A research method and system for an anti-tumor active ingredient

Abstract

This invention relates to the field of biomedical technology and discloses a research method and system for antitumor active ingredients. The method includes: standardizing the structural data and in vitro bioactivity data of antitumor compounds to obtain standard compound data and standardized bioactivity data; analyzing the correlation between the structural characteristics and activity of antitumor compounds, and classifying the antitumor compounds by efficacy based on the correlation to obtain a preliminary efficacy benchmark; performing multidimensional efficacy evaluation and ranking of antitumor compounds to obtain a candidate compound ranking list; identifying highly active candidate compounds and conducting in vivo pharmacodynamic tests on the highly active candidate compounds to obtain in vivo pharmacodynamic data; performing mechanism-of-action analysis on the highly active candidate compounds to obtain preliminary mechanism data; and comprehensively evaluating the in vivo pharmacodynamic data and preliminary mechanism data to obtain a preferred evaluation report. This invention can improve the accuracy of research on antitumor active ingredients.

Description

Technical Field

[0001] This invention relates to the field of biomedical technology, and in particular to a research method and system for an anti-tumor active ingredient. Background Technology

[0002] Existing technologies have significant shortcomings in the data processing and structure-activity relationship analysis of antitumor compounds. The structural and in vitro bioactivity data of antitumor compounds are not systematically and standardized; instead, raw data or data processed from a single dimension are simply used. This results in inconsistent data formats, a lack of comparability of activity indicators, and an inability to provide high-quality data support for subsequent research. Furthermore, the correlation between structural features and activity is not deeply analyzed; compound screening relies solely on experience or simple statistics, making it difficult to accurately extract structure-activity relationship patterns and scientifically calculate the contribution weight of core structural features to activity. This leads to a lack of basis for efficacy classification, inaccurate initial efficacy benchmarks, and difficulty in identifying potential antitumor compounds.

[0003] Existing technologies lack a multi-dimensional efficacy evaluation index system, ranking compounds solely based on a single activity dimension. This neglects key factors such as structural novelty, drug-likeness, and synthetic feasibility, resulting in an insufficiently rational candidate compound ranking list. Furthermore, dynamic threshold selection and targeted in vivo pharmacodynamic testing are not conducted based on the ranking list; only batch testing is performed, leading to significant resource waste and low testing efficiency. Mechanism-of-action analysis of highly active candidate compounds is not performed, focusing only on the efficacy data itself, failing to identify the compound's target and signaling pathways, and hindering the correlation between effects and mechanisms. Finally, in vivo efficacy data and preliminary mechanism exploration data are not comprehensively evaluated; evaluation reports are generated based on only a single data dimension, resulting in biased evaluation results that cannot provide a comprehensive and reliable reference for further research and development of anti-tumor active ingredients. Summary of the Invention

[0004] This invention provides a method and system for researching antitumor active ingredients to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for researching antitumor active ingredients, comprising: S1. The structural data and in vitro bioactivity data of the antitumor compound are standardized to obtain the standard compound data and standardized bioactivity data of the antitumor compound; S2. Based on standard compound data and standardized bioactivity data, analyze the correlation between the structural characteristics and activity of the antitumor compounds, and classify the efficacy of the antitumor compounds based on the correlation to obtain the initial efficacy benchmark of the antitumor compounds; S3. Based on the preliminary efficacy criteria, the antitumor compounds are evaluated and ranked in a multidimensional efficacy manner to obtain a candidate compound ranking list of the antitumor compounds; S4. Based on the candidate compound ranking list, determine the highly active candidate compounds among the antitumor compounds, and perform in vivo pharmacodynamic tests on the highly active candidate compounds to obtain the in vivo pharmacodynamic data of the antitumor compounds; S5. Based on the in vivo efficacy data, perform mechanism-of-action analysis on the highly active candidate compounds to obtain preliminary data on the mechanism of the antitumor compounds; S6. The in vivo efficacy data and the preliminary mechanism data are comprehensively evaluated to obtain a preferred evaluation report of the antitumor compound.

[0006] In a preferred embodiment, the structural data and in vitro bioactivity data of the antitumor compound are normalized to obtain standard compound data and standardized bioactivity data of the antitumor compound, including: To obtain raw structural and biological activity data of antitumor compounds; Extract the characteristic structural data from the original structural data; The original bioactivity data were normalized to obtain the normalized activity data of the antitumor compound. The characterized structural data and the normalized activity data are processed to unify the format, resulting in standard compound data and standardized biological activity data of the antitumor compound.

[0007] In a preferred embodiment, the step of analyzing the correlation between the structural features and activities of the antitumor compounds based on standard compound data and standardized bioactivity data, and classifying the antitumor compounds by efficacy based on the correlation to obtain the preliminary efficacy criteria for the antitumor compounds, includes: Key structural features are extracted from the standard compound data, and activity response data corresponding to the key structural features are matched from the standardized bioactivity data. Structure-activity relationship pattern extraction was performed on the key structural features and the activity response data to obtain the structure-activity relationship pattern spectrum of the antitumor compound; Based on the structure-activity relationship pattern spectrum, the contribution weight of the core structural feature pattern in the antitumor compound to the activity is calculated, and the efficacy gradient of the antitumor compound is divided according to the contribution weight to obtain the efficacy gradient division result of the antitumor compound. Based on the efficacy gradient division results, a preliminary efficacy classification framework for the antitumor compounds is constructed, and the efficacy gradient division results are mapped to the preliminary efficacy classification framework to obtain the preliminary efficacy benchmark for the antitumor compounds.

[0008] In a preferred embodiment, the formula for calculating the contribution weight is: ; In the formula, For the first The weight value of each object, For the first The characteristic index values ​​of an object For the first Another characteristic indicator value of an object It is an exponential adjustment parameter. For the first The characteristic index values ​​of an object This is the sum of the characteristic index values ​​of all objects.

[0009] In a preferred embodiment, the step of performing a multidimensional efficacy evaluation and ranking of the antitumor compounds based on the initial efficacy benchmark to obtain a candidate compound ranking list of the antitumor compounds includes: Based on the efficacy categories and efficacy levels in the preliminary efficacy benchmarks, a multidimensional efficacy evaluation index system for the antitumor compounds is constructed. Based on the aforementioned multidimensional efficacy evaluation index system, the antitumor compound was independently evaluated to obtain the structural novelty evaluation results, drug resistance evaluation results, and synthesis feasibility evaluation results of the antitumor compound. The structural novelty assessment results, the drug-likeness assessment results, and the synthetic feasibility assessment results are fused together from multiple dimensions to obtain the comprehensive efficacy profile of the antitumor compound. Based on the aforementioned multidimensional efficacy evaluation index system, the efficacy of the initially selected efficacy benchmark is evaluated, and a reference benchmark efficacy profile of the antitumor compound is generated. The comprehensive efficacy profile is compared with the reference baseline efficacy profile, and the antitumor compounds are prioritized according to the comparison results to obtain a candidate compound ranking list of the antitumor compounds.

[0010] In a preferred embodiment, the step of constructing a multidimensional efficacy evaluation index system for the antitumor compound based on the efficacy category and efficacy level in the preliminary efficacy benchmark includes: The initial performance benchmark is parsed to obtain the performance category information and performance level information of the initial performance benchmark; Based on the efficacy category information and the efficacy level information, a multi-dimensional evaluation index for the antitumor compound is constructed; Based on the multi-dimensional evaluation indicators, the evaluation criteria for the multi-dimensional evaluation indicators are determined; By integrating the multi-dimensional evaluation indicators and the evaluation criteria, a multi-dimensional efficacy evaluation indicator system for the anti-tumor compound is obtained. The multidimensional efficacy evaluation index system is applied to the antitumor compound to obtain quantitative scoring data of the antitumor compound.

[0011] In a preferred embodiment, the step of determining highly active candidate compounds among the antitumor compounds according to the candidate compound ranking list, and performing in vivo pharmacodynamic testing on the highly active candidate compounds to obtain in vivo pharmacodynamic data of the antitumor compounds includes: The priority ranking logic and performance profile features of the candidate compound ranking list are analyzed, and a dynamic selection threshold based on the difference between the ranking position and the performance profile is set. Based on the dynamic selection threshold, highly active candidate compounds are selected from the candidate compound ranking list; Based on the highly active candidate compounds, an in vivo pharmacodynamic testing protocol adapted to the structural characteristics and expected mechanism of action of the highly active candidate compounds was determined. Based on the aforementioned in vivo pharmacodynamic testing protocol, the highly active candidate compound was tested in vivo to obtain the original pharmacodynamic observation data of the antitumor compound. The original pharmacodynamic observation data were quantified pharmacodynamically to obtain the in vivo pharmacodynamic data of the antitumor compound.

[0012] In a preferred embodiment, the step of performing mechanism-directed analysis on the candidate compound based on the in vivo efficacy data to obtain preliminary mechanism-related data of the antitumor compound includes: The in vivo pharmacodynamic data are deconstructed using pharmacodynamic response patterns to obtain pharmacodynamic response dimension data of the antitumor compound; The pharmacodynamic response dimension data is correlated with the core structural feature pattern and the efficacy gradient division result in a multidimensional correlation mapping to obtain stable correlation pairs of the antitumor compounds. Based on the stable correlation pairs, a preliminary mechanistic hypothesis for the antitumor compound is constructed; Potential targets that match the structural features in the preliminary mechanism hypothesis are screened out. By integrating the preliminary mechanism hypothesis with the potential target, preliminary data on the mechanism of the antitumor compound are obtained.

