Target identification apparatus, method, device, computer readable storage medium and computer program product

Abstract

The application relates to a target point recognition device, method, apparatus, computer readable storage medium and computer program product, relates to the technical field of biological information, and can improve the effectiveness of a target point recognition result. The target point recognition device is configured to perform the following steps: determining a candidate biological molecule in tumor cells that can provide an action target point for a targeted drug; performing expression level prediction on the candidate biological molecule by a trained expression prediction model, determining expression levels of the candidate biological molecule at multiple development stages of a tumor according to a prediction result; determining a first importance detection result of the candidate biological molecule in a tumor development process according to the expression levels of the candidate biological molecule at the multiple development stages of the tumor; and judging whether a drug action target point is determined according to the candidate biological molecule according to a tumor specificity detection result of the candidate biological molecule and the first importance detection result.

Description

Technical Field

[0001] This application relates to the field of bioinformatics, and in particular to a target identification device, method, apparatus, computer-readable storage medium, and computer program product. Background Technology

[0002] With the development of biomedical technology, targeted drugs can precisely target diseased cells by identifying and binding to drug targets on tumor cells. In related technologies, the screening and identification of drug targets on tumor cells is mainly based on their expression levels. For example, biomolecules with high expression levels are selected as drug targets, so that when using corresponding targeted drugs, the drugs can act on tumor cells through the highly expressed drug targets.

[0003] However, in practice, it has been found that high levels of drug target expression do not necessarily equate to better efficacy of the targeted drug. In some cases, even when the drug target is expressed at low levels, the targeted drug can still deliver good clinical results. Therefore, the effectiveness of methods for identifying drug targets based on expression levels in related technologies still needs improvement. Summary of the Invention

[0004] Therefore, it is necessary to provide a target identification device, method, apparatus, computer-readable storage medium, and computer program product to address the aforementioned technical problems.

[0005] In a first aspect, this application provides a target identification device, which is configured to perform the following steps:

[0006] Identify candidate biomolecules in tumor cells that can provide targets for targeted drugs;

[0007] The expression levels of the candidate biomolecules are predicted by a trained expression prediction model, and the expression levels of the candidate biomolecules at multiple stages of tumor development are determined based on the prediction results.

[0008] Based on the expression levels of the candidate biomolecules at multiple stages of tumor development, the first importance detection results of the candidate biomolecules in the tumor development process are determined;

[0009] Based on the tumor-specific detection results of the candidate biomolecules and the first importance detection results, determine whether to identify drug targets based on the candidate biomolecules.

[0010] In one exemplary embodiment, determining the candidate biomolecules to be identified in tumor cells includes:

[0011] Obtain a pre-defined target drug and identify drug targets in tumor cells that can bind to the target drug and / or target genes corresponding to the target drug.

[0012] Based on the drug target and / or the target gene, candidate biomolecules to be screened are obtained.

[0013] In one exemplary embodiment, the first importance detection result includes the first importance detection result of the candidate biomolecule in the tumor development process of at least one tumor;

[0014] The step of determining whether to identify a drug target based on the candidate biomolecule, according to the tumor-specific detection results and the first importance detection results, includes:

[0015] For each type of tumor, if the tumor-specific detection results of the candidate biomolecules and the first importance detection results meet preset conditions, the drug target of the tumor is determined based on the candidate biomolecules.

[0016] In one exemplary embodiment, determining the drug target of the tumor based on the candidate biomolecule includes:

[0017] If the candidate biomolecule includes a drug target that can bind to the target drug, the drug target is identified as the drug target of the target drug in the tumor.

[0018] When the candidate biomolecule includes the target gene corresponding to the target drug, the drug target in the tumor is obtained based on the expression results of the target gene.

[0019] In one exemplary embodiment, the first importance detection result includes the first importance detection result of the candidate biomolecule in the tumor development process of at least one tumor;

[0020] The step of determining whether to identify a drug target based on the candidate biomolecule, according to the tumor-specific detection results and the first importance detection results, includes:

[0021] For each of the tumors, if the tumor-specific detection results of the candidate biomolecules meet preset conditions and the first importance detection results do not meet preset conditions, at least one tumor subgroup of the tumor is determined.

[0022] For each of the tumor subgroups, the expression prediction model predicts the expression level of the candidate biomolecules, and the expression level of the candidate biomolecules in multiple developmental stages of the tumor subgroup is determined based on the prediction results. Based on the expression level of the candidate biomolecules in multiple developmental stages of the tumor subgroup, the second importance detection result of the candidate biomolecules in the tumor subgroup is obtained.

[0023] If the second importance detection result meets the preset conditions, the drug targets of the tumor subgroup are determined based on the candidate biomolecules.

[0024] In one exemplary embodiment, the step of predicting the expression level of the candidate biomolecule using a trained expression prediction model, and determining the expression level of the candidate biomolecule at multiple stages of tumor development based on the prediction results, includes:

[0025] Identify target individuals with cancer;

[0026] The expression levels of the candidate biomolecules in the target body are predicted by a trained expression prediction model, and the expression levels of the candidate biomolecules in the target body at multiple stages of tumor development are determined based on the prediction results.

[0027] In one exemplary embodiment, predicting the expression level of the candidate biomolecule in the target organism by the trained expression prediction model includes:

[0028] Obtain the first transcriptome sequencing results corresponding to the normal tissue of the target object and the second transcriptome sequencing results corresponding to the tumor tissue of the target object;

[0029] The first transcriptome sequencing results and the second transcriptome sequencing results are input into a trained expression prediction model. The expression prediction model predicts the evolution trajectory of the first transcriptome sequencing results into the second transcriptome sequencing results, thereby obtaining the transcriptome change trajectory in the target object.

[0030] Based on the transcriptome change trajectory extracted from the transcriptome change trajectory, the predicted expression level of the candidate biomolecule in the target body is determined.

[0031] In one exemplary embodiment, predicting the evolutionary trajectory of the first transcriptome sequencing result into the second transcriptome sequencing result to obtain the transcriptome change trajectory in the target object includes:

[0032] The starting point of the change trajectory is determined based on the first transcriptome sequencing results, and the ending point of the change trajectory is determined based on the second transcriptome sequencing results;

[0033] The transcriptome sequencing results corresponding to each of the multiple intermediate points of the change trajectory between the starting point and the ending point of the change trajectory are predicted, and the transcriptome change trajectory in the target object is obtained based on the prediction results.

[0034] In one exemplary embodiment, determining the first importance detection result of the candidate biomolecule in tumor development based on its expression level at multiple stages of tumor development includes:

[0035] Based on the expression levels of the candidate biomolecules at multiple stages of tumor development, the persistence of expression of the candidate biomolecules in the later stages of tumor development is determined.

[0036] When the expression persistence reaches a preset persistence level, a first importance detection result indicating the importance of the candidate biomolecule in tumor development is obtained.

[0037] Secondly, this application also provides a target identification method, the method comprising:

[0038] Identify candidate biomolecules in tumor cells that can provide targets for targeted drugs;

[0039] The expression levels of the candidate biomolecules are predicted by a trained expression prediction model, and the expression levels of the candidate biomolecules at multiple stages of tumor development are determined based on the prediction results.

