the PI3K-Akt pathway, thereby improving myocardial ischemia and alleviating the symptoms of coronary heart disease. Answer: Ginseng, through its component ginsenosides, acts on targets such as vascular endothelial growth factor receptors, activates the PI3K-Akt pathway, improves myocardial ischemia, and plays a therapeutic role in coronary heart disease." DeepSeek R1 can infer output results that conform to the example format based on the example.
[0085] S302, generate second interactive information based on the first result; wherein, the second interactive information includes a second instruction, a second logical basis, and a second example.
[0086] In this embodiment, after obtaining the first result from DeepSeek R1, in order to analyze the specific impact of the Astragalus component on Akt gene expression, a second interactive information, i.e., a new prompting process, needs to be designed based on the first result. This prompting process can be input into the DNA sequence model Evo 2 for detection. The second interactive information also includes instruction information, logical basis information, and example information (second example).
[0087] For example, the second instruction could be: based on the target gene (such as the Akt gene) in the DeepSeek R1 output, analyze the DNA sequence of its million-nucleotide context length, predict the possible gene expression regulatory molecules such as RNA and regulatory elements that are unknown to humans, so that researchers can refer to these prediction results to conduct in-depth analysis of the regulatory mechanism of Akt gene expression at the DNA level by RNA and regulatory elements that are unknown to humans, and elaborate on the basis of prediction and potential regulatory modes.
[0088] The second logical basis is that gene expression is influenced by RNA and regulatory elements in the surrounding DNA sequence, such as promoters, enhancers, and silencers. These regulatory elements affect the initiation and rate of gene transcription by binding to transcription factors. While gene regulation mechanisms differ across species and tissues, they exhibit certain conserved patterns.
[0089] It should be noted that, based on the target gene such as the Akt gene in the first result, its million-nucleotide context length DNA sequence can be found in an open-source DNA sequence database, and this sequence can be input into the DNA sequence model Evo 2. At the same time, the DNA sequence model Evo 2 can perform corresponding analysis after receiving the second instruction and the second logical basis.
[0090] S303, based on the second interactive information, detect the second type of biomolecular information related to the first disease in the first Chinese medicinal material, and obtain the second result; wherein the first type is different from the second type.
[0091] In this embodiment, the second type of biomolecular information can be the molecular information corresponding to the million-nucleotide context length DNA sequence of the target gene. The DNA sequence model Evo 2 analyzes the second biomolecular information based on the received prompt engineering (second interaction information) to obtain the regulatory process of traditional Chinese medicine herbs such as Astragalus membranaceus on target genes such as the Akt gene, as well as new bioanalytical information. For example, the second result might be: "A potential enhancer region was found in the DNA sequence of the Akt gene, located approximately 5000 bp upstream of the gene. It is predicted that under the influence of Astragalus membranaceus, specific transcription factors may bind to this enhancer, promoting Akt gene transcription by recruiting transcription-related proteins such as RNA polymerase, thereby affecting its expression level."
[0092] In one embodiment, the second result includes RNA-related regulatory information. Based on the RNA-related regulatory information in the second result and the effect of RNA molecules in the treatment of heart failure by Astragalus membranaceus in the first result, new interaction information can be generated and input into the next large RNA sequence model CRAFTS. Furthermore, based on the RNA information related to the treatment of heart failure by Astragalus membranaceus output by DeepSeek R1 and the analysis results of Evo 2 on the regulation of target gene DNA, the corresponding RNA sequence can be searched in the open-source RNA sequence database and the above sequence can be input into the large RNA sequence model CRAFTS.
[0093] The CRAFTS large-scale RNA sequence model incorporates new logical bases, such as: the structure of RNA molecules determines their function; RNAs with different structures can interact with other RNAs or proteins to participate in various biological processes, including gene expression regulation and cell metabolism. In heart failure, abnormal RNA regulation is closely related to disease progression. Researchers can analyze the predicted output of the CRAFTS large-scale RNA sequence model to understand the unknown structures and potential regulatory functions of RNA molecules, thereby gaining a deeper understanding of RNA-RNA and RNA-protein interactions and explaining their impact on the pathological process of heart failure. The CRAFTS large-scale RNA sequence model analyzes the received information and outputs new results, such as potentially identifying unknown non-coding RNAs associated with Astragalus membranaceus in the treatment of heart failure, which have specific stem-loop structures and are predicted to interact with protein X involved in the regulation of cardiomyocyte apoptosis, inhibiting its pro-apoptotic function, thereby reducing cardiomyocyte apoptosis and improving heart failure symptoms.