[0013] In a preferred embodiment, the step of comprehensively evaluating the in vivo efficacy data and the preliminary mechanism data to obtain the preferred evaluation report of the antitumor compound includes: The in vivo efficacy data were subjected to antitumor effect feature extraction to obtain the tumor inhibition intensity characteristics, tumor regression rate characteristics, and survival benefit characteristics of the antitumor compound; Biological significance analysis was performed on the preliminary data of the mechanism described above to obtain the core target intervention characteristics, signal pathway regulation characteristics, and potential off-target risk characteristics of the anti-tumor compound. The effect-mechanism matching degree matrix of the antitumor compound is obtained by performing consistency matching analysis on the tumor inhibition intensity characteristics, the tumor regression rate characteristics, the survival benefit characteristics, the core target intervention characteristics, the signaling pathway regulation characteristics, and the potential off-target risk characteristics. Based on the effect-mechanism matching matrix, the comprehensive advantage assessment and risk trade-off of the highly active candidate compounds are performed to obtain a comprehensive judgment conclusion on the candidate compounds of the antitumor compounds. Based on the comprehensive evaluation of the candidate compounds, a preferred evaluation report of the antitumor compounds is generated.

[0014] To address the above problems, the present invention also provides a research system for antitumor active ingredients, the system comprising: The data standardization module is used to standardize the structural data and in vitro bioactivity data of the antitumor compounds to obtain standard compound data and standardized bioactivity data of the antitumor compounds. The structure-activity relationship efficacy screening module is used to analyze the correlation between the structural features and activities of the antitumor compounds based on standard compound data and standardized bioactivity data, and to classify the efficacy of the antitumor compounds based on the correlation to obtain the initial efficacy benchmark of the antitumor compounds. The multidimensional efficacy ranking module is used to perform multidimensional efficacy evaluation and ranking of the antitumor compounds based on the initial efficacy benchmark, and obtain a candidate compound ranking list of the antitumor compounds. The candidate compound verification module is used to determine the highly active candidate compounds among the antitumor compounds according to the candidate compound ranking list, and to perform in vivo pharmacodynamic tests on the highly active candidate compounds to obtain in vivo pharmacodynamic data of the antitumor compounds. The mechanism-oriented analysis module is used to perform mechanism-oriented analysis on the candidate compounds based on the in vivo efficacy data, and obtain preliminary mechanism exploration data of the antitumor compounds. The comprehensive evaluation and selection module is used to comprehensively evaluate the in vivo efficacy data and the preliminary mechanism data to obtain an optimal evaluation report of the antitumor compound.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention lays a precise foundation for the study of antitumor active ingredients through standardized data processing and structure-activity relationship analysis. The structural data and in vitro bioactivity data of antitumor compounds are standardized, characteristic structural data are extracted, and activity indicators are normalized to form standard data. The correlation between structural features and activity is analyzed in depth, structure-activity relationship patterns are extracted, and the contribution weight of core structural features is calculated. Efficacy gradients are divided and classified, and preliminary efficacy benchmarks are established, providing a scientific basis for subsequent screening.

[0016] 2. This invention significantly improves the efficiency and reliability of screening antitumor active ingredients through multidimensional evaluation and comprehensive assessment. A multidimensional efficacy evaluation index system is constructed based on initial efficacy criteria, comprehensively evaluating and ranking compounds from dimensions such as structural novelty, drug-likeness, and synthetic feasibility to accurately screen highly active candidate compounds. In vivo pharmacodynamic testing verifies activity, and the mechanism of action is analyzed in conjunction with pharmacodynamic data. A comprehensive evaluation report is generated by integrating in vivo pharmacodynamic data with preliminary mechanism exploration data, providing efficient and precise technical support for the research and development of antitumor active ingredients. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a research method for an antitumor active ingredient according to an embodiment of the present invention. Figure 2 This is a functional block diagram of a research system for antitumor active ingredients provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for researching antitumor active ingredients. The execution subject of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for researching antitumor active ingredients can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating a method for researching an antitumor active ingredient according to an embodiment of the present invention. In this embodiment, the method for researching an antitumor active ingredient includes: S1. The structural data and in vitro bioactivity data of the antitumor compound are standardized to obtain the standard compound data and standardized bioactivity data of the antitumor compound; In this embodiment of the invention, the structural data and in vitro bioactivity data of the antitumor compound are standardized to obtain standard compound data and standardized bioactivity data of the antitumor compound, including: To obtain raw structural and biological activity data of antitumor compounds; Extract the characteristic structural data from the original structural data; The original bioactivity data were normalized to obtain the normalized activity data of the antitumor compound. The characterized structural data and the normalized activity data are processed to unify the format, resulting in standard compound data and standardized biological activity data of the antitumor compound.

[0021] We collect relevant raw data on antitumor compounds through professional chemical database searches and extraction of experimental test records. The raw structural data includes molecular structural information such as the atomic composition, chemical bond type, atomic spatial coordinates, and molecular skeleton of the compound. The raw bioactivity data includes experimental data such as the inhibition rate of the compound on tumor cells in in vitro experiments, the detection value corresponding to the half-maximal inhibitory concentration, and the activity performance related to the action time. This ensures that the collected raw structural data and raw bioactivity data are complete and can be accurately matched to specific antitumor compounds.

[0022] Based on chemical structure analysis tools, the collected raw structural data are systematically analyzed to identify and extract information that reflects the essential structural characteristics of the compound, including the types and numbers of functional groups in the molecule, the types and numbers of ring structures, the connection mode of atoms, and key structural elements such as stereoconfiguration. These extracted structural elements are organized and summarized according to unified rules to form characteristic structural data that can accurately characterize the structural properties of the antitumor compound.

[0023] The detection standards and numerical ranges of different activity indicators in the raw bioactivity data were clearly defined, and a unified reference benchmark for activity indicators was established. This benchmark was determined based on industry standards for activity detection of similar antitumor compounds and statistical results from a large amount of experimental data. The raw bioactivity data of each antitumor compound was compared with the established reference benchmark. Through a unified proportional conversion method, the raw activity values ​​of different ranges and magnitudes were converted into values ​​under the same standard scale, eliminating numerical differences caused by different detection conditions, and obtaining normalized activity data of the antitumor compounds.

[0024] A unified data format standard was established, including data storage structure, field naming rules, numerical expression formats, and text description specifications, to ensure that the data can be recognized and processed by subsequent analysis tools. The characteristic structure data and normalized activity data were formatted separately. The characteristic structure data was organized into standardized structural information records according to the standard format, and the normalized activity data was organized into a unified numerical list and description according to the standard format. The two types of data were then linked and integrated according to their corresponding relationships to ultimately obtain standard compound data and standardized biological activity data for antitumor compounds.

[0025] The beneficial effects are that, through standardized methods such as searching professional chemical databases and extracting experimental test records, comprehensive original structural data of antitumor compounds, including atomic composition, chemical bond types, and molecular skeletons, as well as original biological activity data such as tumor cell inhibition rate and half-maximal inhibitory concentration in in vitro experiments, are collected. This ensures that both types of data fully cover the essential structure and activity of compounds and accurately correspond to specific compounds, providing comprehensive and reliable basic data support for subsequent standardized processing and avoiding the impact of missing or misplaced data on the accuracy of subsequent research.

[0026] Relying on professional chemical structure analysis tools, the original structural data is systematically analyzed to accurately extract key structural elements such as the types and quantities of functional groups, ring system types, atomic connection methods, and stereoconfigurations. These core features are organized and summarized according to unified rules to form characteristic structural data that can accurately characterize the structural properties of compounds. Irrelevant and redundant information in the original data is discarded, highlighting the core elements related to structure and activity, thus laying a focused foundation for subsequent structure-activity relationship analysis.

[0027] Based on industry standards for the detection of similar antitumor compounds and statistical results from a large amount of experimental data, a unified benchmark for activity indicators is established. Through a standardized conversion method, raw bioactivity data from different detection conditions and at different magnitudes are transformed into normalized activity data under the same standard scale. This eliminates the incomparability of activity indicators caused by differences in detection methods and experimental environments, enabling direct comparative analysis of the activity performance of different compounds and ensuring the scientific rigor of subsequent efficacy classification.

[0028] A unified data format standard was established, covering storage structure, field naming rules, numerical expression formats, and text description specifications, to ensure that the data can be recognized and processed by subsequent analysis tools. The format of characteristic structural data and normalized activity data were adjusted separately, and then standardized into standardized records and numerical lists. The two types of data were then linked and integrated based on the unique compound identifier. The resulting standardized compound data and standardized bioactivity data maintain the correspondence between structural features and activity data while achieving data format uniformity and standardization. This provides high-quality data support for subsequent structure-activity relationship analysis and multidimensional efficacy evaluation, improving the efficiency and accuracy of the overall research process.

[0029] S2. Based on standard compound data and standardized bioactivity data, analyze the correlation between the structural characteristics and activity of the antitumor compounds, and classify the efficacy of the antitumor compounds based on the correlation to obtain the initial efficacy benchmark of the antitumor compounds; In this embodiment of the invention, the step of analyzing the correlation between the structural features and activities of the antitumor compounds based on standard compound data and standardized bioactivity data, and classifying the antitumor compounds by efficacy based on the correlation to obtain the preliminary efficacy criteria for the antitumor compounds, includes: Key structural features are extracted from the standard compound data, and activity response data corresponding to the key structural features are matched from the standardized bioactivity data. Structure-activity relationship pattern extraction was performed on the key structural features and the activity response data to obtain the structure-activity relationship pattern spectrum of the antitumor compound; Based on the structure-activity relationship pattern spectrum, the contribution weight of the core structural feature pattern in the antitumor compound to the activity is calculated, and the efficacy gradient of the antitumor compound is divided according to the contribution weight to obtain the efficacy gradient division result of the antitumor compound. Based on the efficacy gradient division results, a preliminary efficacy classification framework for the antitumor compounds is constructed, and the efficacy gradient division results are mapped to the preliminary efficacy classification framework to obtain the preliminary efficacy benchmark for the antitumor compounds.