[0040] Based on the expression levels of the candidate biomolecules at multiple stages of tumor development, the first importance detection results of the candidate biomolecules in the tumor development process are determined;

[0041] Based on the tumor-specific detection results of the candidate biomolecules and the first importance detection results, determine whether to identify drug targets based on the candidate biomolecules.

[0042] Thirdly, this application also provides a target identification device, the device comprising:

[0043] The candidate molecule identification module is used to identify candidate biomolecules in tumor cells that can provide targets for targeted drugs.

[0044] An expression prediction module is used to predict the expression level of the candidate biomolecules by a trained expression prediction model, and to determine the expression level of the candidate biomolecules at multiple stages of tumor development based on the prediction results.

[0045] The first importance determination module is used to determine the first importance detection result of the candidate biomolecule in the tumor development process based on the expression level of the candidate biomolecule in multiple stages of tumor development.

[0046] The target identification module is used to determine whether to identify a drug target based on the candidate biomolecule, according to the tumor-specific detection results and the first importance detection results.

[0047] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the target identification method.

[0048] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of a target identification method.

[0049] The aforementioned target identification device, method, apparatus, computer-readable storage medium, and computer program product identify candidate biomolecules in tumor cells that can provide targets for targeted drugs. Then, a trained expression prediction model predicts the expression levels of these candidate biomolecules, determining their expression levels at multiple stages of tumor development based on the prediction results. Subsequently, based on the expression levels of these candidate biomolecules at multiple stages of tumor development, the first importance detection result of the candidate biomolecule in the tumor development process is determined. Finally, based on the tumor-specific detection result and the first importance detection result of the candidate biomolecule, it is determined whether a drug target should be identified based on the candidate biomolecule. In this embodiment, by predicting the expression levels of candidate biomolecules that can provide targets for targeted drugs, the dynamic expression trend of candidate biomolecules in tumors can be revealed based on their expression levels at multiple stages of tumor development. This allows for a more accurate determination of whether candidate biomolecules can stably and specifically provide targets in tumors over a long period, effectively improving the identification effectiveness of the finally determined drug targets and enabling precise screening of reliable drug targets for targeted drugs. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating a target identification method in one embodiment;

[0052] Figure 2 This is a schematic diagram illustrating the expression level of the CD19 gene in one embodiment;

[0053] Figure 3A One example is a transcriptome image;

[0054] Figure 3B This is another transcriptome image in one embodiment;

[0055] Figure 4 This is a schematic diagram illustrating the expression level of the CEACAM5 gene in one embodiment;

[0056] Figure 5A This is another transcriptome image in one embodiment;

[0057] Figure 5B This is another transcriptome image in one embodiment;

[0058] Figure 6 This is a schematic diagram illustrating the changes in target gene expression levels during tumorigenesis in one embodiment;

[0059] Figure 7A This is a schematic diagram comparing the total survival time in one embodiment;

[0060] Figure 7B This is a comparative schematic diagram of a progression-free survival in one embodiment;

[0061] Figure 8 This is a structural block diagram of a target identification device in one embodiment;

[0062] Figure 9 This is an internal structural diagram of a target identification device in one embodiment;

[0063] Figure 10 This is an internal structural diagram of another target identification device in one embodiment. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0065] To enable those skilled in the art to better understand this application, the relevant technologies are first introduced below.

[0066] With the development of biomedical technology, targeted drugs can precisely target diseased cells by recognizing and binding to drug targets on tumor cells.

[0067] In related technologies, the screening and identification of drug targets on tumor cells is mainly based on the level of expression. Taking antibody-drug conjugates (ADCs) as an example, the screening of drug targets is mainly based on gene mutations or gene expression levels. For instance, in lung and breast cancer, the genes related to epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor-2 (HER2) have very high mutation rates, leading to high levels of receptor expression. Therefore, EGFR and HER2-based antibody-drug conjugates have excellent clinical efficacy in lung and breast cancer. Similarly, in non-small cell lung carcinoma (NSCLC), HER2 is often highly expressed, allowing HER2-based ADCs to be widely used in the clinical treatment of NSCLC.

[0068] However, the inventors discovered in practice that the expression level of the drug target itself is not always related to clinical efficacy. For example, although the expression level of human trophoblast cell surface antigen 2 (Trop2) is not high in triple-negative breast cancer, Trop2-based ADC drugs have shown excellent clinical efficacy in triple-negative breast cancer. Similarly, drug targets such as CD79b, Nectin-4, or CD30 have generally low expression levels in their respective tumors, but ADC drugs targeting these targets have still achieved good clinical results in the treatment of some tumors (such as CD79b-based ADC drugs in the treatment of B-cell non-Hodgkin lymphoma and CD30-based ADC drugs in the treatment of lymphoma).

[0069] It is evident that the method of identifying drug targets based on expression levels in related technologies is difficult to accurately and effectively determine suitable drug targets, and its identification accuracy still needs to be improved.

[0070] Therefore, it is necessary to provide a target identification device, method, apparatus, computer-readable storage medium, and computer program product to address the aforementioned technical problems.

[0071] In one embodiment, a target identification device is provided. This target identification can be used to identify the targets of a drug. In some exemplary embodiments, the target identification device can be a terminal, a server, or a system including both a terminal and a server. In this embodiment, as... Figure 1 As shown, the target identification device is configured to perform the following steps:

[0072] S101 identifies candidate biomolecules in tumor cells that can provide targets for targeted drugs.

[0073] Candidate biomolecules are biomolecules that can provide a target for targeted drugs. Candidate biomolecules can be developed or yet to be developed. In some examples, candidate biomolecules can be at least one of genes, proteins, enzymes, or cell membrane receptors. For example, known receptors on the surface of tumor cells that can serve as targets can be used as candidate biomolecules, or genes that express the corresponding receptors can be used as candidate biomolecules.

[0074] In practice, candidate biomolecules that can provide targets for targeted drugs can be selected from tumor cells. In some exemplary embodiments, one or more biomolecules with drug-binding characteristics can be identified as candidate biomolecules. Drug-binding characteristics refer to one or more of the physical and chemical characteristics that enable the biomolecule to bind to the drug. For example, biomolecules that can bind to the targeted drug and form a covalent bond can be selected as candidate biomolecules.

[0075] S102 uses a trained expression prediction model to predict the expression levels of candidate biomolecules and determines their expression levels at multiple stages of tumor development based on the prediction results.

[0076] In related technologies, the screening of drug targets mainly relies on static expression level analysis to determine whether a receptor is suitable as a target. However, after creative efforts in practice, the inventors discovered that although some targets are highly expressed at specific stages of tumor development, the expression level of these receptors in the tumor decreases as the disease progresses.

[0077] In this application, the expression prediction model can be pre-trained. In some exemplary embodiments, the expression prediction model can be trained by unsupervised learning, such as training a generative adversarial network by unsupervised learning so that it can generate the expression level of candidate biomolecules at a specified stage; or it can be trained by supervised learning, such as training a neural network by training samples carrying expression level labels so that the model can predict the expression level of candidate biomolecules at a specified stage based on past samples.