[0094] In another embodiment, based on the results of DeepSeek R1 and the analysis of DNA regulatory elements (biomolecules) by Evo 2, the corresponding DNA sequence can be searched in an open-source DNA sequence database. The above sequence is then input into the cis-regulatory element large model seq2PRINT. The corresponding DNA sequence can be searched in the source DNA sequence database and input into the cis-regulatory element large model seq2PRINT. The interactive information is generated by combining the results of DeepSeek R1 and the analysis of DNA regulatory elements by Evo 2. The interactive information also includes instructions and corresponding logical basis.
[0095] The logical basis could be: cis-regulatory elements play a crucial role in gene expression regulation by binding to transcription factors. Different cellular environments result in varying expression and activity of transcription factors, leading to alterations in the binding patterns of cis-regulatory elements, which in turn affect gene expression networks and are closely related to the occurrence, development, and treatment of diseases. The instruction could be: identify the cis-regulatory elements associated with Astragalus membranaceus in the treatment of heart failure, analyze the impact of binding changes on gene expression networks, and pay particular attention to genes related to the pathological process of heart failure.
[0096] The seq2PRINT model for cis-regulatory elements analyzes the received information and outputs new results. For example, the output might reveal previously unknown aspects, such as: "During the treatment of heart failure with Astragalus membranaceus, it was found that specific cis-regulatory elements in cardiomyocytes showed significantly enhanced binding to transcription factor Y after treatment. This change in binding affected the expression of multiple genes related to heart failure pathology, including the Akt gene, and improved myocardial function by regulating related signaling pathways."
[0097] In another embodiment, based on the protein information related to Astragalus membranaceus in treating heart failure output by DeepSeek R1, and the analysis results of genes, RNA, and regulation by the previous model, the corresponding amino acid sequence (biomolecular information) is searched in an open-source protein sequence database, and the above sequence is input into the protein structure prediction model AlphaFold3. Furthermore, based on the protein information related to Astragalus membranaceus in treating heart failure output by DeepSeek R1, and the analysis results of genes, RNA, and regulation by the previous model, interactive information is generated. The protein structure prediction model AlphaFold3 then analyzes the regulatory mechanism of the traditional Chinese medicine in the first symptom based on the input information.
[0098] S103, predicting the regulatory mechanism of the first traditional Chinese medicine material on the first disease based on different types of biomolecular information.
[0099] In this embodiment, the final protein structure prediction model, AlphaFold3, combines the analysis results of genes, RNA, regulatory elements, etc., from multiple models to perform a multi-faceted analysis, revealing the specific regulatory mechanism by which the first traditional Chinese medicine regulates the first symptom. Since the function of a protein is determined by its three-dimensional structure, structural changes affect the interaction between proteins and other molecules, thereby influencing biological processes. In disease states, protein structure and function may change, and drug treatment may restore normal function or produce therapeutic effects by regulating protein structure.
[0100] Based on the above information, the protein structure prediction model AlphaFold3 can predict the regulatory mechanism of the first traditional Chinese medicine on the first symptom. For example, the output result may be: predicting protein Z related to Astragalus membranaceus in the treatment of heart failure. In the state of heart failure, a certain helical region in its structure is distorted, affecting its ability to bind to the substrate. Under the action of Astragalus membranaceus, the structure of protein Z is partially restored to normal, enhancing its binding to the substrate. By regulating related metabolic pathways, it improves myocardial energy metabolism in heart failure and exerts a therapeutic effect.
[0101] S104, the first analytical result is generated based on the regulatory mechanism.