[0030] The formula for calculating the contribution weight is as follows: ; In the formula, For the first The weight value of each object, For the first The characteristic index values ​​of an object For the first Another characteristic indicator value of an object It is an exponential adjustment parameter. For the first The characteristic index values ​​of an object This is the sum of the characteristic index values ​​of all objects.

[0031] By deeply analyzing standard compound data, key structural features with significant impacts on tumor suppressor activity were identified. These features include specific functional group combinations, ring system linkages, and differences in stereoconfiguration—structural elements directly related to biological activity. Using a data association mapping method, based on the unique identifier information of each compound, activity response data corresponding to each key structural feature was precisely matched from standardized biological activity data. This ensures a one-to-one correspondence between key structural features and activity response data, providing a precise foundation for subsequent structure-activity relationship analysis.

[0032] A structure-activity relationship (SAR) pattern extraction system was constructed. This system analyzes the correlation between key structural features and corresponding activity response data. By comparing the differences in key structural features of different compounds and the changes in activity response data, regular patterns consistent with the trends in activity response are identified. These regular patterns are categorized and sorted according to structural feature type and activity response intensity. The correspondence between different structural features and activities is presented visually, forming a SAR pattern spectrum that intuitively reflects the correlation between the structure and activity of antitumor compounds.

[0033] Based on structure-activity relationship (SAR) patterns, this study analyzes the activity response intensity of each core structural feature pattern in different compounds. By statistically analyzing the changes in activity data when different structural feature patterns appear, the influence of each core structural feature pattern on overall activity is determined. A corresponding contribution weight is assigned based on the magnitude of the influence; this contribution weight directly reflects the strength of the promoting or inhibiting effect of the structural feature pattern on tumor suppression activity. Based on the cumulative result of the contribution weights, all antitumor compounds are divided into different efficacy levels, each corresponding to a clearly defined range of activity intensity, resulting in an efficacy gradient classification of antitumor compounds.

[0034] Based on the efficacy gradient classification results, the core characteristics and activity performance of different efficacy levels are clarified, and a preliminary efficacy classification framework is constructed, including efficacy level name, core structural feature identifier, activity intensity range, and applicable scenario description. This framework is arranged in descending order of efficacy, with each level having clear defining criteria. Each compound in the efficacy gradient classification results is mapped one-to-one to the corresponding level in the preliminary efficacy classification framework according to its efficacy level and core structural features, achieving a precise mapping between the efficacy gradient classification results and the classification framework. This ultimately forms a preliminary efficacy benchmark capable of systematically classifying the efficacy of antitumor compounds.

[0035] No. The weight value of each object is a core result calculated by combining the values ​​of the object's two characteristic indicators, the exponential adjustment parameter, and the sum of the characteristic indicator values ​​of all objects. It directly reflects the weight value of the first object. The proportion of importance of each object in the overall evaluation system provides a key quantitative basis for subsequent performance gradient division.

[0036] No. The characteristic index values ​​of each object are derived from the extraction and quantification of the activity response data corresponding to the core structural characteristic patterns. By analyzing the structure-activity relationship pattern spectrum, the value of the first object is determined. The activity response intensities induced by the core structural feature patterns in different antitumor compounds were analyzed, and these activity response intensities were converted into specific values ​​according to a unified standard to obtain the [missing information]. The numerical values ​​of the characteristic indicators of an object.

[0037] No. Another characteristic index value of an object is a quantitative representation of the properties of the core structural feature pattern itself. These properties include the stability of the core structural feature, its frequency of occurrence in the compound, and its synergistic ability with other structural features. Through systematic analysis and quantitative transformation of these properties, specific numerical values ​​that reflect the inherent characteristics of the core structural feature pattern are formed, namely the first... Another characteristic indicator value of an object.

[0038] The exponential adjustment parameter is a fixed value set based on the actual needs of structure-activity relationship analysis and the results of historical data verification. It is used to adjust the influence of another characteristic index value on the weight calculation. By adjusting the size of this parameter, the structure-activity relationship characteristics of different types of antitumor compounds can be accurately adapted to ensure that the weight calculation results are consistent with the actual situation.

[0039] No. The characteristic index values ​​of the first object are related to the first... The feature index values ​​of each object have the same source; they are all obtained by extracting and quantifying the activity response data corresponding to the core structural feature patterns. The acquisition process strictly follows the same principle as the first... The same standards and methods are used for the characteristic index values ​​of all objects to ensure that the characteristic index values ​​of all objects are comparable.

[0040] The sum of the feature index values ​​of all objects is the result of accumulating the feature index values ​​corresponding to all core structural feature patterns. By collecting the feature index values ​​of all objects and performing a summation operation according to the numerical accumulation rules, the total value reflecting the overall activity response level of all core structural feature patterns is obtained.

[0041] The significance of this formula lies in its precise quantification of the contribution weight of each core structural feature pattern to tumor inhibitory activity, providing a scientific basis for the efficacy gradient classification of antitumor compounds. The formula obtains the relative proportion of an object's activity response by comparing the feature index value of a single object with the sum of the feature index values ​​of all objects; by introducing another feature index value and an exponential adjustment parameter, it takes into account both the activity performance and inherent properties of the core structural feature patterns; and by comparing it with the comprehensive calculation results of all objects, it achieves weight normalization, ensuring that the sum of the contribution weights of each object is one. The final contribution weights can comprehensively and objectively reflect the importance of each core structural feature pattern, making the efficacy gradient classification results more reasonable and accurate, and laying a solid foundation for the construction of the initial efficacy benchmark.

[0042] The beneficial effects include the ability to accurately screen structural features that play a key role in tumor-suppressive activity from standard compound data. These features are directly related to the activity of the compounds, ensuring that the analysis focuses on core influencing factors. By establishing correlations through unique compound identifiers, corresponding activity response data are matched from standardized biological activity data, achieving a precise correspondence between structural features and activity performance. This provides high-quality, strongly correlated basic data for subsequent structure-activity relationship analysis, avoiding analytical biases caused by data mismatches.

[0043] By analyzing the correlation between key structural features and activity response data, and comparing the structural differences and activity changes of different compounds, regular patterns consistent with structural features and activity responses are accurately identified. These patterns are then categorized and visualized according to structural type and activity intensity, forming a structure-activity relationship pattern spectrum. This visually demonstrates the intrinsic relationship between structure and activity, providing a clear and quantifiable analytical basis for subsequent contribution weight calculations and efficacy classification.

[0044] Based on structure-activity relationship (SAR) pattern mapping, the influence of core structural feature patterns on tumor suppressive activity was quantitatively analyzed, and corresponding contribution weights were assigned to objectively reflect the activity contribution value of different structural features. Based on the cumulative result of the contribution weights, antitumor compounds were divided into different efficacy levels, each corresponding to a clearly defined range of activity intensity, resulting in an efficacy gradient classification. This result achieves precise stratification of compound efficacy, avoiding the subjectivity of traditional screening based solely on experience, and providing a scientific basis for the subsequent construction of a classification framework.

[0045] Based on the efficacy gradient classification results, a preliminary efficacy classification framework was constructed, incorporating elements such as efficacy level names, core structural feature identifiers, and activity intensity ranges. The defining criteria for each level were clarified, forming a systematic classification system. Each compound in the efficacy gradient classification results was precisely mapped to the corresponding level in the classification framework according to its efficacy level and core structural features, achieving orderly classification of compounds. The resulting preliminary efficacy benchmarks provide a clear initial classification basis for subsequent multi-dimensional efficacy evaluation and ranking, ensuring a more targeted and efficient screening process.

[0046] The formula clarifies the relative proportion of an object's activity response within the overall total by comparing the value of a single object's characteristic index with the sum of the values ​​of all objects' characteristic indices. It then combines another characteristic index value and an exponential adjustment parameter to comprehensively consider the activity performance and inherent properties of the core structural feature pattern. By comparing the overall calculation results with all objects, weight normalization is achieved, ensuring that the sum of the contribution weights of each object is equal. This accurately quantifies the contribution weight of each core structural feature pattern to tumor suppression activity, avoiding the subjectivity of traditional empirical judgments.

[0047] The formula incorporates both activity response data corresponding to the core structural feature patterns and structural property data. The former reflects the direct impact of the structure on activity, while the latter reflects the inherent characteristics of the structure, such as stability, frequency of occurrence, and synergistic effects. By flexibly adjusting the influence of structural properties on the weights through exponential adjustment parameters, the formula adapts to the structure-activity relationship characteristics of different types of antitumor compounds, ensuring that the weight calculation closely reflects actual activity performance while also taking into account the intrinsic structural characteristics, resulting in more comprehensive and objective results.

[0048] The calculated weight values ​​directly reflect the importance of each core structural feature pattern in the overall evaluation system, providing clear data support for the efficacy gradient classification of antitumor compounds. Based on the weight accumulation results, compounds can be accurately classified into different efficacy levels, each level corresponding to a clear range of activity intensity. This replaces the traditional single-dimensional screening method, making efficacy classification more logical and convincing, and laying a solid foundation for the construction of initial efficacy benchmarks.

[0049] All parameters in the formula have clear sources and scientific definitions. The values ​​of the characteristic indicators are derived from the quantitative extraction of activity response data, while the values ​​of another characteristic indicator come from the systematic analysis of the structure's own properties. The exponential adjustment parameter has been verified and set using historical data to ensure the rigor of the calculation process. Through standardized calculation logic, the impact of data errors on the results is reduced, making the correlation analysis between core structural features and activity more reliable, and providing high-quality data support for subsequent mechanism analysis and compound screening.