[0078] Furthermore, after identifying candidate biomolecules, a trained expression prediction model can predict their expression levels to determine their expression levels at multiple stages of tumor development. These multiple stages can represent changes throughout the entire tumor lifecycle, such as the period before tumor appearance, and the early, middle, and late stages after tumor appearance. It is understood that this embodiment determines the dynamic changes of candidate biomolecules in the later stages of tumor development by predicting their expression levels at multiple stages, rather than simply determining their expression levels at a single static period.

[0079] S103, based on the expression levels of candidate biomolecules at multiple stages of tumor development, determines the primary importance of candidate biomolecules in the tumor development process.

[0080] After determining the expression levels of candidate biomolecules at multiple stages of tumor development, the expression trends of these biomolecules can be analyzed. These trends can then be used to assess the importance of the candidate biomolecules to the tumor development process, yielding an importance detection result. This importance detection result characterizes the degree of importance of the candidate biomolecules to the presence of the tumor; for clarity, the result characterizing the overall importance of the candidate biomolecules to the tumor is referred to as the first importance detection result.

[0081] In this step, by analyzing the expression levels of candidate biomolecules at multiple stages of tumor development, it is possible to determine whether the candidate biomolecules are key elements in the tumor's existence process (such as whether the candidate biomolecules are essential to the tumor, i.e., cannot escape), thereby obtaining the first importance test results of the candidate biomolecules to the tumor development process.

[0082] S104. Based on the tumor-specific detection results and first importance detection results of the candidate biomolecules, determine whether a candidate drug target has been obtained based on the candidate biomolecules.

[0083] In addition, tumor-specific detection results of candidate biomolecules can be obtained. These results can indicate whether the candidate biomolecules are specifically expressed in tumors. When the results indicate that the candidate biomolecules are specifically expressed in tumor cells, it can be determined that the candidate biomolecules can enable the drug to bind more precisely to the tumor cells.

[0084] For example, targeting the CD19 gene, Figure 2The CD19 transcript levels (log2-transformed transcripts per million+1, og2TPM+1) are shown. Figure 2 In this context, "Ctrl" refers to a whole blood sample from a normal sample subject, while "BCL" refers to a blood sample from an abnormal sample subject suffering from B-cell acute lymphoblastic leukemia (BCL). Analysis... Figure 2 It can be seen that the average expression level of the CD19 gene in the blood of abnormal samples was 13.2 (log2TPM+1), while the average expression level in the blood of normal samples was only 9.1 (log2TPM+1).

[0085] Furthermore, based on the tumor-specific and first-importance test results of the candidate biomolecules, it can be determined whether candidate drug targets can be derived from them. Specifically, tumor-specific test results indicate whether the candidate biomolecule is specifically expressed in tumor cells, while first-importance test results indicate whether the candidate biomolecule is necessary for tumor development. By combining tumor-specific and first-importance test results, it is possible to identify whether the candidate biomolecule is tumor-specific relative to normal tissues and whether it is continuously expressed during tumor development. Thus, it can be determined whether it is appropriate to determine the corresponding drug targets based on the candidate biomolecules.

[0086] The aforementioned target identification device can identify candidate biomolecules in tumor cells that can provide targets for targeted drugs. A trained expression prediction model then predicts the expression levels of these candidate biomolecules, determining their expression levels at multiple stages of tumor development. Subsequently, based on these expression levels, the first importance of each candidate biomolecule in tumor development is determined. Finally, based on the tumor-specific and first importance results of the candidate biomolecules, it is determined whether a drug target should be identified. In this embodiment, by predicting the expression levels of candidate biomolecules that can provide targets for targeted drugs, the dynamic expression trends of these biomolecules in tumors can be revealed based on their expression levels at multiple stages of tumor development. This allows for a more accurate assessment of whether candidate biomolecules can stably and specifically provide targets in tumors over a long period, effectively improving the identification effectiveness of the final drug targets and enabling precise screening of reliable drug targets for targeted drugs.

[0087] In an exemplary embodiment, step S101 may include the following steps:

[0088] Obtain the pre-defined target drug and identify the drug target and / or the target gene corresponding to the target drug in tumor cells; based on the drug target and / or target gene, obtain the candidate biomolecules to be screened.

[0089] In practical applications, a pre-defined target drug can be identified. In some exemplary embodiments, the target drug can be an antibody-drug conjugate, such as an antibody-drug conjugate that has already been developed. For example, when the target drug is developed, its suitable drug target may not be clearly defined. Or, for a target drug with a known drug target, relevant personnel may want to identify more suitable drug targets that match the target drug.

[0090] In this embodiment, drug targets that can bind to the targeted drug in tumor cells can be identified. Specifically, for example, proteins, enzymes, or cell membrane receptors that can bind to the targeted drug can be identified among biomolecules on the surface or inside tumor cells as drug targets. For instance, when a receptor can bind to the targeted drug via covalent or non-covalent bonds, it can be considered a drug target. Alternatively, target genes corresponding to the targeted drug can be identified. These target genes can be genes that express the drug target related to the targeted drug. Furthermore, at least one of the drug target and target gene can be used as candidate biomolecules for screening.

[0091] In this embodiment, by identifying drug targets or target genes corresponding to the target drug in tumor cells that can bind to the target drug as candidate biomolecules, and using the target identification device provided in this application to determine whether they can be used to obtain candidate drug targets, it helps to more quickly and efficiently identify more targets that can bind to known target drugs, thereby reducing clinical trial costs and rapidly expanding the clinical application scope of the target drug.

[0092] In one exemplary embodiment, the first importance detection result includes the first importance detection result of the candidate biomolecule in the tumor development process of at least one type of tumor. That is, in the embodiments of this application, for each type of tumor in at least one tumor, candidate biomolecules that can provide targets for targeted drugs in the tumor cells of that tumor can be identified, and then the first importance detection result of the candidate biomolecule for that type of tumor in the tumor development process is obtained by predicting its expression level. It can be understood that by switching different target tumors, the importance of the candidate biomolecule to the corresponding target tumor can be identified separately.

[0093] Accordingly, in step S104, determining whether to identify a drug target based on the candidate biomolecule, according to the tumor-specific detection results of the candidate biomolecule and the first importance detection results, may include:

[0094] For each type of tumor, if the tumor-specific detection results and first importance detection results of the candidate biomolecules meet the preset conditions, the drug target of the tumor is determined based on the candidate biomolecules.

[0095] Specifically, for each type of tumor, a specific drug target for that tumor can be identified based on the candidate biomolecule, provided that the tumor-specific detection results and the first importance detection results of the candidate biomolecule meet preset conditions. In some examples, meeting the preset conditions for tumor-specific detection results and first importance detection results may mean that the tumor-specific detection results indicate that the candidate biomolecule is tumor-specific, and that the first importance detection results indicate that the candidate biomolecule is important to the tumor development process (e.g., it is a key or necessary element for the existence of the tumor).