[0102] In this application embodiment, the first analytical result is a preliminary systematic analysis conclusion of traditional Chinese medicine materials in treating diseases, covering multiple aspects. A detailed analytical report (first analytical result) can be generated based on the obtained regulatory mechanisms, including identifying the main therapeutic components in Astragalus membranaceus and the key biomolecular targets of these components; explaining the signaling pathways through which these components regulate related biomolecules; and demonstrating how this regulation improves the pathological process of heart failure, such as reducing cardiomyocyte apoptosis and lowering oxidative stress levels. The first analytical result provides a foundation for further in-depth research, allowing researchers to further verify and refine their understanding of the mechanism by which Astragalus membranaceus treats heart failure, and also provides important theoretical basis for clinical application and drug development.
[0103] The above method obtains user input regarding a specific symptom (the first symptom) and a specific traditional Chinese medicine (TCM) herb (the first TCM herb), and analyzes the various types of biomolecular information contained in the TCM herb. This multi-type biomolecular information allows for analysis of the input interaction information from multiple perspectives, leading to a more comprehensive and accurate understanding of the regulatory mechanism of the biomolecules in the extracted TCM herb on the first symptom, resulting in a more valuable analytical result, i.e., the first analytical result. By utilizing various different types of biomolecular information, a more comprehensive understanding of the regulatory mechanism of the biomolecules in the first TCM herb on the first symptom can be achieved. Therefore, the above method can analyze the regulatory mechanisms inherent in biomolecules within the mechanisms of traditional Chinese medicine.
[0104] In one embodiment, the first interaction information includes a first example, which indicates the format of the input and output of the first interaction information; step S104 includes:
[0105] A first parsing result conforming to the format of the first example is generated based on the regulatory mechanism.
[0106] In this embodiment of the application, the first parsing result also needs to be in the format given in the above example, i.e., the first example.
[0107] In one embodiment, see Figure 4 This is a schematic diagram of the process for evaluating the parsing results provided in an embodiment of this application, such as... Figure 4 As shown, the method also includes:
[0108] S401, Evaluate the first analysis result.
[0109] In this embodiment of the application, the opinions of traditional Chinese medicine experts are fully considered to establish a scientific evaluation system to ensure the rationality of its structure and content. The first analysis result is evaluated according to the evaluation system to determine its rationality.
[0110] S402, if the evaluation result of the first parsing result is unqualified, then the first interactive information is readjusted to obtain the adjusted third interactive information.
[0111] In this embodiment of the application, if the evaluation result indicates that the first analysis result is unqualified, it is necessary to redesign the interaction information input to the large artificial intelligence model to obtain the newly designed interaction information (third interaction information).
[0112] S403, based on the third interactive information, predict the regulatory mechanism of the first Chinese medicinal material on the first symptom.
[0113] In this application, new interactive information is input into the model to re-predict the regulatory mechanism of Chinese herbal medicines on the primary symptom.
[0114] S404, a second analytical result is generated based on the regulatory mechanism, and the second analytical result is re-evaluated.
[0115] In this embodiment, a new analytical result (second analytical result) is generated based on the predicted new regulatory mechanism, and the new result is re-evaluated. The interaction information is iteratively optimized based on the evaluation result to improve inference quality and efficiency.
[0116] S105, display the first parsing result to the user.
[0117] In this embodiment of the application, if the first analysis result is reasonable, it will be displayed to the user so that the user can have a more comprehensive understanding of the regulatory mechanism of Chinese medicine on the disease.
[0118] In one embodiment, see Figure 5 This is a schematic diagram of the model analysis of the mechanism of traditional Chinese medicine provided in the embodiments of this application, such as... Figure 5 As shown, the method also includes:
[0119] S501, Obtain a preset model; wherein, the preset model is used to detect different types of biomolecular information related to the first disease in the first traditional Chinese medicine material based on the first interactive information, predict the regulatory mechanism of the first traditional Chinese medicine material on the first disease based on the different types of biomolecular information, and generate the first analysis result based on the regulatory mechanism.
[0120] In this embodiment, the preset model is an integrated artificial intelligence large model cluster, which integrates multiple artificial intelligence large models with different functions and specialties using logical thinking chains to build a collaborative cluster system.