[0050] S3. Based on the preliminary efficacy criteria, the antitumor compounds are evaluated and ranked in a multidimensional efficacy manner to obtain a candidate compound ranking list of the antitumor compounds; In this embodiment of the invention, the step of performing multidimensional efficacy evaluation and ranking of the antitumor compounds based on the initial efficacy benchmark to obtain a candidate compound ranking list of the antitumor compounds includes: Based on the efficacy categories and efficacy levels in the preliminary efficacy benchmarks, a multidimensional efficacy evaluation index system for the antitumor compounds is constructed. Based on the aforementioned multidimensional efficacy evaluation index system, the antitumor compound was independently evaluated to obtain the structural novelty evaluation results, drug resistance evaluation results, and synthesis feasibility evaluation results of the antitumor compound. The structural novelty assessment results, the drug-likeness assessment results, and the synthetic feasibility assessment results are fused together from multiple dimensions to obtain the comprehensive efficacy profile of the antitumor compound. Based on the aforementioned multidimensional efficacy evaluation index system, the efficacy of the initially selected efficacy benchmark is evaluated, and a reference benchmark efficacy profile of the antitumor compound is generated. The comprehensive efficacy profile is compared with the reference baseline efficacy profile, and the antitumor compounds are prioritized according to the comparison results to obtain a candidate compound ranking list of the antitumor compounds.

[0051] The construction of a multidimensional efficacy evaluation index system for the antitumor compounds based on the efficacy categories and efficacy levels in the preliminary efficacy benchmarks includes: The initial performance benchmark is parsed to obtain the performance category information and performance level information of the initial performance benchmark; Based on the efficacy category information and the efficacy level information, a multi-dimensional evaluation index for the antitumor compound is constructed; Based on the multi-dimensional evaluation indicators, the evaluation criteria for the multi-dimensional evaluation indicators are determined; By integrating the multi-dimensional evaluation indicators and the evaluation criteria, a multi-dimensional efficacy evaluation indicator system for the anti-tumor compound is obtained. The multidimensional efficacy evaluation index system is applied to the antitumor compound to obtain quantitative scoring data of the antitumor compound.

[0052] This study delves into the efficacy categories and levels within the initial efficacy benchmarks, clarifying the core efficacy requirements and evaluation priorities corresponding to different categories and levels. Combining the R&D goals and application scenarios of antitumor compounds, structural novelty, drug-likeness, and synthetic feasibility are selected as core evaluation dimensions, with specific sub-indicators set for each dimension. Structural novelty sub-indicators include the uniqueness of the molecular skeleton and the innovativeness of functional group combinations; drug-likeness indicators cover solubility, metabolic stability, and toxicity risk; synthetic feasibility sub-indicators involve the complexity of reaction steps, the availability of raw materials, and the mildness of reaction conditions. These dimensions and sub-indicators are logically integrated, clarifying the evaluation standards and judgment criteria for each indicator, thus constructing a comprehensive and systematic multi-dimensional efficacy evaluation index system for antitumor compounds.

[0053] Based on the requirements of the multidimensional performance evaluation index system, independent performance evaluations were conducted for each dimension. For structural novelty evaluation, global chemical databases were searched, and the structural differences between the target compound and published compounds were compared. The innovative points in terms of molecular skeleton, functional group arrangement, etc., were analyzed, and the structural novelty evaluation results were given according to the set standards. For drug-likeness evaluation, professional drug screening models were used to simulate the absorption, distribution, metabolism, and excretion processes of the compound in vivo. The solubility, metabolic stability, and other indicators were tested to see if they met the drug-likeness standards, and the drug-likeness evaluation results were generated. For synthetic feasibility evaluation, experts in the field of chemical synthesis were organized to assess the ease of compound synthesis, taking into account factors such as raw material market supply and reaction process maturity, and the synthetic feasibility evaluation results were determined.

[0054] We collected specific indicator data from structural novelty assessment, drug-likeness assessment, and synthetic feasibility assessment results, clarifying the efficacy level and quantitative description corresponding to each assessment result. Using a data fusion method, we integrated the assessment results from different dimensions according to a unified standard, eliminating the problems of dimensional differences and varying evaluation scales among the data from different dimensions. By analyzing the inherent relationships between the assessment results from each dimension, we comprehensively presented the overall performance of the compound in terms of structure, drug-likeness, and synthesis, forming a comprehensive efficacy profile that fully reflects the multidimensional efficacy of antitumor compounds.

[0055] Using a multi-dimensional performance evaluation index system as the evaluation basis, representative compounds from each performance category and level in the initial performance benchmark were selected. These representative compounds must possess the typical performance characteristics of their respective category and level. Following the same procedures and standards as independent performance evaluation, these representative compounds were comprehensively evaluated, and the results of each evaluation index were recorded. These results were then integrated according to the multi-dimensional performance dimensions to form a reference benchmark performance profile that reflects the optimal performance level of the performance category and level, providing a clear reference standard for subsequent compound comparisons.

[0056] A collaborative comparison mechanism was established between the overall efficacy profile and the reference baseline efficacy profile, comparing the differences between the two at each evaluation dimension and sub-indicator level. The degree of fit between each indicator in the overall efficacy profile and its corresponding indicator in the reference baseline efficacy profile was calculated; a higher degree of fit indicates that the compound's efficacy is closer to the benchmark level. Appropriate weights were assigned to each dimension based on its importance, and a weighted calculation was performed to obtain the compound's overall fit score. All antitumor compounds were ranked from highest to lowest overall fit score, with higher-scoring compounds having higher priority, ultimately forming a ranked list of candidate antitumor compounds.

[0057] We thoroughly studied the complete content of the preliminary performance benchmark, sorted out the basis for classifying performance categories, and clarified the core characteristics corresponding to each performance category. These characteristics include key information such as the range of compound activity intensity and core structure type. At the same time, we extracted the classification criteria for performance levels, clarified the performance differences and definition conditions between different levels, and obtained the performance category information and performance level information of the preliminary performance benchmark through systematic analysis.

[0058] Based on the obtained efficacy category and efficacy level information, and combined with the research and development goals and application scenarios of antitumor compounds, evaluation indicators were constructed from multiple dimensions, including structure-related, drug-related, and synthesis-related dimensions. The structure-related dimension focuses on aspects such as the uniqueness of the molecular skeleton and the innovativeness of functional group combinations; the drug-related dimension revolves around drug-like characteristics such as solubility, metabolic stability, and toxicity risk; and the synthesis-related dimension focuses on factors such as the complexity of reaction steps, the availability of raw materials, and the mildness of reaction conditions. The specific considerations under these dimensions were clarified as multi-dimensional evaluation indicators.

[0059] For each multi-dimensional evaluation indicator, clear evaluation criteria were formulated by referring to relevant industry standards, historical R&D data, and expert experience. For the structural novelty indicator, criteria such as the threshold for structural differences from published compounds and the required number of innovative points were defined; for the drug-likeness indicator, qualified ranges and grading standards for specific characteristics such as solubility and metabolic stability were set; for the synthetic feasibility indicator, the upper limit of the number of reaction steps and the degree of raw material supply assurance were determined to ensure that each indicator has clear and operable evaluation criteria.

[0060] The multi-dimensional evaluation indicators are categorized and organized according to dimensions such as structure, pharmacodynamics, and synthesis. The specific definition and scope of each indicator are clearly defined, and the corresponding evaluation criteria are linked to each indicator, explaining the judgment conditions for different evaluation results for each indicator. All indicators and criteria are integrated in a logical order to form a clear and complete framework. This framework clarifies the core dimensions and specific indicators of the evaluation and standardizes the evaluation criteria for each indicator, resulting in a multi-dimensional efficacy evaluation indicator system for antitumor compounds.

[0061] The constructed multidimensional efficacy evaluation index system was applied to each antitumor compound. Each evaluation index in the system was compared against the corresponding evaluation criteria to determine the actual performance of the compound. For each index, a score was given based on the degree to which the compound met the criteria. The scoring process strictly followed the requirements of the evaluation criteria to ensure the objectivity and accuracy of the scoring results. The scores of all indicators were summarized to obtain the quantitative score data of each antitumor compound under the multidimensional efficacy evaluation index system. This data comprehensively reflects the efficacy performance of the compound in various dimensions.

[0062] The beneficial effects are that, based on the efficacy category and efficacy level of the initial efficacy benchmark, it accurately matches the R&D needs of anti-tumor compounds and establishes a multi-dimensional evaluation framework covering structural novelty, drug-likeness, and synthetic feasibility. Each dimension has clear sub-indicators and evaluation criteria, which not only meet the core requirements of different efficacy levels but also achieve the unification and standardization of evaluation standards. This overcomes the limitations of traditional single-dimensional evaluation and provides scientific support for comprehensive and objective efficacy evaluation.

[0063] Based on a multidimensional efficacy evaluation index system, each antitumor compound underwent independent evaluation for structural novelty, drug resistance, and synthetic feasibility. Structural novelty evaluation highlighted the compound's innovative value through database comparison; drug-likeness evaluation verified its drug potential through simulated in vivo processes; and synthetic feasibility evaluation assessed the likelihood of commercialization by combining expert experience with process conditions. These three types of results accurately reflect the core characteristics of the compounds from different dimensions, providing comprehensive and detailed foundational data for subsequent integrated evaluation.

[0064] A standardized fusion approach was adopted to eliminate dimensional differences and discrepancies in evaluation scales among the three evaluation results: structural novelty, drug-likeness, and synthetic feasibility. By analyzing the inherent relationships among various results, a comprehensive performance profile was formed that fully presents the overall performance of the compound. This profile retains the core information of each dimension while intuitively showcasing the compound's strengths and weaknesses, avoiding the one-sidedness of single-dimensional results and providing a complete performance reference for subsequent comparisons with benchmark profiles.