[0096] In this embodiment, by obtaining the first importance detection results of candidate biomolecules in various tumors, and when the tumor-specific detection results of candidate biomolecules and the first importance detection results of candidate biomolecules in the corresponding tumors meet preset conditions, the drug targets of tumors can be determined based on the candidate biomolecules. This can quickly and accurately identify the indications of various candidate biomolecules, reduce the cost of clinical trials, and quickly identify suitable drug targets for various types of tumors.

[0097] In one exemplary embodiment, determining a drug target for a tumor based on candidate biomolecules may include the following steps:

[0098] When the candidate biomolecule includes a drug target that can bind to the target drug, the drug target is identified as the drug target of the target drug in the tumor; when the candidate biomolecule includes a target gene corresponding to the target drug, the drug target of the target drug in the tumor is obtained based on the expression results of the target gene.

[0099] In practice, candidate biomolecules can be newly identified biomolecules with drug-binding characteristics, such as receptors for which there are no corresponding targeted drugs. By using newly identified biomolecules as candidate biomolecules and predicting their first importance for a specific tumor through expression level analysis, relevant guidance can be provided for the clinical treatment of the corresponding tumor. This can help improve the clinical treatment efficiency of the corresponding tumor by developing targeted drugs that can be adapted to the drug's target.

[0100] On the other hand, candidate biomolecules can also be candidate biomolecules related to known targeted drugs. In this embodiment, if the candidate biomolecule is a drug target that can bind to the targeted drug, then the drug target can be used as the drug target of the targeted drug in the tumor. If the candidate biomolecule is a target gene corresponding to the targeted drug, then the drug target of the targeted drug in the tumor can be obtained based on the expression results of the target gene, such as proteins or cell membrane surface receptors.

[0101] In this embodiment, by using the drug target or target gene corresponding to a known targeted drug as a candidate biomolecule, when the tumor-specific detection results and the first importance detection results in a specific tumor of the candidate biomolecule meet preset conditions, it can be identified as the drug target in a specific tumor. This can quickly determine the indications of a known targeted drug, which helps to apply the developed targeted drug to more tumor treatment scenarios and expand the clinical application scope of the same targeted drug and the scope of benefit for related subjects.

[0102] In an exemplary embodiment, the first importance detection result includes the first importance detection result of the candidate biomolecule in the tumor development process of at least one tumor. That is, in the embodiments of this application, for each type of tumor in at least one tumor, candidate biomolecules that can provide targets for targeted drugs in the tumor cells of that tumor can be identified, and then the first importance detection result of the candidate biomolecule in the tumor development process of that type of tumor can be obtained by predicting its expression level.

[0103] Accordingly, in step S104, determining whether to identify a drug target based on the candidate biomolecules, according to the tumor-specific detection results and the first importance detection results, may include the following steps:

[0104] For each of the tumors, if the tumor-specific detection results of the candidate biomolecules meet preset conditions and the first importance detection results do not meet preset conditions, at least one tumor subgroup is determined; for each tumor subgroup, the expression level of the candidate biomolecules is predicted by an expression prediction model, and the expression level of the candidate biomolecules in multiple developmental stages of the tumor subgroup is determined based on the prediction results; based on the expression level of the candidate biomolecules in multiple developmental stages of the tumor subgroup, the second importance detection result of the candidate biomolecules in the tumor subgroup is obtained; if the second importance detection result meets preset conditions, the drug target of the tumor subgroup is determined based on the candidate biomolecules.

[0105] In practical applications, after selecting one or more designated target tumors, the first importance test results of candidate biomolecules in the tumor development process of each target tumor can be determined. Then, by combining the first importance test results and the tumor-specific test results, candidate biomolecules that are universally induced to be expressed in the target tumor population, continue to be induced to be expressed in the later stages of tumor development, and have tumor specificity can be identified.

[0106] In some embodiments, the tumor-specific detection results and the first importance detection results of the screened candidate biomolecules can simultaneously meet preset conditions. However, in other embodiments, the relevant characteristics of the candidate biomolecules may not meet the preset conditions. For example, the tumor-specific detection results of the candidate biomolecules meet the preset conditions, but the first importance detection results do not meet the preset conditions. That is, although the candidate biomolecules are specifically expressed in tumor tissues, the candidate biomolecules are generally of general importance to the tumor as a whole. The candidate biomolecules are not key or necessary elements in the growth and development of this type of tumor.

[0107] In this embodiment, for each type of tumor, if the tumor-specific detection results of the candidate biomolecules meet the preset conditions but the first importance detection results do not meet the preset conditions, at least one tumor subgroup can be further identified.

[0108] Tumor subgroups refer to subsets of tumors further subdivided based on different characteristics. In some examples, tumor subgroups can be classified according to one or more of the following clinical features: malignancy, tumor cell differentiation, tumor stage, pathological type, gene mutation type, and physiological characteristics of the target population (such as age or gender). For example, lung cancer patients can be divided into tumor subgroups such as small cell lung cancer and non-small cell lung cancer (including adenocarcinoma, squamous cell carcinoma, and large cell carcinoma) based on pathological type. Breast cancer can be divided into different tumor subgroups such as HER2-positive and triple-negative breast cancer.

[0109] It is understandable that although different tumor subgroups belong to the same type of tumor, the clinical characteristics of the target subjects in each tumor subgroup are different, which may have different effects on the expression levels of candidate biomolecules during tumor development. Therefore, in this embodiment, for each tumor subgroup, the expression level of candidate biomolecules can be predicted using an expression prediction model. This allows for the targeted determination of the expression levels of candidate biomolecules in multiple stages of tumor development within the tumor subgroup. Based on the expression levels of candidate biomolecules in multiple stages of tumor development within the tumor subgroup, the importance detection result of the candidate biomolecule in the tumor cell subgroup can be determined. For ease of distinction, this importance detection result can be referred to as the second importance detection result.

[0110] The second importance detection result can indicate the importance of candidate biomolecules to tumor growth and development in a tumor subpopulation. For example, the second importance detection result can indicate whether a candidate biomolecule is a key or necessary element for tumor growth and development in a tumor subpopulation. In some exemplary embodiments, a target object within a tumor subpopulation can be identified, and then an expression prediction model can be used to predict the expression level of candidate biomolecules in the target object within that tumor subpopulation. The second importance detection result is determined based on the prediction result.

[0111] Furthermore, if the second importance test result meets the preset conditions, the drug target of the tumor subgroup can be determined based on the candidate biomolecule. For example, if the second importance test result indicates that the candidate biomolecule is important for tumor growth and development in the tumor subgroup (such as the candidate biomolecule having a sustained inducing property in the tumor subgroup), the second importance test result can be determined to meet the preset conditions.

