Data-driven pathogenic microorganism target analysis method and system

Abstract

The invention discloses a pathogenic microorganism target analysis method and system based on data driving, and belongs to the technical field of data analysis. The method comprises the following steps: establishing a pathogenic microorganism target experiment sample library, and separating pathogen types, environmental induction factors and morning and evening period characteristic appearance time data clusters; configuring parameters and generating a driving characteristic state; constructing early and late period target feature matrixes, and analyzing feature relevancy; iteratively evaluating to generate a potential target matrix and marking an environment induction factor. The system comprises a target sample filing module, a driving feature generation module, a feature matrix analysis module and a potential target classification processing module. According to the method, through data-driven modeling, the time correlation of pathogen early and late characteristics is quantified, target dynamic identification and environmental factor correlation analysis are realized, and the target screening efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and specifically to a data-driven pathogenic microorganism target analysis method and system. Background Art

[0002] Pathogenic microorganism target analysis is a core link in the research and development of antibacterial drugs and pathogen detection. The existing technology mainly relies on static analysis of single features, lacking systematic research on the feature associations of pathogens at different growth stages (such as early metabolism and late pathogenesis) and the influence of environmental factors. For example, the dynamic association between early-expressed metabolites and late virulence factors of pathogens has not been effectively quantified, resulting in the inability of target screening to reflect the integrity of the pathogen life cycle; at the same time, the induction effect of environmental factors (such as temperature and nutrient concentration) on target features lacks dynamic modeling, making it difficult to achieve accurate target prediction and risk warning. Summary of the Invention

[0003] Aiming at the one-sidedness of target analysis and the lag of environmental response in the existing technology, the present invention proposes a data-driven pathogenic microorganism target analysis method and system. By integrating the characteristic data of the entire life cycle of pathogens, a time-dimensional association model is constructed to realize the quantification of the synergy between early and late features and the dynamic identification of environmental induction factors, providing a new paradigm for the accurate screening of pathogen targets.

[0004] The present invention provides the following technical solutions: A data-driven pathogenic microorganism target analysis system, which includes: a target sample filing module, a driving feature generation module, a feature matrix analysis module, and a potential target classification and processing module; The target sample filing module is used to establish a pathogenic microorganism target experimental sample library, store target experimental sample data, and generate data clusters; The driving feature generation module is used to configure data cluster parameters, establish a time-dimensional relationship, and capture and generate driving feature states; The feature matrix analysis module is used to lock samples and construct a target feature matrix, and analyze the correlation between early and late features; The potential target classification and processing module is used to iteratively evaluate the feature correlation, generate a potential target matrix, and mark environmental induction factors.

[0005] Further, the target sample filing module includes a sample storage unit and a data cluster generation unit; The sample storage unit is used to store uniquely encoded pathogenic microorganism target experimental samples, record pathogen types, environmental induction factors, and the occurrence times of early and late features; The data cluster generation unit separates data clusters based on index category information, including pathogen type data clusters, environmental inducing factor data clusters, early feature appearance time data clusters, and late feature appearance time data clusters.

[0006] Furthermore, the driving feature generation module includes a parameter configuration unit and a driving capture unit; The parameter configuration unit is used to configure data cluster parameters and establish a time delay scale correspondence relationship between the early feature appearance time and the late feature appearance time; The driving capture unit is used to generate driving feature states with pathogen types and environmental inducing factors as data driving objects.

[0007] Furthermore, the feature matrix analysis module includes a matrix construction unit and a correlation analysis unit; The matrix construction unit is used to construct early and late target feature matrices with driving feature states as matrix elements; The correlation analysis unit quantifies the early and late feature synergy and calculates the feature correlation based on Boolean matrix intersection and union operations.

[0008] Furthermore, the potential target classification and processing module includes an iterative evaluation unit and a potential target marking unit; The iterative evaluation unit iteratively adjusts the time delay scale through a preset feature correlation threshold to screen out target feature matrices with high correlation; The potential target marking unit is used to accumulate the screened matrices to generate a potential target matrix, extract the row sequence, and mark the potential environmental inducing factors corresponding to the pathogen types.