[0121] These large-scale artificial intelligence models include DeepSeekR1, a large-scale language model with strong logical reasoning; Evo2, a large-scale model based on DNA molecules; CRAFTS, a large-scale model based on RNA sequences; seq2PRINT, a deep learning network model for multi-scale footprint cis-regulatory elements; and AlphaFold3, a protein structure prediction model. Each of them excels at processing different modalities of data, such as natural language, DNA sequences, RNA sequences, and protein sequences, in the analysis of mechanisms in traditional Chinese medicine. They possess long biological sequence analysis capabilities far exceeding human cognition and demonstrate knowledge emergence capabilities through mutual prompting.
[0122] S502, input the first interactive information into the preset model and output the first parsing result.
[0123] In this embodiment, the preset model designs a cross-modal thought chain causal relationship discovery architecture through methods such as reinforcement learning and cue engineering (interactive information), performing causal self-consistent chain-like logical reasoning and verification. Through multiple rounds of iterative decision-making by an interactive agent (interaction between multiple artificial intelligence models), the causal logic, correctness, consistency, and completeness of the collaborative cross-modal reasoning of the large model are improved. Therefore, after inputting the first output information into the preset model, the corresponding parsing result (first parsing result) can be output.
[0124] S503 displays the first parsing result to the user.
[0125] In this embodiment of the application, the first parsing result is evaluated. If the evaluation is qualified, the first parsing result is displayed to the user. If it is not qualified, the first interactive information input is adjusted.
[0126] See Figure 6 This is a schematic diagram of the overall structure for analyzing the mechanisms of traditional Chinese medicine provided in the embodiments of this application, as shown below. Figure 6As shown, the specific process is as follows:
[0127] Artificial Intelligence Large Language Model Analysis:
[0128] Input interactive information into the AI large language model (AI large language model analysis), such as the interactive information related to Astragalus membranaceus and mental failure, to enable it to mine biomolecular information related to the mechanism of traditional Chinese medicine from the literature, such as target genes, RNA, proteins and other information.
[0129] DNA Series Large Model Analysis
[0130] Based on target genes (such as the Akt gene) output by DeepSeek R1, the Evo 2 model searches for their contextual DNA sequences (within a million nucleotides) in open-source DNA sequence databases. Evo 2, trained with 40 billion parameters and based on over 9.3 trillion DNA sequence data, can perform comprehensive and in-depth analysis and prediction of regulatory mechanisms in input DNA sequences, uncovering previously unknown DNA molecular regulatory mechanisms related to traditional Chinese medicine.
[0131] Large-scale RNA sequence model analysis
[0132] Based on the RNA information related to the mechanisms of traditional Chinese medicine output by DeepSeek R1 and the analysis results of Evo 2 on the DNA regulation of target genes, corresponding RNA sequences were searched in open-source RNA sequence databases and input into the CRAFTS model. CRAFTS extracts semantic features of RNA sequences through self-supervised learning, mines folding patterns from massive RNA sequences, predicts complex conformations of multimeric RNAs, and then analyzes the unknown RNA molecular regulatory mechanisms related to traditional Chinese medicine.
[0133] Large-scale model analysis of cis-regulatory elements
[0134] Based on the results of DeepSeek R1 and the analysis of DNA regulatory elements by Evo 2, corresponding DNA sequences were searched in open-source DNA sequence databases and input into the seq2PRINT model. This model was used to resolve the sequence-level organization of multi-scale footprints in cis-regulatory elements, predict transcription factor binding, and discover the molecular regulatory mechanisms of cis-regulatory elements related to traditional Chinese medicine treatment.
[0135] Protein structure prediction model analysis:
[0136] Based on the protein information output by DeepSeek R1, and the analysis results of genes, RNA, and regulation from other models, the corresponding amino acid sequences are searched in open-source protein sequence databases and input into the AlphaFold3 model. AlphaFold3 predicts the three-dimensional structure of proteins through one-dimensional sequences, analyzes the relationship between protein structural changes and functions, and explores the unknown molecular regulation and interaction mechanisms of proteins related to traditional Chinese medicine.
[0137] This application combines natural language processing with an interactive intelligent agent based on a large-scale biological model. This agent can accurately extract biomolecular information, including target genes, RNA, and proteins, related to known TCM mechanisms from a vast amount of academic literature, targeting core syndromes in TCM. Furthermore, based on the DNA sequences containing target genes, RNA, and proteins—up to millions of nucleotides—this agent can deeply deduce the complex interactions and correlations between epigenetic regulatory networks, cis-regulatory elements, and protein structures, thereby revealing the fundamental mechanisms and laws governing the expression regulation of target genes, RNA, cis-regulatory elements, and proteins. This innovative method provides a completely new approach and means for the mechanistic research of TCM.