[0112] For example, transcriptome sequencing results showed that the expression level of the TROP2 gene in lung adenocarcinoma tissue was significantly higher than that in normal lung tissue. Meanwhile, the expression level of the TROP2 gene in lung squamous cell carcinoma tissue was even higher than that in lung adenocarcinoma. Single-cell sequencing results showed that the expression level of the TROP2 gene in cells derived from normal lung tissue was very low, while many cells from lung adenocarcinoma and lung squamous cell carcinoma tissues expressed high levels of the TROP2 gene. Specifically, cells derived from lung squamous cell carcinoma expressed more TROP2 than cells derived from lung adenocarcinoma. Based on the target screening logic of related technologies, it would be assumed that the therapeutic effect of the TROP2 gene in lung squamous cell carcinoma might be better than that in adenocarcinoma. However, expression prediction models suggest that although the TROP2 gene is indeed more induced in squamous cell carcinoma, its expression level shows a decreasing trend in the later stages of lung squamous cell carcinoma development. Therefore, it can be predicted that the TROP2 gene is mainly effective against lung adenocarcinoma rather than lung squamous cell carcinoma. Clinical data also show that TROP2 gene-related ADC drugs can significantly prolong progression-free survival in patients with lung adenocarcinoma. However, although lung squamous cell carcinoma cells can express the TROP2 gene at higher levels, these patients do not benefit from TROP2 gene-related ADC drugs. Therefore, obtaining information on the expression and regulatory trends of existing ADC drug targets at different stages of cancer development can help predict whether corresponding target patients can benefit from current ADC drugs.

[0113] It is understood that, through this embodiment, when the importance of candidate biomolecules to the overall tumor is not obvious, by refining the expression level of candidate biomolecules in various tumor subgroups and determining the corresponding second importance detection results, it is possible to efficiently and quickly identify whether candidate biomolecules can serve as drug targets for specific tumor subgroups. Thus, for one or more tumor subgroups, the above processing method can refine and determine whether the target can benefit from the drug targets identified by candidate biomolecules, and more quickly and effectively identify drug targets applicable to specific subgroups. This provides a reference basis for the development of new targeted drugs or the promotion of existing targeted drugs, and helps to increase the types of selectable targeted drugs for target groups in specific tumor subgroups.

[0114] In an exemplary embodiment, in step S102, the expression level of candidate biomolecules is predicted by a trained expression prediction model, and the expression level of candidate biomolecules at multiple stages of tumor development is determined based on the prediction results. This may include the following steps:

[0115] S1021, Identify target individuals with tumors.

[0116] In a specific implementation, a specific target object suffering from a tumor can be identified, where the target object can be any object suffering from any type of tumor.

[0117] S1021 uses a trained expression prediction model to predict the expression level of candidate biomolecules in the target body, and determines the expression level of candidate biomolecules in multiple stages of tumor development in the target body based on the prediction results.

[0118] After identifying the target, the expression levels of candidate biomolecules in the target body can be predicted using an expression prediction model. This allows for the determination of the expression levels of candidate biomolecules at multiple stages of tumor development within the target body. Subsequently, the primary importance of candidate biomolecules to the tumor in the target body can be determined based on these expression levels.

[0119] In this embodiment, by identifying target subjects with tumors and determining the expression levels of candidate biomolecules in the target subjects at multiple stages of tumor development through an expression prediction model, it is possible to specifically identify the drug targets applicable to each target subject. This helps to quickly match suitable targeted drugs based on these drug targets. As a result, it is possible to quickly identify the suitable subjects for existing targeted drugs, further reducing the cost of clinical trials in the targeted drug development process, and also to quickly locate suitable targeted drugs for specific target subjects, thus gaining valuable treatment time.

[0120] In an exemplary embodiment, step S1021, predicting the expression level of candidate biomolecules in the target organism using a trained expression prediction model, may include the following steps:

[0121] The first transcriptome sequencing results for normal tissue of the target object and the second transcriptome sequencing results for tumor tissue of the target object are obtained. The first transcriptome sequencing results and the second transcriptome sequencing results are input into a trained expression prediction model. The expression prediction model predicts the evolution trajectory of the first transcriptome sequencing results into the second transcriptome sequencing results, thereby obtaining the transcriptome change trajectory in the target object. Based on the transcriptome change trajectory of the candidate biomolecules extracted from the transcriptome change trajectory in the target object, the predicted expression level of the candidate biomolecules in the target object is determined.

[0122] In practice, the tumor development process can also be called the tumorigenesis process, which refers to the transformation process from normal tissue to cancerous tissue. In some examples, the normal tissue transcriptome can be defined as the average of the set of transcriptomes of normal tissue corresponding to normal sample objects or adjacent normal tissue of abnormal sample objects with tumors; the cancerous tissue transcriptome can be defined as the average of the set of transcriptomes of cancerous tissue of abnormal sample objects with tumors.

[0123] In this embodiment, normal tissue and tumor tissue of the target object can be obtained. Normal tissue can be tissue that does not contain diseased tissue or tumor cells. In one example, cancerous tissue and adjacent normal tissue of the target object can be obtained through biopsy sampling, which are respectively used as the tumor tissue and normal tissue of the target object.

[0124] Then, transcriptome sequencing can be performed on normal tissues and tumor tissues separately to obtain transcriptome sequencing results for normal tissues and tumor tissues. For ease of distinction, the transcriptome sequencing results for normal tissues are referred to as the first transcriptome sequencing results, and the transcriptome sequencing results for tumor tissues are referred to as the second transcriptome sequencing results. The transcriptome can be understood as the collection of transcribed products within cells (such as normal tissue cells or tumor tissue cells in this embodiment) under specified physiological conditions, and may include messenger RNA (mRNA), ribosomal RNA, transfer RNA, and non-coding RNA; or it can be understood as the collection of all mRNAs. Transcriptome sequencing can determine gene expression within cells.

[0125] After obtaining the first and second transcriptome sequencing results, an expression prediction model can be used to predict the evolutionary trajectory from the first to the second transcriptome sequencing results, thus obtaining the transcriptome change trajectory within the target organism. Since this transcriptome change trajectory includes transcriptome information corresponding to multiple genes, the transcriptome change trajectory corresponding to candidate biomolecules can be extracted from it. For example, if the candidate biomolecule is a target gene, the transcriptome change trajectory related to that target gene can be obtained; if the candidate biomolecule is a protein, the transcriptome change trajectory related to the expression of that protein can be obtained. Because the transcriptome change trajectory corresponding to a candidate biomolecule can characterize the regulatory trajectory (or regulatory mechanism) of the candidate biomolecule, the predicted expression level of the candidate biomolecule in the target organism can be determined based on the transcriptome change trajectory corresponding to the candidate biomolecule.

[0126] In this embodiment, by obtaining the first transcriptome sequencing results corresponding to normal tissue and the second transcriptome sequencing results corresponding to tumor tissue, and then using the expression prediction model to predict the change trajectory between the two, the expression trend of the transcriptome trajectory of candidate biomolecules in the target object can be efficiently and accurately deduced. This helps to determine whether candidate biomolecules benefit the target object in a targeted manner, thereby effectively improving the identification effectiveness, accuracy and efficiency when identifying drug targets for different target objects.

[0127] In an exemplary embodiment, predicting the evolutionary trajectory from a first transcriptome sequencing result to a second transcriptome sequencing result to obtain the transcriptome change trajectory within the target subject may include the following steps:

[0128] The starting point of the change trajectory is determined based on the first transcriptome sequencing results, and the ending point of the change trajectory is determined based on the second transcriptome sequencing results; the transcriptome sequencing results corresponding to the intermediate points of the change trajectory between the starting point and the ending point of the change trajectory are predicted, and the transcriptome change trajectory in the target object is obtained based on the prediction results.

