Marker for screening aggressive sociopathic behavior, screening method, device, and storage medium

Abstract

The application discloses a marker for screening aggressive social harmful behavior, a screening method and device and a storage medium, and relates to the technical field of aggressive social behavior screening. The reagent comprises a reagent for detecting relative abundance data of a microbial combination at the genus level. When the 13 kinds of bacteria are used for screening aggressive social harmful behavior, the screening accuracy is extremely high, and the screening method is non-invasive and objective. When the 13 kinds of bacteria are combined with social demographic information as screening factors of aggressive harmful social behavior, the screening accuracy is also extremely high. The application provides a new way for screening and intervention of aggressive social harmful behavior.

Description

Technical Field

[0001] This invention relates to the field of aggressive social behavior screening technology, and more specifically, to markers, screening methods, devices, and storage media for screening aggressive social harmful behaviors. Background Technology

[0002] The gut microbiota, composed of a diverse array of symbiotic microorganisms, plays a crucial role in host development, including immune system maturation, brain development, and host behavior. The gut-microbe-brain axis mediates a complex connection between the brain and the gut, involving multiple key systems such as the central nervous system, autonomic nervous system, enteric nervous system, hypothalamic-pituitary-adrenal (HPA) axis, neuroendocrine system, and immune system. Bidirectional communication between the gut microbiota and the central nervous system is achieved through various mechanisms, including immune system activation, the production of neuroactive metabolites (such as short-chain fatty acids), and vagus nerve stimulation. Research into this gut-brain interaction mechanism has driven extensive exploration of the association between microbial community function and metabolites and mental and behavioral disorders.

[0003] Aggressive behavior is any form of purposeful harm to another organism that is undesirable to that organism. Socially harmful aggressive behavior includes violent crime, which is criminal behavior that uses violence to achieve its criminal purpose. According to the Institute for Economics & Peace (IEP) Global Peace Index 2024, the economic losses caused by violence worldwide in 2023 were equivalent to 13.5% of global GDP or $2,380 per person. Despite the heavy social burden of violent crime, our understanding of the biological mechanisms of human violence remains limited.

[0004] Multiple animal studies have demonstrated a link between gut microbiota and aggressive behavior. Studies in fruit flies and mice have found that gut microbiota can stimulate the production of the neurotransmitter octopamine, leading to increased aggressive behavior. Studies in germ-free mice have shown that regulating gut microbiota in early development can suppress aggression. Furthermore, disrupting the gut microbiota of pregnant animals increases the aggression of their offspring. A review synthesizing animal studies in mouse, dog, hamster, and fruit fly models suggests that aggression may be influenced by the "microbe-gut-brain axis." While species-specific patterns of aggressive behavior in animals are considered among the earliest evolved emotional behavioral patterns, certain forms of aggression, particularly life-threatening aggression in humans (such as violent crime), impose a significant economic burden on society.

[0005] Currently, research on human aggression (especially violent crime) remains very limited.

[0006] Given that differences in gut microbiota composition are believed to be related to neurodevelopment and play an important role in neurodevelopment and behavioral control, it is essential to conduct in-depth research on the gut microbiota of aggressive socially harmful behaviors to provide useful insights for clinical treatment and disease prevention.

[0007] In view of this, the present invention is proposed. Summary of the Invention

[0008] The purpose of this invention is to provide markers, screening methods, devices, and storage media for screening aggressive socially harmful behaviors to solve the aforementioned technical problems.

[0009] This invention is implemented as follows: In a first aspect, the present invention provides the application of a reagent for detecting microbial assemblages in the preparation of products for screening aggressive socially harmful behaviors, the reagent comprising: a reagent for detecting relative abundance data at the genus level of the microbial assemblages; The microbial assemblage includes Clostridium ( Clostridium ), genus *Trichophyton* Lachnospira ), Enterobacteriaceae ( Enterocloster ), Fusobacterium spp. Fusobacterium Mediterranean bacteria () Mediterraneibacter ), Broutbacterium spp. Blautia ), Flavonoids ( Flavonifractor ), symbiotic pectinobacterium ( Symbiopectobacterium ), Anaerobic rod-shaped bacteria ( Anaerostipes ), Dorperella spp. Dorea Bifidobacterium spp. Bifidobacterium ), Prevotella spp. Prevotella ) and Mycoplasma genus ( Mycoplasma ).

[0010] In a preferred embodiment of the present invention, the above-mentioned product is a reagent kit, a chip, or a screening device.

[0011] In a preferred embodiment of the present invention, the reagents are selected from primers and / or probes.

[0012] Secondly, this invention provides the application of biomarkers for aggressive socially harmful behaviors in the preparation of products for screening aggressive socially harmful behaviors. The biomarkers for aggressive socially harmful behaviors include: a microbial ensemble and sociodemographic information, wherein the microbial ensemble includes Clostridium species (…). Clostridium ), genus *Trichophyton* Lachnospira ), Enterobacteriaceae ( Enterocloster ), Fusobacterium genus ( FusobacteriumMediterranean bacteria () Mediterraneibacter ), Broutbacterium spp. Blautia ), Flavonoids ( Flavonifractor ), symbiotic pectinobacterium ( Symbiopectobacterium ), Anaerobic rod-shaped bacteria ( Anaerostipes ), Dorperella spp. Dorea Bifidobacterium spp. Bifidobacterium ), Prevotella spp. Prevotella ) and Mycoplasma genus ( Mycoplasma ); Sociodemographic information includes: age, body mass index (BMI), education level, marital status, employment status, smoking and drinking.