[0065] Using a multidimensional efficacy evaluation index system as a unified standard, a comprehensive evaluation was conducted on representative compounds of each efficacy category and level in the initial efficacy benchmark. The optimal efficacy performance of these representative compounds was integrated to form a reference benchmark efficacy profile, clarifying the ideal efficacy level of compounds at different efficacy levels. This profile provides a unified and comparable reference standard for all antitumor compounds, ensuring the objectivity and fairness of subsequent priority ranking.

[0066] A systematic collaborative comparison mechanism was established to compare the degree of fit between the overall performance profile and the reference benchmark performance profile across each evaluation dimension and sub-indicator. A weighted calculation was used to obtain an overall fit score, and compounds were prioritized according to their scores; higher scores indicate that the compound's overall performance is closer to the benchmark level. The resulting candidate compound ranking list clearly defined the order of compounds' merits, accurately screening compounds with outstanding overall potential. This provided a direct basis for identifying subsequent highly active candidate compounds, significantly improving screening efficiency and accuracy.

[0067] A comprehensive review of the initial performance benchmarks was conducted, systematically outlining the criteria and core characteristics for classifying performance categories. Key information such as the activity intensity range and core structural types of compounds corresponding to different categories was clarified. Simultaneously, the standards for classifying performance levels were extracted, clarifying the performance differences and defining conditions between each level. This precise analysis of performance category and level information provides clear guidance for the subsequent construction of the indicator system, ensuring that the indicator system aligns with the core requirements of different performance categories and levels, and avoiding arbitrary indicator design.

[0068] Based on efficacy category and efficacy level information, and combined with the actual needs of antitumor compound development, a targeted multi-dimensional evaluation index is constructed from three core dimensions: structure, pharmacodynamics, and synthesis. The structural dimension focuses on structural features closely related to activity, such as the uniqueness of the molecular skeleton and the innovativeness of functional group combinations; the pharmacodynamics dimension revolves around drug-like characteristics such as solubility, metabolic stability, and toxicity risk; and the synthesis dimension considers feasibility factors such as reaction step complexity and raw material availability. These indicators not only cover the key considerations in compound development but also align with the classification logic of the initial efficacy benchmarks, laying the foundation for a comprehensive evaluation of compound efficacy.

[0069] Drawing on relevant industry standards, historical R&D data, and the experience of domain experts, clear and operational evaluation criteria were developed for each multi-dimensional evaluation indicator. For the structural novelty indicator, the threshold for structural differences from previously published compounds and the required number of innovative points were defined. For the drug-likeness indicator, acceptable ranges and grading standards for properties such as solubility and metabolic stability were set. For the synthetic feasibility indicator, the upper limit for the number of reaction steps and the degree of raw material supply assurance were determined. These clear evaluation criteria eliminate subjective bias in the evaluation process, ensuring the objectivity and consistency of the evaluation results for each indicator and providing a unified standard for subsequent quantitative scoring.

[0070] The multi-dimensional evaluation indicators are categorized and organized according to structure, pharmacodynamics, and synthesis. The specific definition, scope, and correspondence with efficacy categories of each indicator are clearly defined. Each indicator is then linked to its corresponding evaluation criteria, clearly explaining the judgment conditions for different evaluation results. This logical integration forms a hierarchical and comprehensive multi-dimensional efficacy evaluation indicator system. This system clarifies both "what to evaluate" and "how to evaluate," achieving an organic unity between evaluation indicators and evaluation criteria, and providing a scientific framework for systematic and standardized compound efficacy evaluation.

[0071] The constructed multidimensional efficacy evaluation index system was applied to each antitumor compound, and the actual situation of each compound was objectively judged and assigned a corresponding score according to the evaluation criteria of each index. The scores of all indicators were summarized to obtain quantitative score data. This data comprehensively and accurately reflects the efficacy performance of the compounds in various dimensions such as structural novelty, drug-likeness, and synthetic feasibility, and is comparable across different compounds. The quantitative score data provides solid data support for the subsequent construction of the comprehensive efficacy profile and the collaborative comparison with the reference baseline efficacy profile, ensuring the scientific validity and reliability of the candidate compound ranking results.

[0072] S4. Based on the candidate compound ranking list, determine the highly active candidate compounds among the antitumor compounds, and perform in vivo pharmacodynamic tests on the highly active candidate compounds to obtain the in vivo pharmacodynamic data of the antitumor compounds; In this embodiment of the invention, the step of determining highly active candidate compounds among the antitumor compounds according to the candidate compound ranking list, and performing in vivo pharmacodynamic testing on the highly active candidate compounds to obtain in vivo pharmacodynamic data of the antitumor compounds includes: The priority ranking logic and performance profile features of the candidate compound ranking list are analyzed, and a dynamic selection threshold based on the difference between the ranking position and the performance profile is set. Based on the dynamic selection threshold, highly active candidate compounds are selected from the candidate compound ranking list; Based on the highly active candidate compounds, an in vivo pharmacodynamic testing protocol adapted to the structural characteristics and expected mechanism of action of the highly active candidate compounds was determined. Based on the aforementioned in vivo pharmacodynamic testing protocol, the highly active candidate compound was tested in vivo to obtain the original pharmacodynamic observation data of the antitumor compound. The original pharmacodynamic observation data were quantified pharmacodynamically to obtain the in vivo pharmacodynamic data of the antitumor compound.

[0073] A thorough analysis of the candidate compound ranking list clarifies the core criteria for priority ranking. This criterion is determined by the degree of fit between the overall efficacy profile and the reference baseline efficacy profile. Simultaneously, the core characteristics of each compound's efficacy profile are analyzed, including differences in structural novelty, drug-likeness, and synthetic feasibility. Considering the R&D needs and resource allocation of antitumor compounds, a dynamic selection threshold is set, taking into account the overall efficacy level reflected by the ranking position, as well as the uniqueness and complementarity of efficacy profile characteristics. This threshold clarifies the range of eligible compounds while also accommodating the differentiated requirements of efficacy profiles, ensuring that the selected compounds possess both high priority and coverage of diverse efficacy characteristics.

[0074] Each compound in the candidate compound ranking list is compared with its ranking range in the dynamic selection threshold, while its efficacy profile characteristics are checked to see if they meet the differentiation requirements of the threshold. Compounds whose ranking is within the set range and whose efficacy profile characteristics meet the differentiation conditions are directly included in the screening results; for compounds whose ranking meets the criteria but whose efficacy profile characteristics highly overlap with those of already screened compounds, they are discarded according to the requirements of the dynamic selection threshold, with priority given to retaining compounds with more unique efficacy profiles, and finally, the highly active candidate compounds in the candidate compound ranking list are selected.

[0075] This study comprehensively analyzes the structural characteristics of highly active candidate compounds, including key structural information such as molecular skeleton, functional group composition, and stereoconfiguration. It also clarifies their expected mechanisms of action, such as inhibiting tumor cell proliferation, inducing tumor cell apoptosis, and blocking tumor angiogenesis. Referring to industry standards for in vivo pharmacodynamic testing and testing experience with similar compounds, and combining structural characteristics with the expected mechanisms of action, the study determines the animal model to be used for testing. This model must match the antitumor target and tumor type of the compound. Simultaneously, it establishes core testing parameters such as dosage gradient, route of administration, and dosing period, and clarifies the pharmacodynamic indicators to be measured, such as tumor volume change, tumor weight inhibition rate, and animal survival time. These elements are integrated to form an in vivo pharmacodynamic testing protocol suitable for highly active candidate compounds.

[0076] Following the in vivo pharmacodynamic testing protocol, experimental animals with similar health conditions and weights were randomly divided into a control group and multiple treatment groups. The control group received a solvent without the highly active candidate compound, while the treatment groups received the corresponding highly active candidate compound according to the set dose gradient and route of administration. Throughout the entire treatment period, the temperature, humidity, and light conditions of the rearing environment were strictly controlled to ensure consistency of experimental conditions. At the time points set in the protocol, relevant indicators were measured in the experimental animals, and raw data such as tumor volume, animal weight, and survival status were recorded. Simultaneously, any abnormal reactions in the animals were observed, and comprehensive raw pharmacodynamic observation data on the effects of the highly active candidate compound in vivo were collected.

[0077] The original pharmacodynamic observation data were systematically organized, and invalid data due to experimental errors or individual animal abnormalities were removed to ensure the authenticity and reliability of the data. For the recorded tumor volume data, the tumor growth inhibition rate at different time points was calculated; combined with tumor weight data, the inhibitory effect of the compounds on tumors was further quantified; and based on animal survival time records, the effect of the compounds on prolonging animal survival time was analyzed. These quantified results were organized according to unified standards to form systematic data that can intuitively reflect the in vivo antitumor effects of highly active candidate compounds, thus obtaining in vivo pharmacodynamic data of antitumor compounds.

[0078] The beneficial effects include a deep dive into the priority ranking logic of the candidate compound ranking list, clarifying that the core of the ranking is based on the degree of fit between the comprehensive efficacy profile and the reference baseline efficacy profile. Simultaneously, it systematically analyzes the efficacy profile characteristics of each compound in dimensions such as structural novelty, drug-likeness, and synthetic feasibility. Considering the actual needs of resource allocation and development cycle in anti-tumor compound research, and taking into account the comprehensive efficacy level reflected by the ranking position, as well as the uniqueness and complementarity of the efficacy profile, a dynamic selection threshold is set. This threshold not only defines the ranking range of high-priority compounds but also takes into account the coverage requirements of different efficacy characteristics, avoiding homogenization of the selected compounds' efficacy and providing a scientific basis for subsequent precise screening of highly active candidate compounds.