[0129] In specific implementation, if one or more target objects have corresponding first transcriptome sequencing results and second transcriptome sequencing results, the first transcriptome sequencing results and second transcriptome sequencing results of each target object can be paired and input into the trained expression prediction model. For example, when the expression prediction model is a generative artificial intelligence model for generating transcriptomes, the first transcriptome sequencing results and second transcriptome sequencing results can be paired and flipped into the model.

[0130] After acquiring the paired inputs of the first and second transcriptome sequencing results, the model can determine the feature points corresponding to the first transcriptome sequencing result as the starting point of the trajectory change in the model's feature space (also known as the model's latent space), and the feature points corresponding to the second transcriptome sequencing result as the ending point of the trajectory change. Then, multiple intermediate points can be set between the starting and ending points of the trajectory change, and the transcriptome sequencing results corresponding to each intermediate point can be predicted. This allows the model to obtain the dynamic transcriptome change trajectory of each target object during tumor development. For example, the number of intermediate points to be set between the starting and ending points of the trajectory change can be determined as x. Then, x feature points can be set between the starting and ending points of the trajectory change as intermediate points of the trajectory change with the same step size. The model then outputs the transcriptome corresponding to these intermediate points in the feature space, thereby obtaining the personalized transcriptome transition process of each target object. Subsequently, based on the transcriptome data derived by the model, candidate biomolecules, such as the regulatory trajectory of the CEACAM5 gene in the tumor development and progression of the target object, can be extracted.

[0131] In this embodiment, by predicting the transcriptome sequencing results corresponding to multiple intermediate points of the change trajectory between the starting point and the ending point of the change trajectory, corresponding transcriptome change trajectories can be constructed for different target objects, providing a basis for subsequently determining suitable drug targets for each target object.

[0132] In an exemplary embodiment, step S103, determining the first importance detection result of the candidate biomolecule in the tumor development process based on the expression level of the candidate biomolecule at multiple stages of tumor development, may include the following steps:

[0133] Based on the expression levels of candidate biomolecules at multiple stages of tumor development, the persistence of expression of candidate biomolecules in the later stages of tumor development is determined; when the persistence of expression reaches a preset persistence level, the first importance detection result indicating the importance of candidate biomolecules in the process of tumor development is obtained.

[0134] In practical applications, the expression persistence of candidate biomolecules in later stages of tumor development can be determined based on their expression levels at multiple stages of tumor development. This persistence refers to the extent to which the candidate biomolecule can be continuously expressed, even in later stages of tumor progression. In other words, this embodiment uses the expression levels of candidate biomolecules at multiple stages of tumor development to determine whether the candidate biomolecule can be expressed sustainably for a long period and whether it can still be expressed in later stages of tumor progression.

[0135] When the expression persistence reaches a preset persistence level, the candidate biomolecule can be identified as a key factor in tumor growth and development, thus obtaining the first importance detection result indicating the importance of the candidate biomolecule in the tumor development process.

[0136] For example, single-cell transcriptome sequencing results indicate that the CD19 gene is mainly expressed in B cells of tumor patients, suggesting that the CD19 gene may be a suitable target for ADC drugs. In some embodiments, single-cell transcriptome sequencing results are... Figure 3A and Figure 3B The transcriptome image shown, Figure 3A CD19 gene expression levels in single-cell transcriptomes of blood samples from individuals with BCL (bacillary cystic keratitis), where cells... Figure 3A The position within can be calculated using nonlinear dimensionality reduction algorithms, such as the Uniform Manifold Approximation and Projection (UMAP) algorithm. Figure 3B This shows the location of the B cell population in UMAP. (Comparison) Figure 3A and Figure 3B It can be known that the CD19 gene is a B cell-specific gene.

[0137] Similarly, transcriptome sequencing data showed that the CEACAM5 gene was significantly elevated in tumor tissue (non-small cell lung cancer) compared to normal tissue, for example... Figure 4 As shown, the expression levels of the CEACAM5 gene in normal tissues (including normal samples and adjacent normal tissues of non-small cell lung cancer samples) and cancerous tissues of abnormal samples with tumors are illustrated; simultaneously, Figure 5A The transcriptome image shown illustrates the single-cell expression level of the CEACAM5 gene in lung cancer tissue. Figure 5B The location of tumor cell clusters in UMAP is shown. Comparison Figure 5A and Figure 5B It is known that the CEACAM5 gene is expressed at a higher level in tumor cells of non-small cell lung cancer tissue, meaning that the CEACAM5 gene is mainly expressed at a high level in cancer cells, and therefore may also be a suitable target for ADC drugs.

[0138] However, although both CD19 and CEACAM5 genes are induced during the development and progression of their respective tumors, expression prediction models can predict, for example... Figure 6 The changes in target gene expression levels during tumorigenesis are shown. The horizontal axis, 0-100, represents the distance between two points in the model's feature space from the normal tissue transcriptome to the adjacent normal tissue transcriptome, each divided into 100 parts. Figure 6 This study investigated the changes in CEACAM5 gene expression levels during the development of non-small cell lung cancer (NSCLC) and CD19 expression levels during B-cell lymphoma. A significant difference in regulatory trends was observed: CD19 expression remained high after induction during B-cell lymphoma development, while CEACAM5 expression was downregulated in the later stages of NSCLC, suggesting that CEACAM5 is no longer a necessary gene for the subsequent development of lung cancer. Related clinical studies have also shown that ADC drugs associated with the CD19 gene can significantly prolong the survival of patients with B-cell lymphoma, while ADC drugs associated with the CEACAM5 gene are less likely to achieve the same effect.

[0139] For example, Figure 7A This illustrates the difference in overall survival (OS) between patients using CD19 gene-related ADC drugs and those not using CD19 gene-related ADC drugs. Figure 7A It is evident that CD19 antibody-drug conjugate therapy significantly improves overall survival in patients who have relapsed after chemotherapy and immunotherapy, indicating the effectiveness of CD19 antibody-drug conjugates in treating relapsed B-cell lymphoma; on the other hand, Figure 7B The study shows the progression-free survival (PFS) of patients using CEACAM5 gene-related antibody-drug conjugates (ADCs) and those not using CEACAM5 gene-related ADCs. The comparison reveals that CEACAM5 antibody-drug conjugates do not delay the progression of non-small cell lung cancer. Therefore, genes that can be persistently induced at high levels in later stages of tumor development are more suitable as drug targets for ADCs.

[0140] In this embodiment, by determining the degree of expression persistence of candidate biomolecules in the late stage of tumor development, and when the degree of expression persistence reaches a preset degree of persistence, a first importance detection result indicating the importance of candidate biomolecules in tumor development is obtained. This allows the target identification device to identify candidate biomolecules that are continuously expressed in the tumor. Even if the expression level of the candidate biomolecule at a certain static moment is not very high or low, for example, greater than or equal to the first expression level threshold required when the targeted drug just takes effect, or less than the second expression level threshold corresponding to a high expression level, the first importance detection result of the candidate biomolecule meets the preset conditions. This means that the candidate biomolecule can be expressed in multiple stages of tumor development and can bind to the targeted drug relatively stably during treatment. Thus, effective drug targets can be quickly identified in tumor cells.