[0009] A data-driven method for analyzing pathogen targets, the method comprising the following steps: Step S1: Establish a pathogen target experimental sample library, store target experimental sample data, record pathogen types, environmental inducing factors, early feature appearance time, and late feature appearance time, and separate the corresponding data clusters; Step S2: Configure parameters for each data cluster and establish a time dimension relationship between the early feature appearance time and the late feature appearance time; use pathogen types and environmental inducing factors as data driving objects to capture and generate driving feature states; Step S3: Based on the time dimension relationship, respectively lock the pathogen target experimental samples at the early feature appearance time or the late feature appearance time to respectively form target feature matrices at the early feature appearance time and the late feature appearance time, and perform feature correlation analysis between the pathogen target experimental samples in the early and late stages; Step S4: Iteratively evaluate the feature relevance to form a potential target matrix; based on the different row sequences in the potential target matrix corresponding to the pathogen type, mark the potential environmental inducing factors and output them to the experimenter's port.

[0010] Further, the specific implementation process of the step S1 includes: Establish a pathogen target experimental sample library, in which different pathogen target experimental samples are recorded, and each pathogen target experimental sample is attached with different index category information, and the index category information includes pathogen type, environmental inducing factor, early feature appearance time, and late feature appearance time; Based on the index category information, separate the pathogen target experimental samples to obtain data clusters with different index category information attributes, and the data clusters include pathogen type data cluster, environmental inducing factor data cluster, early feature appearance time data cluster, and late feature appearance time data cluster.

[0011] Further, the specific implementation process of the step S2 includes: Configure the pathogen type data cluster 、the environmental inducing factor data cluster 、the early feature appearance time data cluster and the late feature appearance time data cluster , where represents the h-th pathogen target experimental sample, and H represents the total number of pathogen target experimental samples, 、 and respectively represent the pathogen type 、the environmental inducing factor and the early feature appearance time obtained when separating the pathogen target experimental sample , n, i, and r are the coding serial numbers of the pathogen type, environmental inducing factor, and early feature appearance time respectively, and N, I, and R represent the maximum values of the coding serial numbers of the pathogen type, environmental inducing factor, and early feature appearance time respectively, and 、 and , g is the time delay scale, and the early feature appearance time and the late feature appearance time have a time dimension corresponding relationship with a time delay scale of g; Capture the data-driven objects: the pathogen type and the environmental inducing factor , generate the driving feature state , in the pathogen target experimental sample , if the pathogen type is captured simultaneously and environmental inducing factors , then set the drive feature state . If the pathogen type and environmental inducing factors are not captured simultaneously, then set the drive feature state .

[0012] Furthermore, the specific implementation process of step S3 includes: Take the drive feature state as the matrix element with row number n and column number i to respectively form the target feature matrix at the early feature appearance time and the late feature appearance time and ; Based on the target feature matrix, evaluate the feature correlation degree between the early and late stages of the pathogen microorganism target experimental sample . In the formula, represents the number of matrix elements with a value of 1 included after the Boolean intersection operation and between the target feature matrices represents the number of matrix elements with a value of 1 included after the Boolean union operation and between the target feature matrices; In the above method, the physical principle of the feature correlation degree is: the feature correlation degree is equal to the number of "feature - environment combinations that appear simultaneously in the early and late stages" divided by the number of "feature - environment combinations that appear at least once in the early and late stages". The closer the value of the feature correlation degree is to 1, the stronger the coordination of the early and late features and the greater the potential target association risk.

[0013] Furthermore, the specific implementation process of step S4 includes:

[0014] Preset a feature correlation threshold and set to perform iterative evaluation of the feature correlation degree, and stop the iterative evaluation when it reaches. G is the time node number at the end of the target experiment. When the iterative evaluation stops, screen out the feature correlation degrees greater than or equal to the feature correlation threshold corresponding target feature matrices ; Perform matrix cumulative summation operation on the screened target feature matrices , and denote the matrix after the cumulative summation operation as the potential target matrix . Extract the matrix elements in the nth row of the potential target matrix to form the row sequence Underlying environmental inducing factors ; In the above method, by dynamically iterating to exclude low-correlation data and focusing on high-risk target combinations, the dynamic correlation mining between environmental factors and pathogen characteristics is realized.