[0079] Using a dynamic selection threshold as the screening criterion, the ranking and efficacy profile characteristics of each compound in the candidate compound ranking list are compared one by one. Compounds whose ranking is within the set range and whose efficacy profile meets the differentiation requirements are directly included in the screening results. Compounds that meet the ranking criteria but whose efficacy profiles highly overlap with those of already screened compounds are discarded, prioritizing the retention of compounds with more unique efficacy characteristics. This screening method accurately identifies highly active candidate compounds with outstanding overall efficacy and differentiated characteristics, avoiding the blindness of traditional batch screening, significantly improving screening efficiency, and ensuring the diversity and representativeness of the screening results.

[0080] This study comprehensively analyzes the structural characteristics of highly active candidate compounds, including their molecular skeleton, functional group composition, and stereoconfiguration, to clarify their expected mechanisms of action, such as inhibiting tumor cell proliferation and inducing tumor cell apoptosis. Referring to industry standards for in vivo pharmacodynamic testing and experience with similar compounds, animal models matching the compound's target and tumor type were selected. Core parameters such as dosage gradient, route of administration, and dosing cycle were scientifically formulated, and key detection indicators such as tumor volume change, tumor weight inhibition rate, and animal survival time were defined. This testing protocol is highly compatible with the structural characteristics and mechanisms of action of the highly active candidate compounds, ensuring that the test results accurately reflect the compound's in vivo antitumor effect and providing reliable support for subsequent efficacy evaluation.

[0081] Following a strict in vivo pharmacodynamic testing protocol, experimental animals of similar health status and weight were selected and randomly grouped. Temperature, humidity, and light conditions in the rearing environment were controlled to ensure consistency and stability of experimental conditions. The control group was given a solvent without the highly active candidate compound, while the treatment groups were administered the compound according to the set dosage gradient and route of administration. Raw data such as tumor volume, animal weight, and survival status were measured and recorded at specific time points, while abnormal animal reactions were closely observed. Through this standardized in vivo testing procedure, comprehensive raw pharmacodynamic data on the compound's effects in vivo were collected, ensuring the completeness and accuracy of the data and laying a solid foundation for subsequent quantitative pharmacodynamic analysis.

[0082] The original pharmacodynamic observation data were systematically organized, and invalid data caused by experimental errors and individual animal abnormalities were eliminated to ensure data reliability. Growth inhibition rates at different time points were calculated based on tumor volume data, and the inhibitory effect on tumors was quantified by combining tumor weight data. The effect on prolonging survival time was analyzed based on animal life cycle records. These quantitative results were integrated according to a unified standard to form in vivo pharmacodynamic data that intuitively and systematically reflects the in vivo antitumor effects of highly active candidate compounds. This data eliminates the fragmentation of the original data, providing accurate and comparable quantitative evidence for subsequent mechanism of action analysis and comprehensive judgment, and facilitating the efficient screening of high-quality antitumor active ingredients.

[0083] S5. Based on the in vivo efficacy data, perform mechanism-of-action analysis on the highly active candidate compounds to obtain preliminary data on the mechanism of the antitumor compounds; In this embodiment of the invention, the step of performing mechanism-directed analysis on the candidate compounds based on the in vivo efficacy data to obtain preliminary mechanism-related data of the antitumor compounds includes: The in vivo pharmacodynamic data are deconstructed using pharmacodynamic response patterns to obtain pharmacodynamic response dimension data of the antitumor compound; The pharmacodynamic response dimension data is correlated with the core structural feature pattern and the efficacy gradient division result in a multidimensional correlation mapping to obtain stable correlation pairs of the antitumor compounds. Based on the stable correlation pairs, a preliminary mechanistic hypothesis for the antitumor compound is constructed; Potential targets that match the structural features in the preliminary mechanism hypothesis are screened out. By integrating the preliminary mechanism hypothesis with the potential target, preliminary data on the mechanism of the antitumor compound are obtained.

[0084] This study comprehensively analyzes in vivo pharmacodynamic data, breaking down core information reflecting the antitumor effects of compounds, including key response manifestations such as tumor growth inhibition rate, extension of animal survival time, and changes in tumor tissue morphology. These response manifestations are categorized and deconstructed according to time, dose, and effect intensity dimensions. The time dimension focuses on the onset time, duration, and decay pattern of the drug effect; the dose dimension analyzes the changing trends of drug effect at different doses; and the effect intensity dimension quantifies the differences in the degree of tumor inhibition. Through this systematic deconstruction, pharmacodynamic response dimension data of antitumor compounds are obtained, which can accurately characterize the pharmacodynamic effects of the compounds.

[0085] The core structural feature patterns and efficacy gradient classification results obtained in the previous stage were retrieved. The core structural feature patterns contain the key structural elements and activity contribution characteristics of the compounds, while the efficacy gradient classification results clearly define the efficacy level and activity range of the compounds. Using pharmacodynamic response data as the core, a multidimensional correlation mapping system was established. The pharmacodynamic response data of each dimension were compared one by one with the structural elements of the core structural feature patterns and the efficacy level of the efficacy gradient classification results. Combinations showing a stable correspondence between pharmacodynamic response and structural features / efficacy level were selected, and combinations with random correlations were removed to obtain stable correlation pairs of antitumor compounds.

[0086] Based on the correspondence between the core structural feature patterns, efficacy levels, and pharmacodynamic response dimensions reflected by stable correlation pairs, and combined with the known knowledge base of antitumor drug mechanisms of action, this study analyzes the possible pathways underlying these stable correlation pairs. Focusing on how core structural feature patterns, through binding to specific biological targets, trigger corresponding pharmacodynamic responses, and the correlation between efficacy level differences and mechanism of action intensity, hypotheses are proposed to reasonably explain the in vivo pharmacodynamic performance of compounds. This clarifies the possible modes of action, target types, and pathways of influence on tumor cells, thus constructing preliminary mechanistic hypotheses for antitumor compounds.

[0087] A database of tumor-related antitumor targets was retrieved, storing structural information, functional properties, pathways of action, and corresponding ligand structural features of various known antitumor targets. The core structural feature patterns involved in the preliminary mechanistic hypothesis were compared with the ligand structural features of each target in the database to screen potential targets whose structural features could complement each other and whose functional properties matched the pathways of action hypothesized in the preliminary mechanism hypothesis. The screened potential targets were then validated to confirm their core roles in tumor development and progression, ensuring the correlation between the selected targets and the preliminary mechanistic hypothesis.

[0088] This process involves compiling core information from the initial mechanism hypothesis, including the pathway and mode of action, as well as structural information, functional characteristics, and pathways of the selected potential targets. The potential targets are then integrated with the initial mechanism hypothesis to clarify the specific role of each target within the hypothesis, explain the binding mechanism between the target and the core structural features, and the logical chain that triggers the pharmacodynamic response. This process supplements the correspondence between target function and pharmacodynamic performance, eliminates information conflicts, and forms a systematic and complete set of preliminary data for the mechanism exploration of antitumor compounds.

[0089] The beneficial effects include a comprehensive analysis of core response metrics in in vivo pharmacodynamic data, such as tumor growth inhibition rate and the extent of animal survival extension. This data is categorized and deconstructed according to time, dosage, and efficacy intensity, clearly presenting key information such as onset time, efficacy changes at different dosages, and the degree of tumor inhibition. The resulting pharmacodynamic response data accurately characterizes the effects of compounds, eliminating the fragmentation of raw data and providing a clear and focused analytical foundation for subsequent correlation mapping, ensuring more targeted analysis of the mechanism of action.

[0090] The core structural feature patterns and efficacy gradient classification results were retrieved. The former covers the key structural elements and activity contribution characteristics of the compound, while the latter clarifies the compound's efficacy level and activity range. Using pharmacodynamic response data as the core, a correlation mapping system was established. The correspondence between pharmacodynamic response and core structural features and efficacy levels in each dimension was compared one by one, and stable and non-random correlation combinations were screened out. Stable correlation pairs tightly bind the compound's structure, efficacy, and pharmacodynamic response, providing a solid correlation basis for the subsequent construction of mechanism hypotheses and ensuring that the hypotheses are not divorced from actual data support.

[0091] Based on stable correlation pairs reflecting the correspondence between structure, efficacy, and pharmacodynamic response, and combined with a knowledge base of known antitumor drug mechanisms of action, this study delves into the potential pathways of action. It clarifies the possible modes of action, target types, and pathways of influence on tumor cells of the compounds, constructing preliminary mechanistic hypotheses that can reasonably explain the in vivo efficacy. These hypotheses provide direction for subsequent target screening, ensuring that target selection is not blind but rather focused on a clearly defined mechanism, thus improving the accuracy of target screening.

[0092] A database of tumor-related antitumor targets was retrieved, storing structural information, functional properties, pathways of action, and ligand structural features of known antitumor targets. The core structural feature patterns from the initial mechanistic hypothesis were compared with the ligand structural features of the targets in the database. Potential targets with complementary structures and functions consistent with the hypothesized mechanistic pathways were identified, and their core roles in tumorigenesis and development were validated. The identified potential targets showed a high degree of agreement with the initial mechanistic hypothesis, providing crucial target support for further elucidating the mechanisms of action of these compounds.

[0093] This study compiles data on the action pathways, modes of action, and structures, functions, and pathways of potential targets based on preliminary mechanism hypotheses. It clarifies the specific role of each target within the hypotheses, explains the binding mechanisms between targets and core structural features, and elucidates the logical chain that triggers the pharmacodynamic response. It also supplements the correspondence between target functions and pharmacodynamic performance, eliminating information conflicts. The resulting preliminary mechanism-level data system is comprehensive and clearly presents the relationships between compound structure, targets, pathways, and pharmacodynamics, providing substantial mechanistic data support for subsequent comprehensive analysis and making compound selection and evaluation more scientific.

[0094] S6. The in vivo efficacy data and the preliminary mechanism data are comprehensively evaluated to obtain a preferred evaluation report of the antitumor compound.