[0141] It is understood that the various embodiments provided above in this application can be combined in any way. For example, for a specified target tumor, any candidate biomolecule can be selected and detected in the tumor cells of the target tumor to determine a suitable drug target for the target tumor. Subsequently, a related drug can be developed specifically for the drug target. On this basis, the second importance detection results of the candidate biomolecule in each tumor subgroup of the target tumor can also be determined to determine a suitable drug target for the tumor subgroup and develop a related targeted drug.

[0142] For example, for known targeted drugs and their corresponding candidate biomolecules, the first or second importance of the candidate biomolecule in various tumors can be determined, and the various indications of the developed targeted drugs and the appropriate drug targets in specific tumor types can be identified.

[0143] In addition, for a target subject with a tumor, the first or second most important test results of various candidate biomolecules on the target subject can be determined (e.g., the tumor of the target subject belongs to a certain tumor subgroup). Based on the candidate biomolecule, a targeted drug that can benefit the target subject can be identified. For example, when the target gene can be continuously induced in the target subject, especially when it continues to induce in the later stages of tumor development, it can be determined that the target subject can benefit from the targeted drug related to the target gene.

[0144] In one embodiment, this application also provides a target identification method. It is understood that this method can be applied to a terminal, a server, or a system including both a terminal and a server. In this embodiment, the method includes the following steps:

[0145] S201 identifies candidate biomolecules in tumor cells that can provide targets for targeted drugs.

[0146] S202 uses a trained expression prediction model to predict the expression levels of candidate biomolecules and determines their expression levels at multiple stages of tumor development based on the prediction results.

[0147] S203, based on the expression levels of candidate biomolecules at multiple stages of tumor development, determines the primary importance of candidate biomolecules in the tumor development process.

[0148] S204. Based on the tumor-specific detection results and first importance detection results of the candidate biomolecules, determine whether to identify the drug target based on the candidate biomolecules.

[0149] The specific implementation of the above method can be found in one or more of the embodiments described above, and will not be repeated here.

[0150] In the aforementioned target identification method, candidate biomolecules that can provide targets for targeted drugs in tumor cells can be identified. Then, a trained expression prediction model predicts the expression levels of these candidate biomolecules, determining their expression levels at multiple stages of tumor development. Subsequently, based on the expression levels at these stages, the first importance of the candidate biomolecules in tumor development can be determined. Finally, based on the tumor-specific and first importance results of the candidate biomolecules, it is determined whether a drug target should be identified. In this embodiment, by predicting the expression levels of candidate biomolecules that can provide targets for targeted drugs, the dynamic expression trends of these biomolecules in tumors can be revealed based on their expression levels at multiple stages of tumor development. This allows for a more accurate assessment of whether candidate biomolecules can stably and specifically provide targets in tumors over a long period, effectively improving the identification effectiveness of the final drug targets and enabling precise screening of reliable drug targets for targeted drugs.

[0151] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0152] Based on the same inventive concept, this application also provides a target identification device for implementing the target identification method described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more target identification device embodiments provided below can be found in the limitations of the target identification method described above, and will not be repeated here.

[0153] In one exemplary embodiment, such as Figure 8 As shown, a target identification device is provided, comprising:

[0154] Candidate molecule identification module 801 is used to identify candidate biomolecules in tumor cells that can provide targets for targeted drugs.

[0155] The expression prediction module 802 is used to predict the expression level of the candidate biomolecule by a trained expression prediction model, and determine the expression level of the candidate biomolecule at multiple stages of tumor development based on the prediction results.

[0156] The first importance determination module 803 is used to determine the first importance detection result of the candidate biomolecule in the tumor development process based on the expression level of the candidate biomolecule in multiple stages of tumor development.

[0157] The target identification module 804 is used to determine whether to identify a drug target based on the candidate biomolecule, according to the tumor-specific detection results and the first importance detection results of the candidate biomolecule.

[0158] In one embodiment, the candidate molecule determination module 801 is configured to:

[0159] Obtain a pre-defined target drug and identify drug targets in tumor cells that can bind to the target drug and / or target genes corresponding to the target drug;

[0160] Based on the drug target and / or the target gene, candidate biomolecules to be screened are obtained.

[0161] In one embodiment, the first importance detection result includes the first importance detection result of the candidate biomolecule in the tumor development process of at least one tumor;

[0162] The target identification module 804 is used for:

[0163] For each type of tumor, if the tumor-specific detection results of the candidate biomolecules and the first importance detection results meet preset conditions, the drug target of the tumor is determined based on the candidate biomolecules.

[0164] In one embodiment, the target identification module 804 is used for:

[0165] If the candidate biomolecule includes a drug target that can bind to the target drug, the drug target is identified as the drug target of the target drug in the tumor.

[0166] When the candidate biomolecule includes the target gene corresponding to the target drug, the drug target in the tumor is obtained based on the expression results of the target gene.

[0167] In one embodiment, the first importance detection result includes the first importance detection result of the candidate biomolecule in the tumor development process of at least one tumor;

[0168] The target identification module 804 is used for:

[0169] For each of the tumors, if the tumor-specific detection results of the candidate biomolecules meet preset conditions and the first importance detection results do not meet preset conditions, at least one tumor subgroup of the tumor is determined.

[0170] For each of the tumor subgroups, the expression prediction model predicts the expression level of the candidate biomolecules, and the expression level of the candidate biomolecules in multiple developmental stages of the tumor subgroup is determined based on the prediction results. Based on the expression level of the candidate biomolecules in multiple developmental stages of the tumor subgroup, the second importance detection result of the candidate biomolecules in the tumor subgroup is obtained.

[0171] If the second importance detection result meets the preset conditions, the drug targets of the tumor subgroup are determined based on the candidate biomolecules.

[0172] In one embodiment, the expression prediction module 802 is configured to:

[0173] Identify target individuals with cancer;

[0174] The expression levels of the candidate biomolecules in the target body are predicted by a trained expression prediction model, and the expression levels of the candidate biomolecules in the target body at multiple stages of tumor development are determined based on the prediction results.

[0175] In one embodiment, the expression prediction module 802 is configured to:

[0176] Obtain the first transcriptome sequencing results corresponding to the normal tissue of the target object and the second transcriptome sequencing results corresponding to the tumor tissue of the target object;

[0177] The first transcriptome sequencing results and the second transcriptome sequencing results are input into a trained expression prediction model. The expression prediction model predicts the evolution trajectory of the first transcriptome sequencing results into the second transcriptome sequencing results, thereby obtaining the transcriptome change trajectory in the target object.

[0178] Based on the transcriptome change trajectory extracted from the transcriptome change trajectory, the predicted expression level of the candidate biomolecule in the target body is determined.