[0015] Compared with the prior art, the beneficial effects achieved by the present invention include: Dynamic identification of targets driven by multi-dimensional data: By establishing a pathogen microorganism target experimental sample library and systematically integrating multi-dimensional data such as pathogen types, environmental inducing factors, early and late characteristic times, etc., the limitations of traditional single-characteristic analysis are broken through. For example, in bacterial target analysis, time series data of "Escherichia coli - high temperature - early metabolites - late endotoxin" can be associated simultaneously to quantify the synergy between characteristics at different growth stages and environmental factors, and improve the comprehensiveness of target identification; Precise modeling with time dimension correlation: By constructing an early and late characteristic matrix and calculating the time correlation coefficient, the system can quantify the correlation intensity of pathogen characteristics at different time nodes. For example, in virus research, by analyzing the time correlation coefficient between early antigen expression and late pathogenic proteins, strong-correlation target combinations such as "Dengue virus - 37°C - NS1 protein - E protein" can be accurately located, providing a quantitative basis for virus life cycle research; Dynamic marking of environmental inducing factors: Through iterative evaluation and potential target matrix analysis, the system can automatically mark key environmental inducing factors under specific pathogen types. For example, in fungal experiments, by accumulating and summing operations to screen high-correlation matrices, environmental-characteristic combinations such as "Candida albicans - acidic pH - hyphal morphology - invasive enzymes" can be identified, providing precise environment-responsive targets for the development of antifungal drugs; Improvement of experimental efficiency and accuracy: The system significantly reduces the manual analysis cost through an automated process of data cluster separation, matrix modeling, and iterative evaluation. At the same time, through the quantitative evaluation of the time correlation coefficient, subjective judgment errors are avoided, and the accuracy and repeatability of target screening are improved. Brief Description of the Drawings

[0016] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention.

[0017] Figure 1 It is a schematic diagram of the steps of a data-driven pathogen microorganism target analysis method of the present invention. Detailed Embodiments

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] In the first embodiment: A data-driven pathogenic microorganism target analysis system is provided, and the system includes: a target sample filing module, a driving feature generation module, a feature matrix analysis module, and a potential target classification and processing module; The target sample filing module is used to establish a pathogenic microorganism target experimental sample library, store target experimental sample data, and generate data clusters; Among them, the target sample filing module includes a sample storage unit and a data cluster generation unit; The sample storage unit is used to store uniquely encoded pathogenic microorganism target experimental samples, record the pathogen type, environmental inducing factors, and the occurrence time of early and late features; The data cluster generation unit separates data clusters based on index category information, including pathogen type data clusters, environmental inducing factor data clusters, early feature occurrence time data clusters, and late feature occurrence time data clusters; The driving feature generation module is used to configure data cluster parameters, establish a time dimension relationship, and capture and generate driving feature states; Among them, the driving feature generation module includes a parameter configuration unit and a driving capture unit; The parameter configuration unit is used to configure data cluster parameters and establish a time delay scale correspondence relationship between the early feature occurrence time and the late feature occurrence time; The driving capture unit is used to use the pathogen type and environmental inducing factors as data driving objects to generate driving feature states; The feature matrix analysis module is used to lock samples and construct a target feature matrix, and analyze the correlation between early and late features; Among them, the feature matrix analysis module includes a matrix construction unit and a correlation analysis unit; The matrix construction unit is used to use the driving feature state as matrix elements to construct early and late target feature matrices; The correlation analysis unit quantifies the cooperation of early and late features based on Boolean matrix intersection and union operations, and calculates the feature correlation; The potential target classification and processing module is used to iteratively evaluate the feature correlation, generate a potential target matrix, and mark environmental inducing factors; Among them, the potential target classification and processing module includes an iterative evaluation unit and a potential target marking unit; Iterative evaluation unit, which iteratively adjusts the time delay scale through a preset feature correlation threshold to screen out a target feature matrix with a high degree of correlation. Potential target marking unit, which is used to accumulate the screened matrices to generate a potential target matrix, extract the row sequence, and mark the potential environmental inducing factors corresponding to the pathogen types.