[0138] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0139] Corresponding to the TCM mechanism analysis method in the above embodiments, Figure 7 This is a structural block diagram of the traditional Chinese medicine mechanism analysis device 7 provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.
[0140] Reference Figure 7 The TCM mechanism analysis device 7 includes:
[0141] The interactive information acquisition module 71 is used to acquire the first interactive information input by the user, which is related to the first symptom and the first Chinese medicinal material.
[0142] Molecular information detection module 72 is used to detect different types of biomolecular information related to the first disease in the first Chinese herbal medicine based on the first interactive information;
[0143] The regulatory mechanism prediction module 73 is used to predict the regulatory mechanism of the first Chinese medicine on the first disease based on different types of biomolecular information;
[0144] The analysis result generation module 74 is used to generate the first analysis result based on the regulation mechanism;
[0145] The TCM mechanism analysis device 7 also includes: an analysis result evaluation module 75, used for:
[0146] The first analysis result is evaluated;
[0147] If the evaluation result of the first analysis result is unqualified, the first interaction information is readjusted to obtain the adjusted third interaction information;
[0148] Predict the regulatory mechanism of the first Chinese medicinal material on the first symptom based on the third interactive information;
[0149] A second analytical result is generated based on the regulatory mechanism, and the second analytical result is re-evaluated.
[0150] The parsing result display module 76 is used to display the first parsing result to the user.
[0151] Optionally, the interactive information acquisition module 71 is also used for:
[0152] Obtain information on the first symptom and the first Chinese medicinal herb according to the first instruction.
[0153] Optionally, the molecular information detection module 72 is also used for:
[0154] Based on logical reasoning, obtain the component information of the first Chinese medicinal herb and the case information corresponding to the first disease;
[0155] Based on the component information of the first Chinese medicinal material and the case information corresponding to the first disease, we detected different types of biomolecules in the first Chinese medicinal material that are related to the first disease.
[0156] Optionally, the molecular information detection module 72 is also used for:
[0157] Based on the component information of the first Chinese medicinal material and the case information corresponding to the first disease, the first type of biomolecule information related to the first disease in the first Chinese medicinal material is detected to obtain the first result;
[0158] The second interactive information is generated based on the first result; wherein the second interactive information includes a second instruction, a second logical basis, and a second example;
[0159] Based on the second interactive information, the second type of biomolecular information related to the first disease in the first Chinese medicinal material is detected to obtain the second result; wherein the first type is different from the second type.
[0160] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.
[0161] in addition, Figure 7The TCM mechanism analysis device shown can be a software unit, hardware unit, or a combination of software and hardware built into existing terminal equipment, or it can be integrated into the terminal equipment as an independent accessory, or it can exist as an independent terminal equipment.
[0162] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0163] Figure 8 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. For example... Figure 8 As shown, the terminal device 8 of this embodiment includes: at least one processor 80 ( Figure 8 (Only one is shown) a processor, a memory 81, and a computer program 82 stored in the memory 81 and capable of running on at least one processor 80. When the processor 80 executes the computer program 82, it implements the steps in any of the above embodiments of the traditional Chinese medicine mechanism analysis device and method.
[0164] The terminal device can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 8 This is merely an example of terminal device 8 and does not constitute a limitation on terminal device 8. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.
[0165] The processor 80 may be a Central Processing Unit (CPU), or it may 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. A general-purpose processor may be a microprocessor or any conventional processor.
[0166] In some embodiments, memory 81 may be an internal storage unit of the terminal device 8, such as a hard disk or memory of the terminal device 8. In other embodiments, memory 81 may be an external storage device of the terminal device 8, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the terminal device 8. Furthermore, memory 81 may include both internal storage units and external storage devices of the terminal device 8. Memory 81 is used to store operating systems, applications, bootloaders, data, and other programs, such as program code of computer programs. Memory 81 can also be used to temporarily store data that has been output or will be output.