[0095] In this embodiment of the invention, the step of comprehensively evaluating the in vivo efficacy data and the preliminary mechanism data to obtain the preferred evaluation report of the antitumor compound includes: The in vivo efficacy data were subjected to antitumor effect feature extraction to obtain the tumor inhibition intensity characteristics, tumor regression rate characteristics, and survival benefit characteristics of the antitumor compound; Biological significance analysis was performed on the preliminary data of the mechanism described above to obtain the core target intervention characteristics, signal pathway regulation characteristics, and potential off-target risk characteristics of the anti-tumor compound. The effect-mechanism matching degree matrix of the antitumor compound is obtained by performing consistency matching analysis on the tumor inhibition intensity characteristics, the tumor regression rate characteristics, the survival benefit characteristics, the core target intervention characteristics, the signaling pathway regulation characteristics, and the potential off-target risk characteristics. Based on the effect-mechanism matching matrix, the comprehensive advantage assessment and risk trade-off of the highly active candidate compounds are performed to obtain a comprehensive judgment conclusion on the candidate compounds of the antitumor compounds. Based on the comprehensive evaluation of the candidate compounds, a preferred evaluation report of the antitumor compounds is generated.

[0096] This study comprehensively analyzes in vivo efficacy data, focusing on core indicators reflecting anti-tumor effects. By comparing tumor weight and volume changes between the treatment and control groups, the degree of inhibition of tumor growth by the compounds is quantified, and the tumor-inhibiting intensity characteristics of the anti-tumor compounds are extracted. Based on tumor volume monitoring data at different time points, the rate and duration of tumor volume reduction are calculated to clarify the pattern of how quickly the compounds induce tumor regression, obtaining the tumor regression rate characteristics. Data on the survival time and progression-free survival of experimental animals are statistically analyzed to assess the prolongation of animal survival time and the improvement of quality of life, extracting survival benefit characteristics.

[0097] This study delves into the preliminary data on the mechanisms involved, outlining the specific characteristics of intervention at the core targets. It clarifies the binding ability, regulatory mechanisms, and intensity of intervention of the compounds to these targets, and analyzes the impact of this intervention on the physiological processes of tumor cells. Regarding signaling pathway regulation, it analyzes how the compounds regulate the expression levels and activity states of upstream and downstream signaling molecules, clarifying the activation or inhibition pathways of signaling pathways and their regulatory logic on tumor development. Considering potential off-target risks, it identifies non-target sites that the compounds may act on, assesses the probability and severity of adverse reactions caused by these off-target effects, and derives the corresponding biological significance through systematic analysis.

[0098] A consistency matching analysis framework was established, using tumor inhibition intensity, tumor regression rate, and survival benefit as effect-side features, and core target intervention, signaling pathway regulation, and potential off-target risk as mechanism-side features. The intrinsic correlation between effect-side and mechanism-side features was compared one by one to determine whether core target intervention and signaling pathway regulation could reasonably explain the effects of tumor inhibition, regression, and survival benefit, and to assess the correlation between potential off-target risk and effect features. The matching results were quantified according to the degree of fit, constructing an effect-mechanism matching matrix for antitumor compounds that can intuitively present the correspondence between effect and mechanism.

[0099] Based on the effect-mechanism matching matrix, evaluation weights are assigned according to the importance of each feature. Advantageous features such as tumor inhibition intensity and specificity of core target intervention are given positive weights, while potential off-target risks are given negative weights. The advantageous features of highly active candidate compounds are cumulatively calculated to obtain a comprehensive advantage score. Simultaneously, the negative impact of potential off-target risks is quantified, completing the risk-risk trade-off. Through comparative analysis of the comprehensive advantage score and risk score, the core advantages, potential risks, and overall suitability of each compound are clarified, forming a comprehensive evaluation conclusion for candidate antitumor compounds.

[0100] This report summarizes the core findings of the comprehensive evaluation of candidate compounds, including key information such as the effect characteristics, mechanism analysis, effect-mechanism matching, overall advantages, and potential risks of each compound. Following a standardized report format, it first outlines the overall evaluation process and basis, then elaborates on the evaluation results for each compound, clearly stating the ranking of each compound and the reasons for recommendation, and supplementing the optimization directions and precautions for subsequent research and development. Standardized professional terminology is used to ensure the report is logically clear, data is accurate, and conclusions are explicit, generating a superior evaluation report for antitumor compounds.

[0101] The beneficial effects include comprehensive analysis of in vivo efficacy data, focusing on core indicators such as changes in tumor weight and volume. By comparing data from the treatment group and the control group, the degree of inhibition of tumor growth by the compound was precisely quantified, and the characteristics of tumor inhibition intensity were extracted. Based on tumor volume monitoring results at different time points, the rate and duration of tumor shrinkage were clarified, and the characteristics of tumor regression rate were obtained. Data such as the survival time and progression-free survival of experimental animals were statistically analyzed to examine the effects of the compound on prolonging animal survival time and improving quality of life, extracting survival benefit characteristics. These three types of characteristics comprehensively present the in vivo antitumor effect of the compound from different dimensions, providing precise effect-side data support for subsequent mechanism matching and comprehensive judgment.

[0102] A thorough analysis of preliminary mechanistic data systematically elucidates the binding affinity, regulatory mechanisms, and intervention intensity of compounds to core targets, clarifying the impact of this intervention on tumor cell physiological processes and identifying the intervention characteristics of core targets. The study analyzes the regulatory effects of compounds on the expression levels and activity states of upstream and downstream signaling molecules, clarifying the activation or inhibition pathways of signaling pathways and their logical impact on tumorigenesis and development, and extracting signaling pathway regulatory characteristics. The study also identifies potential non-target targets of the compounds, assesses the probability and severity of adverse reactions caused by off-target effects, and identifies potential off-target risk characteristics. These three types of mechanistic characteristics clearly reveal the intrinsic pathways and potential risks of the compounds' anti-tumor effects, providing a clear mechanistic basis for effect-mechanism matching.

[0103] A scientific consistency-matching analysis framework was established to compare three types of effect characteristics—tumor inhibition intensity, tumor regression rate, and survival benefit—with three types of mechanism characteristics—core target intervention, signaling pathway regulation, and potential off-target risk—one by one. The framework was used to determine whether core target intervention and signaling pathway regulation could reasonably explain the corresponding anti-tumor effects, assess the correlation between potential off-target risks and effect characteristics, and quantify the degree of fit. The resulting effect-mechanism matching matrix visually presents the correspondence between effects and mechanisms, clearly reflecting "why the compound is effective" and "how it is effective," providing a visualized core basis for comprehensive advantage and risk assessment.

[0104] Using an effect-mechanism matching matrix as the core, evaluation weights are assigned based on the importance of each feature. Positive weights are given to advantageous features such as tumor inhibition intensity and specificity of core target intervention, while negative weights are given to potential off-target risks. The overall advantage score of highly active candidate compounds is calculated by summing these scores, while simultaneously quantifying the negative impact of potential off-target risks, thus completing the risk-risk trade-off. Through comparative analysis of advantages and risks, the core advantages, potential shortcomings, and overall suitability of each compound are clarified. The resulting comprehensive evaluation conclusions for candidate compounds provide a comprehensive and objective decision-making basis for compound selection, avoiding the one-sidedness of single-dimensional evaluation.

[0105] This report summarizes the core findings of the comprehensive evaluation of candidate compounds, organized according to a standardized report format. It first outlines the overall evaluation process and basis, then details the efficacy characteristics, mechanism analysis, effect-mechanism matching, overall advantages, and potential risks for each compound. The report clearly ranks the compounds by their merits and provides the rationale for recommendation, supplementing the optimization directions and considerations for subsequent research. Standardized professional terminology is used to ensure the report's logical clarity, data accuracy, and explicit conclusions. The generated optimal evaluation report integrates comprehensive information at both the efficacy and mechanism levels, providing a systematic and reliable reference for further research, clinical trials, and industrial applications of antitumor active ingredients, significantly improving the scientific rigor and efficiency of research and development decisions.

[0106] like Figure 2 The diagram shown is a functional block diagram of a research system for anti-tumor active ingredients provided in an embodiment of the present invention.

[0107] The research system 100 for antitumor active ingredients described in this invention can be installed in an electronic device. Depending on the functions implemented, the research system 100 may include a data standardization module 101, a structure-activity relationship efficacy screening module 102, a multidimensional efficacy ranking module 103, a candidate compound verification module 104, a mechanism-oriented analysis module 105, and a comprehensive evaluation and selection module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0108] In this embodiment, the functions of each module / unit are as follows: The data standardization module 101 is used to standardize the structural data and in vitro bioactivity data of the antitumor compound to obtain standard compound data and standardized bioactivity data of the antitumor compound. The structure-activity relationship efficacy screening module 102 is used to analyze the correlation between the structural features and activities of the antitumor compounds based on standard compound data and standardized bioactivity data, and classify the efficacy of the antitumor compounds based on the correlation to obtain the initial efficacy benchmark of the antitumor compounds. The multidimensional efficacy ranking module 103 is used to perform multidimensional efficacy evaluation and ranking of the antitumor compounds based on the initial efficacy benchmark, and obtain a candidate compound ranking list of the antitumor compounds. The candidate compound verification module 104 is used to determine the highly active candidate compounds among the antitumor compounds according to the candidate compound sorting list, and to perform in vivo pharmacodynamic tests on the highly active candidate compounds to obtain in vivo pharmacodynamic data of the antitumor compounds. The mechanism-directed analysis module 105 is used to perform mechanism-directed analysis on the candidate compound based on the in vivo efficacy data to obtain preliminary mechanism exploration data of the antitumor compound. The comprehensive evaluation and selection module 106 is used to comprehensively evaluate the in vivo efficacy data and the preliminary mechanism data to obtain a preferred evaluation report of the antitumor compound.