[0020] Please refer to Figure 1 , in the second embodiment: Provide a data-driven pathogen target analysis method to adapt to the above-mentioned first embodiment. The method includes the following steps: Step S1: Establish a pathogen target experimental sample library, store the target experimental sample data, record the pathogen type, environmental inducing factors, early feature appearance time, and late feature appearance time, and separate the corresponding data clusters. Exemplarily, establish a pathogen target experimental sample library. The pathogen target experimental sample library records different pathogen target experimental samples, and each pathogen target experimental sample is attached with different index category information. The index category information includes the pathogen type, environmental inducing factors, early feature appearance time, and late feature appearance time. Based on the index category information, separate the pathogen target experimental samples to obtain data clusters with different index category information attributes. The data clusters include a pathogen type data cluster, an environmental inducing factor data cluster, an early feature appearance time data cluster, and a late feature appearance time data cluster.

[0021] Step S2: Configure parameters for each data cluster and establish a time dimension relationship between the early feature appearance time and the late feature appearance time; use the pathogen type and environmental inducing factors as data-driven objects to capture and generate the driving feature state. Exemplarily, configure the pathogen type data cluster , the environmental inducing factor data cluster , the early feature appearance time data cluster and the late feature appearance time data cluster , where represents the h-th pathogen target experimental sample, H represents the total number of pathogen target experimental samples, , and respectively represent the pathogen type , environmental inducing factor , and early feature appearance time obtained when separating the pathogen target experimental sample , and n, i, and r are the coding serial numbers of the pathogen type, environmental inducing factor, and early feature appearance time respectively. N, I, and R represent the maximum values of the coding serial numbers of the pathogen type, environmental inducing factor, and early feature appearance time respectively, and , and , where \(g\) is the time delay scale, the early feature appearance time and the late feature appearance time have a time - dimension correspondence relationship with the time delay scale \(g\); Capture data - driven objects: pathogen types and environmental inducing factors to generate a driving feature state . In the pathogen - microorganism target experimental samples , if both the pathogen type and the environmental inducing factor are captured, then set the driving feature state . If both the pathogen type and the environmental inducing factor are not captured simultaneously, then set the driving feature state .

[0022] Step S3: Based on the time - dimension relationship, lock the pathogen - microorganism target experimental samples at the early feature appearance time or the late feature appearance time respectively, to form a target feature matrix at the early feature appearance time and a target feature matrix at the late feature appearance time respectively, and conduct an analysis of the feature correlation degree between the early and late pathogen - microorganism target experimental samples; Exemplarily, use the driving feature state as the matrix element with row number \(n\) and column number \(i\) to form a target feature matrix at the early feature appearance time and a target feature matrix and at the late feature appearance time; Based on the target feature matrix, evaluate the feature correlation degree between the early and late of the pathogen - microorganism target experimental samples . Wherein, represents the number of matrix elements with value 1 contained after the Boolean intersection operation between the target feature matrix and , and represents the number of matrix elements with value 1 contained after the Boolean union operation between the target feature matrix and .