[0167] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can implement the steps in the above-described method embodiments.
[0168] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.
[0169] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / terminal equipment, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.
[0170] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.
[0171] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0172] In the embodiments provided in this application, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / terminal device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.
[0173] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0174] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.
Claims
1. A method for analyzing the mechanisms of traditional Chinese medicine, characterized in that, include: Obtain first interactive information input by the user, which is related to a first symptom and a first traditional Chinese medicine material; Based on the first interactive information, detect different types of biomolecule information in the first Chinese medicinal material that are related to the first symptom; Predict the regulatory mechanism of the first traditional Chinese medicine on the first symptom based on the different types of biomolecular information; The first analytical result is generated based on the described regulatory mechanism; The first parsing result is displayed to the user.
2. The method for analyzing the mechanism of traditional Chinese medicine as described in claim 1, characterized in that, The first interactive information includes a first instruction; The acquisition of the first interaction information input by the user includes: Information about the first symptom and information about the first Chinese medicinal material are obtained according to the first instruction.
3. The method for analyzing the mechanism of traditional Chinese medicine as described in claim 1 or 2, characterized in that, The first interactive information includes a first logical basis, which is used to provide solutions to the problem and related information; The detection of different types of biomolecular information related to the first symptom in the first traditional Chinese medicine includes: Based on the aforementioned logical basis, obtain the component information of the first Chinese medicinal material and the case information corresponding to the first disease; Based on the component information of the first Chinese medicinal material and the case information corresponding to the first disease, we detect different types of biomolecules in the first Chinese medicinal material that are related to the first disease.
4. The method for analyzing the mechanism of traditional Chinese medicine as described in claim 3, characterized in that, The step of detecting different types of biomolecules related to the first disease in the first traditional Chinese medicine based on the component information of the first medicinal material and the case information corresponding to the first disease includes: Based on the component information of the first Chinese medicinal material and the case information corresponding to the first disease, the first type of biomolecule information related to the first disease in the first Chinese medicinal material is detected to obtain the first result; A second interactive information is generated based on the first result; wherein the second interactive information includes a second instruction, a second logical basis, and a second example; Based on the second interactive information, the second type of biomolecular information related to the first disease in the first Chinese herbal medicine is detected to obtain a second result; wherein the first type is different from the second type.
5. The method for analyzing the mechanism of traditional Chinese medicine as described in claim 1, characterized in that, The first interactive information includes a first example, which is used to indicate the format of the input and output of the first interactive information; The step of generating the first analytical result based on the regulatory mechanism includes: A first parsing result conforming to the format of the first example is generated based on the described regulation mechanism.
6. The method for analyzing the mechanism of traditional Chinese medicine as described in claim 1, characterized in that, The method further includes: The first analysis result is evaluated; If the evaluation result of the first parsing result is unqualified, the first interaction information is readjusted to obtain the adjusted third interaction information; Based on the third interactive information, predict the regulatory mechanism of the first Chinese herbal medicine on the first symptom; A second analytical result is generated based on the described regulatory mechanism, and the second analytical result is re-evaluated.
7. The method for analyzing the mechanism of traditional Chinese medicine as described in any one of claims 1 to 6, characterized in that, The method further includes: Obtain a preset model; wherein the preset model is used to detect different types of biomolecular information related to the first disease in the first Chinese medicine based on the first interaction information, predict the regulatory mechanism of the first Chinese medicine on the first disease based on the different types of biomolecular information, and generate a first analysis result based on the regulatory mechanism. The first interactive information is input into the preset model, and the first parsing result is output. The first parsing result is displayed to the user.
8. A device for analyzing the mechanism of traditional Chinese medicine, characterized in that, include: An interactive information acquisition module is used to acquire first interactive information input by the user, which is related to a first symptom and a first traditional Chinese medicine material. The molecular information detection module is used to detect different types of biomolecular information related to the first disease in the first Chinese herbal medicine based on the first interactive information. The regulatory mechanism prediction module is used to predict the regulatory mechanism of the first traditional Chinese medicine on the first disease based on different types of biomolecular information. The analysis result generation module is used to generate a first analysis result based on the control mechanism. The parsing result display module is used to display the first parsing result to the user.
9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.