[0179] In one embodiment, the expression prediction module 802 is configured to:

[0180] The starting point of the change trajectory is determined based on the first transcriptome sequencing results, and the ending point of the change trajectory is determined based on the second transcriptome sequencing results;

[0181] The transcriptome sequencing results corresponding to each of the multiple intermediate points of the change trajectory between the starting point and the ending point of the change trajectory are predicted, and the transcriptome change trajectory in the target object is obtained based on the prediction results.

[0182] In one embodiment, the first importance determination module 803 is configured to:

[0183] Based on the expression levels of the candidate biomolecules at multiple stages of tumor development, the persistence of expression of the candidate biomolecules in the later stages of tumor development is determined.

[0184] When the expression persistence reaches a preset persistence level, a first importance detection result indicating the importance of the candidate biomolecule in tumor development is obtained.

[0185] Each module in the aforementioned target identification device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0186] In one exemplary embodiment, a target identification device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 9 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores candidate biological factor data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a target identification method.

[0187] In one exemplary embodiment, another target identification device is provided. This computer device can be a terminal, and its internal structure diagram can be as follows: Figure 10As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a target identification method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0188] Those skilled in the art will understand that Figure 9 and Figure 10 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0189] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0190] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0191] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0192] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0193] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0194] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0195] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A target identification device, characterized in that, The target identification device is configured to perform the following steps: Identify candidate biomolecules in tumor cells that can provide targets for targeted drugs; The expression levels of the candidate biomolecules are predicted by a trained expression prediction model, and the expression levels of the candidate biomolecules at multiple stages of tumor development are determined based on the prediction results. Based on the expression levels of the candidate biomolecules at multiple stages of tumor development, the persistence of expression of the candidate biomolecules in the later stages of tumor development is determined. When the expression persistence reaches a preset persistence level, a first importance detection result indicating the importance of the candidate biomolecule in tumor development is obtained; Based on the tumor-specific detection results of the candidate biomolecules and the first importance detection results, determine whether to identify drug targets based on the candidate biomolecules.

2. The device according to claim 1, characterized in that, The identification of candidate biomolecules to be recognized in tumor cells includes: Obtain a pre-defined target drug and identify drug targets in tumor cells that can bind to the target drug and / or target genes corresponding to the target drug. Based on the drug target and / or the target gene, candidate biomolecules to be screened are obtained.

3. The device according to claim 1 or 2, characterized in that, The first importance detection result includes the first importance detection result of the candidate biomolecule in the tumor development process of at least one tumor; The step of determining whether to identify a drug target based on the candidate biomolecule, according to the tumor-specific detection results and the first importance detection results, includes: For each type of tumor, if the tumor-specific detection results of the candidate biomolecules and the first importance detection results meet preset conditions, the drug target of the tumor is determined based on the candidate biomolecules.

4. The device according to claim 3, characterized in that, The step of determining the drug target of the tumor based on the candidate biomolecules includes: If the candidate biomolecule includes a drug target that can bind to the target drug, the drug target is identified as the drug target of the target drug in the tumor. When the candidate biomolecule includes the target gene corresponding to the target drug, the drug target in the tumor is obtained based on the expression results of the target gene.

5. The device according to claim 1, characterized in that, The first importance detection result includes the first importance detection result of the candidate biomolecule in the tumor development process of at least one tumor; The step of determining whether to identify a drug target based on the candidate biomolecule, according to the tumor-specific detection results and the first importance detection results, includes: For each of the tumors, if the tumor-specific detection results of the candidate biomolecules meet preset conditions and the first importance detection results do not meet preset conditions, at least one tumor subgroup of the tumor is determined. For each of the tumor subgroups, the expression prediction model predicts the expression level of the candidate biomolecules, and the expression level of the candidate biomolecules in multiple developmental stages of the tumor subgroup is determined based on the prediction results. Based on the expression level of the candidate biomolecules in multiple developmental stages of the tumor subgroup, the second importance detection result of the candidate biomolecules in the tumor subgroup is obtained. If the second importance detection result meets the preset conditions, the drug targets of the tumor subgroup are determined based on the candidate biomolecules.

6. The device according to claim 1, characterized in that, The process of predicting the expression levels of the candidate biomolecules using a trained expression prediction model, and determining the expression levels of the candidate biomolecules at multiple stages of tumor development based on the prediction results, includes: Identify target individuals with cancer; The expression levels of the candidate biomolecules in the target body are predicted by a trained expression prediction model, and the expression levels of the candidate biomolecules in the target body at multiple stages of tumor development are determined based on the prediction results.

7. The device according to claim 6, characterized in that, The prediction of the expression level of the candidate biomolecule in the target organism by the trained expression prediction model includes: Obtain the first transcriptome sequencing results corresponding to the normal tissue of the target object and the second transcriptome sequencing results corresponding to the tumor tissue of the target object; The first transcriptome sequencing results and the second transcriptome sequencing results are input into a trained expression prediction model. The expression prediction model predicts the evolution trajectory of the first transcriptome sequencing results into the second transcriptome sequencing results, thereby obtaining the transcriptome change trajectory in the target object. Based on the transcriptome change trajectory extracted from the transcriptome change trajectory, the predicted expression level of the candidate biomolecule in the target body is determined.

8. The device according to claim 7, characterized in that, The step of predicting the evolutionary trajectory from the first transcriptome sequencing result to the second transcriptome sequencing result, and obtaining the transcriptome change trajectory in the target object, includes: The starting point of the change trajectory is determined based on the first transcriptome sequencing results, and the ending point of the change trajectory is determined based on the second transcriptome sequencing results; The transcriptome sequencing results corresponding to each of the multiple intermediate points of the change trajectory between the starting point and the ending point of the change trajectory are predicted, and the transcriptome change trajectory in the target object is obtained based on the prediction results.

9. A target identification method, characterized in that, The method includes: Identify candidate biomolecules in tumor cells that can provide targets for targeted drugs; The expression levels of the candidate biomolecules are predicted by a trained expression prediction model, and the expression levels of the candidate biomolecules at multiple stages of tumor development are determined based on the prediction results. Based on the expression levels of the candidate biomolecules at multiple stages of tumor development, the persistence of expression of the candidate biomolecules in the later stages of tumor development is determined. When the expression persistence reaches a preset persistence level, a first importance detection result indicating the importance of the candidate biomolecule in tumor development is obtained; Based on the tumor-specific detection results of the candidate biomolecules and the first importance detection results, determine whether to identify drug targets based on the candidate biomolecules.

10. A target identification device, characterized in that, The device includes: The candidate molecule identification module is used to identify candidate biomolecules in tumor cells that can provide targets for targeted drugs. An expression prediction module is used to predict the expression level of the candidate biomolecules by a trained expression prediction model, and to determine the expression level of the candidate biomolecules at multiple stages of tumor development based on the prediction results. The first importance determination module is used to determine the degree of expression persistence of the candidate biomolecule in the later stages of tumor development based on the expression level of the candidate biomolecule in multiple stages of tumor development; and when the degree of expression persistence reaches a preset degree of persistence, a first importance detection result indicating the importance of the candidate biomolecule in the process of tumor development is obtained. The target identification module is used to determine whether to identify a drug target based on the candidate biomolecule, according to the tumor-specific detection results and the first importance detection results.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method of claim 9.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method of claim 9.

Images