[0023] Step S4: Iteratively evaluate the feature correlation degree to form a potential target matrix; Based on the different row sequences in the potential target matrix corresponding to the pathogen type, mark the potential environmental inducing factors and output them to the experimental personnel port; Exemplarily, preset a feature - correlation threshold and let conduct an iterative evaluation of the feature correlation degree, and When the iterative evaluation stops, G is the time node number at the end of the target experiment. When the iterative evaluation stops, the feature correlation degrees greater than or equal to the feature correlation threshold are screened out. The corresponding target feature matrix ; For the screened target feature matrix Perform matrix cumulative summation operation, and denote the matrix after the cumulative summation operation as the potential target matrix , and extract the matrix elements of the nth row in the potential target matrix to form a row sequence , and mark the pathogen type The potential environmental induction factors under ; For example, in the bacterial target analysis scenario, the experimental object is Pseudomonas aeruginosa, the main environmental factor studied is the iron deficiency concentration. In the early stage , the main form is siderophore, and in the late stage , biofilms are formed. After iterative evaluation, a potential target matrix is generated, and "iron deficiency" is marked as the key environmental induction factor. Then, siderophores and biofilms are strongly associated under iron-deficient conditions and can be used as combined antibacterial targets; in the viral target analysis scenario, the experimental object is influenza virus, the main environmental factor studied is the host cell temperature. Hemagglutinin HA is observed in the early stage, and neuraminidase NA is observed in the late stage. After iterative evaluation, a potential target matrix is generated, and "37°C" is marked as the key environmental factor. HA and NA are cooperative targets. Then, the expression of HA and NA in influenza virus is strongly correlated at 37°C and can be used as a dual-target combination for antiviral drugs.

[0024] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0025] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A data-driven method for analyzing pathogen targets, characterized in that The method includes the following steps: Step S1: Establish a pathogenic microorganism target experimental sample library, store the target experimental sample data, record the pathogen type, environmental inducing factors, the occurrence time of early features, and the occurrence time of late features, and isolate the corresponding data clusters; Step S2: Configure parameters for each data cluster, and establish the time dimension relationship between the occurrence time of early features and the occurrence time of late features; taking the pathogen type and environmental inducing factors as data-driven objects, capture and generate the driving feature state; Step S3: Based on the time dimension relationship, lock the pathogenic microorganism target experimental samples at the occurrence time of early features or the occurrence time of late features respectively, to form the target feature matrices at the occurrence time of early features and the occurrence time of late features respectively, and conduct the feature correlation analysis of the pathogenic microorganism target experimental samples between the early stage and the late stage; Step S4: Iteratively evaluate the feature correlation to form a potential target matrix; based on the different row sequences in the potential target matrix corresponding to the pathogen type, mark the potential environmental inducing factors and output them to the experimenter's port.

2. The method for analyzing pathogenic microorganism targets based on data driving according to claim 1, wherein The specific implementation process of step S1 includes: Establish a pathogenic microorganism target experimental sample library, in which different pathogenic microorganism target experimental samples are recorded, and each pathogenic microorganism target experimental sample is attached with different index category information, and the index category information includes pathogen type, environmental inducing factors, the occurrence time of early features, and the occurrence time of late features; Based on the index category information, separate the pathogenic microorganism target experimental samples to obtain data clusters with different index category information attributes, and the data clusters include pathogen type data clusters, environmental inducing factor data clusters, early feature occurrence time data clusters, and late feature occurrence time data clusters.

3. The method for analyzing pathogen microbial targets based on data driving according to claim 2, wherein The specific implementation process of step S2 includes: Configure the pathogenic type data cluster , the environmental inducing factor data cluster , the early feature appearance time data cluster and the late feature appearance time data cluster , where represents the h-th pathogenic microorganism target experimental sample, and H represents the total number of pathogenic microorganism target experimental samples, , and respectively represent the pathogenic type obtained when separating the pathogenic microorganism target experimental sample , the environmental inducing factor and the early feature appearance time , where n, i, and r are the coding sequence numbers of the pathogenic type, environmental inducing factor, and early feature appearance time respectively, and N, I, and R respectively represent the maximum values of the coding sequence numbers of the pathogenic type, environmental inducing factor, and early feature appearance time, and , and , where g is the time delay scale, and the early feature appearance time and the late feature appearance time have a time dimension correspondence relationship with the time delay scale g; Capture data-driven objects: pathogen types and environmental inducing factors , generate the driving feature state , in the experimental samples of pathogenic microorganism targets , if both the pathogen type and the environmental inducing factor are captured simultaneously, then set the driving feature state , if the pathogen type and the environmental inducing factor are not captured simultaneously, then set the driving feature state .