[0109] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0110] The modules described as separate components may or may not be physically separate. The components shown as modules 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 modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0111] Furthermore, the functional modules in the various embodiments of the present invention 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 in the form of hardware plus software functional modules.

[0112] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0113] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for researching antitumor active ingredients, characterized in that, The method includes: S1. The structural data and in vitro bioactivity data of the antitumor compound are standardized to obtain the standard compound data and standardized bioactivity data of the antitumor compound; S2. Based on standard compound data and standardized bioactivity data, analyze the correlation between the structural characteristics and activity of the antitumor compounds, and classify the efficacy of the antitumor compounds based on the correlation to obtain the initial efficacy benchmark of the antitumor compounds; S3. Based on the preliminary efficacy criteria, the antitumor compounds are evaluated and ranked in a multidimensional efficacy manner to obtain a candidate compound ranking list of the antitumor compounds; S4. Based on the candidate compound ranking list, determine the highly active candidate compounds among the antitumor compounds, and perform in vivo pharmacodynamic tests on the highly active candidate compounds to obtain the in vivo pharmacodynamic data of the antitumor compounds; S5. Based on the in vivo efficacy data, perform mechanism-of-action analysis on the highly active candidate compounds to obtain preliminary data on the mechanism of the antitumor compounds; S6. The in vivo efficacy data and the preliminary mechanism data are comprehensively evaluated to obtain a preferred evaluation report of the antitumor compound.

2. The research method for an antitumor active ingredient as described in claim 1, characterized in that, The structural data and in vitro bioactivity data of the antitumor compounds are standardized to obtain standard compound data and standardized bioactivity data of the antitumor compounds, including: To obtain raw structural and biological activity data of antitumor compounds; Extract the characteristic structural data from the original structural data; The original bioactivity data were normalized to obtain the normalized activity data of the antitumor compound. The characterized structural data and the normalized activity data are processed to unify the format, resulting in standard compound data and standardized biological activity data of the antitumor compound.

3. The research method for an antitumor active ingredient as described in claim 1, characterized in that, The process involves analyzing the correlation between the structural characteristics and activities of the antitumor compounds based on standard compound data and standardized bioactivity data. Based on this correlation, the antitumor compounds are categorized by efficacy to obtain preliminary efficacy criteria, including: Key structural features are extracted from the standard compound data, and activity response data corresponding to the key structural features are matched from the standardized bioactivity data. Structure-activity relationship pattern extraction was performed on the key structural features and the activity response data to obtain the structure-activity relationship pattern spectrum of the antitumor compound; Based on the structure-activity relationship pattern spectrum, the contribution weight of the core structural feature pattern in the antitumor compound to the activity is calculated, and the efficacy gradient of the antitumor compound is divided according to the contribution weight to obtain the efficacy gradient division result of the antitumor compound. Based on the efficacy gradient division results, a preliminary efficacy classification framework for the antitumor compounds is constructed, and the efficacy gradient division results are mapped to the preliminary efficacy classification framework to obtain the preliminary efficacy benchmark for the antitumor compounds.

4. The research method for an antitumor active ingredient as described in claim 3, characterized in that, The formula for calculating the contribution weight is as follows: ； In the formula, For the first The weight value of each object, For the first The characteristic index values ​​of an object For the first Another characteristic indicator value of an object It is an exponential adjustment parameter. For the first The characteristic index values ​​of an object This is the sum of the characteristic index values ​​of all objects.

5. The research method for an antitumor active ingredient as described in claim 1, characterized in that, Based on the initial efficacy benchmark, the antitumor compounds are ranked through multidimensional efficacy evaluation to obtain a candidate compound ranking list, including: Based on the efficacy categories and efficacy levels in the preliminary efficacy benchmarks, a multidimensional efficacy evaluation index system for the antitumor compounds is constructed. Based on the aforementioned multidimensional efficacy evaluation index system, the antitumor compound was independently evaluated to obtain the structural novelty evaluation results, drug resistance evaluation results, and synthesis feasibility evaluation results of the antitumor compound. The structural novelty assessment results, the drug-likeness assessment results, and the synthetic feasibility assessment results are fused together from multiple dimensions to obtain the comprehensive efficacy profile of the antitumor compound. Based on the aforementioned multidimensional efficacy evaluation index system, the efficacy of the initially selected efficacy benchmark is evaluated, and a reference benchmark efficacy profile of the antitumor compound is generated. The comprehensive efficacy profile is compared with the reference baseline efficacy profile, and the antitumor compounds are prioritized according to the comparison results to obtain a candidate compound ranking list of the antitumor compounds.

6. The research method for an antitumor active ingredient as described in claim 5, characterized in that, The construction of a multidimensional efficacy evaluation index system for the antitumor compounds based on the efficacy categories and efficacy levels in the preliminary efficacy benchmarks includes: The initial performance benchmark is parsed to obtain the performance category information and performance level information of the initial performance benchmark; Based on the efficacy category information and the efficacy level information, a multi-dimensional evaluation index for the antitumor compound is constructed; Based on the multi-dimensional evaluation indicators, the evaluation criteria for the multi-dimensional evaluation indicators are determined; By integrating the multi-dimensional evaluation indicators and the evaluation criteria, a multi-dimensional efficacy evaluation indicator system for the anti-tumor compound is obtained. The multidimensional efficacy evaluation index system is applied to the antitumor compound to obtain quantitative scoring data of the antitumor compound.

7. The research method for an antitumor active ingredient as described in claim 1, characterized in that, The step involves determining highly active candidate compounds among the antitumor compounds based on the candidate compound ranking list, and performing in vivo pharmacodynamic testing on the highly active candidate compounds to obtain in vivo pharmacodynamic data of the antitumor compounds, including: The priority ranking logic and performance profile features of the candidate compound ranking list are analyzed, and a dynamic selection threshold based on the difference between the ranking position and the performance profile is set. Based on the dynamic selection threshold, highly active candidate compounds are selected from the candidate compound ranking list; Based on the highly active candidate compounds, an in vivo pharmacodynamic testing protocol adapted to the structural characteristics and expected mechanism of action of the highly active candidate compounds was determined. Based on the aforementioned in vivo pharmacodynamic testing protocol, the highly active candidate compound was tested in vivo to obtain the original pharmacodynamic observation data of the antitumor compound. The original pharmacodynamic observation data were quantified pharmacodynamically to obtain the in vivo pharmacodynamic data of the antitumor compound.

8. The research method for an antitumor active ingredient as described in claim 3, characterized in that, Based on the in vivo efficacy data, the mechanism of action of the candidate compounds is analyzed to obtain preliminary data on the mechanism of action of the antitumor compounds, including: The in vivo pharmacodynamic data are deconstructed using pharmacodynamic response patterns to obtain pharmacodynamic response dimension data of the antitumor compound; The pharmacodynamic response dimension data is correlated with the core structural feature pattern and the efficacy gradient division result in a multidimensional correlation mapping to obtain stable correlation pairs of the antitumor compounds. Based on the stable correlation pairs, a preliminary mechanistic hypothesis for the antitumor compound is constructed; Potential targets that match the structural features in the preliminary mechanism hypothesis are screened out. By integrating the preliminary mechanism hypothesis with the potential target, preliminary data on the mechanism of the antitumor compound are obtained.

9. The research method for an antitumor active ingredient as described in claim 1, characterized in that, The process of comprehensively evaluating the in vivo efficacy data and the preliminary mechanism data to obtain the preferred evaluation report of the antitumor compound includes: The in vivo efficacy data were subjected to antitumor effect feature extraction to obtain the tumor inhibition intensity characteristics, tumor regression rate characteristics, and survival benefit characteristics of the antitumor compound; Biological significance analysis was performed on the preliminary data of the mechanism described above to obtain the core target intervention characteristics, signal pathway regulation characteristics, and potential off-target risk characteristics of the anti-tumor compound. The effect-mechanism matching degree matrix of the antitumor compound is obtained by performing consistency matching analysis on the tumor inhibition intensity characteristics, the tumor regression rate characteristics, the survival benefit characteristics, the core target intervention characteristics, the signaling pathway regulation characteristics, and the potential off-target risk characteristics. Based on the effect-mechanism matching matrix, the comprehensive advantage assessment and risk trade-off of the highly active candidate compounds are performed to obtain a comprehensive judgment conclusion on the candidate compounds of the antitumor compounds. Based on the comprehensive evaluation of the candidate compounds, a preferred evaluation report of the antitumor compounds is generated.

10. A research system for antitumor active ingredients, characterized in that, A system for implementing a research method for an antitumor active ingredient according to claim 1, the system comprising: The data standardization module is used to standardize the structural data and in vitro bioactivity data of the antitumor compounds to obtain standard compound data and standardized bioactivity data of the antitumor compounds. The structure-activity relationship efficacy screening module is used to analyze the correlation between the structural features and activities of the antitumor compounds based on standard compound data and standardized bioactivity data, and to classify the efficacy of the antitumor compounds based on the correlation to obtain the initial efficacy benchmark of the antitumor compounds. The multidimensional efficacy ranking module is used to perform multidimensional efficacy evaluation and ranking of the antitumor compounds based on the initial efficacy benchmark, and obtain a candidate compound ranking list of the antitumor compounds. The candidate compound verification module is used to determine the highly active candidate compounds among the antitumor compounds according to the candidate compound ranking list, and to perform in vivo pharmacodynamic tests on the highly active candidate compounds to obtain in vivo pharmacodynamic data of the antitumor compounds. The mechanism-oriented analysis module is used to perform mechanism-oriented analysis on the candidate compounds based on the in vivo efficacy data, and obtain preliminary mechanism exploration data of the antitumor compounds. The comprehensive evaluation and selection module is used to comprehensively evaluate the in vivo efficacy data and the preliminary mechanism data to obtain an optimal evaluation report of the antitumor compound.

Images