4. A data-driven pathogenic microorganism target analysis method according to claim 3, characterized in that, The specific implementation process of step S3 includes: Taking the drive feature status as the matrix element with row number n and column number i to respectively form the early feature appearance time and the late feature appearance time of the target feature matrix and ; Evaluate the feature correlation between the early and late stages of a pathogenic microorganism target experimental sample based on the target feature matrix where the number of matrix elements with a value of 1 after the Boolean intersection operation between and represents the target feature matrix and is the number of matrix elements with a value of 1 after the Boolean union operation between represents the target feature matrix and is the number of matrix elements with a value of 1 after the Boolean union operation between 5. A data-driven pathogenic microorganism target analysis method according to claim 4, characterized in that The specific implementation process of step S4 includes: Set a threshold related to the preset feature, and let Perform iterative evaluation of the feature correlation, and When the iterative evaluation stops, where G is the time node number at the end of the target experiment. When the iterative evaluation stops, select the feature correlation that is greater than or equal to the feature correlation threshold The corresponding target feature matrix ; For the selected target feature matrix Perform matrix cumulative summation operation, and denote the matrix after the cumulative summation operation as the potential target matrix , extract the potential target matrix The matrix elements of the nth row in, to form a row sequence , and mark the potential environmental inducing factors under the pathogen type .​ 6. A data-driven pathogenic microorganism target analysis system that executes the pathogenic microorganism target analysis method according to any one of claims 1-5, characterized in that, The system includes: a target sample filing module, a driving feature generation module, a feature matrix analysis module, and a potential target classification and processing module; The target sample filing module is used to establish a pathogenic microorganism target experimental sample library, store the target experimental sample data, and generate data clusters; The driving feature generation module is used to configure the data cluster parameters, establish the time dimension relationship, and capture and generate the driving feature state; The feature matrix analysis module is used to lock the samples and construct the target feature matrix, and analyze the correlation between early and late features; The potential target classification and processing module is used to iteratively evaluate the feature correlation, generate a potential target matrix, and mark the environmental inducing factors.

7. The data-driven pathogenic microorganism target analysis system according to claim 6, characterized in that The target sample filing module includes a sample storage unit and a data cluster generation unit; The sample storage unit is used to store the uniquely encoded pathogenic microorganism target experimental samples, and record the pathogen type, environmental inducing factors, and the occurrence time of early and late features; The data cluster generation unit separates the data clusters based on the index category information, including pathogen type data clusters, environmental inducing factor data clusters, early feature occurrence time data clusters, and late feature occurrence time data clusters.

8. The data-driven pathogenic microorganism target analysis system according to claim 6, wherein The driving feature generation module includes a parameter configuration unit and a driving capture unit; The parameter configuration unit is used to configure data cluster parameters and establish a time delay scale correspondence relationship between the early feature occurrence time and the late feature occurrence time; The driving capture unit is used to generate driving feature states with the pathogen type and environmental inducing factors as data driving objects.

9. A data-driven pathogenic microorganism target analysis system according to claim 6, wherein The feature matrix analysis module includes a matrix construction unit and a correlation analysis unit; The matrix construction unit is used to construct early and late target feature matrices with the driving feature states as matrix elements; The correlation analysis unit quantifies the early and late feature synergy and calculates the feature correlation based on the Boolean matrix intersection and union operations.

10. A data-driven pathogenic microorganism target analysis system according to claim 6, characterized in that, The potential target classification and processing module includes an iterative evaluation unit and a potential target marking unit; The iterative evaluation unit iteratively adjusts the time delay scale through a preset feature correlation threshold to screen out the target feature matrices with high correlation; The potential target marking unit is used to accumulate the screened matrices to generate a potential target matrix, extract the row sequence, and mark the potential environmental inducing factors corresponding to the pathogen type.

Images