[0013] Thirdly, the present invention provides a method for constructing a model for screening aggressive socially harmful behaviors, which includes the following steps: We acquire feature data representing markers of aggressive socially harmful behaviors in the training samples, and construct a model of aggressive socially harmful behaviors using a machine learning model.

[0014] In a preferred embodiment of the present invention, the machine learning model is selected from at least one of the following: random forest, K-nearest neighbors, Gaussian Bayes, Naive Bayes, support vector machine, AdaBoost, XGBoost, and DT decision tree; the feature data includes the abundance of microbial assemblages and sociodemographic information.

[0015] Fourthly, the present invention provides a method for screening aggressive socially harmful behaviors, the steps of which are performed by a computer; the method includes the following steps: Based on the characteristic data of the markers contained in the subjects, predictive results of the subjects' aggressive socially harmful behaviors are generated; The markers were selected from the markers of the aforementioned aggressive socially harmful behaviors.

[0016] In a preferred embodiment of the present invention, the feature data of the markers contained in the subject are input into a machine learning model; the predicted result of the subject's aggressive social harmful behavior is output; the machine learning model is selected from at least one of the following: random forest, K-nearest neighbors, Gaussian Bayes, Naive Bayes, support vector machine, AdaBoost, XGBoost and DT decision tree; the feature data includes the abundance of microbial assemblage and sociodemographic information.

[0017] Fifthly, the present invention provides an apparatus for screening aggressive socially harmful behaviors, comprising: an input module, a control module, and an output module; The input module is configured to: acquire feature data of markers contained in the subject, the markers being selected from the aforementioned markers of aggressive socially harmful behavior; The control module includes: a results analysis module, which is configured to: generate prediction results of the subject's aggressive socially harmful behavior based on the characteristic data of the markers contained in the subject; The output module is configured to output the predicted results of the subject's aggressive socially harmful behavior.

[0018] In a sixth aspect, the present invention provides a computer-readable storage medium storing an executable program instruction set, wherein at least one instruction or the executable program instruction set is loaded and executed by a processor to implement the method for screening aggressive socially harmful behaviors as described above.

[0019] The present invention has the following beneficial effects: This invention, through collecting fecal samples from male inmates exhibiting both aggressive and non-aggressive social harm, performing metagenomic sequencing, and using bioinformatics to statistically analyze the sequencing data, has identified gut microbiota associated with aggressive social harm behaviors.

[0020] Clostridium ( Clostridium ), genus *Trichophyton* Lachnospira ), Enterobacteriaceae ( Enterocloster ), Fusobacterium genus ( Fusobacterium Mediterranean bacteria () Mediterraneibacter ), Broutbacterium spp. Blautia ), Flavonoids ( Flavonifractor ), symbiotic pectinobacterium ( Symbiopectobacterium ), Anaerobic rod-shaped bacteria ( Anaerostipes ), Dorperella spp. Dorea Bifidobacterium spp. Bifidobacterium ), Prevotella spp. Prevotella ) and Mycoplasma genus ( Mycoplasma A total of 13 microbial genera were associated with individuals exhibiting aggressive and socially harmful behaviors. Specifically, the abundance of *Prevotella* and *Mycoplasma* species was significantly higher in individuals with aggressive and socially harmful behaviors than in those without such behaviors. *Clostridium* species were also more prevalent in individuals with aggressive and socially harmful behaviors. Clostridium ), genus *Trichophyton* Lachnospira ), Enterobacteriaceae ( Enterocloster ), Fusobacterium genus ( Fusobacterium Mediterranean bacteria () Mediterraneibacter ), Broutbacterium spp. Blautia ), Flavonoids ( Flavonifractor ), symbiotic pectinobacterium ( Symbiopectobacterium ), Anaerobic rod-shaped bacteria ( Anaerostipes ), Dorperella spp. Dorea Bifidobacterium spp. BifidobacteriumThe abundance of all 11 microorganisms was significantly lower than that of individuals with non-aggressive socially harmful behaviors. Using these 13 bacteria for screening for aggressive socially harmful behaviors yielded extremely high accuracy, and the screening method was non-invasive and objective.

[0021] When these 13 bacteria are combined with sociodemographic information as screening factors for aggressive and socially harmful behaviors, the screening accuracy is extremely high. This invention provides a new approach for screening and intervening in aggressive and socially harmful behaviors. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 The results of analyzing the relative abundance of the aggressive social harm behavior group and the non-aggressive social harm behavior group at the genus level using the LefSe software for the training set are shown in the figure. Figure 2 Box scatter plots of Clostridium, Anaerostipes, Bifidobacterium, Blautia, Dorea, and Enterococcus. Figure 3 Box scatter plots of Flavonoids, Fusobacterium, Lachnospira, Mediterranean bacteria, Prevotella, and Mycoplasma. Figure 4 Box scatter plot of Symbiopectobacterium; Figure 5 ROC curve results for screening aggressive socially harmful behaviors using 13 microbial biomarkers combined with sociodemographic information; Figure 6 Ordination graph for SHAP interpretability analysis; Figure 7 Scatter plot for SHAP Summary; Figure 8 The ROC curve results for 13 bacteria were plotted based on the random forest model. Figure 9 The ROC curve results for 13 microbial biomarkers combined with sociodemographic information, plotted based on the XGBoost model. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Where specific conditions are not specified in the embodiments, conventional conditions or conditions recommended by the manufacturer shall apply. Reagents or instruments whose manufacturers are not specified are all conventional products that can be purchased commercially.

[0025] Definition of noun The term "marker" broadly refers to any detectable compound or cell present in or derived from a sample, such as a protein, peptide, proteoglycan, glycoprotein, lipoprotein, cell, or any of the foregoing substances, that is differentiating molecule or differentiating fragment. For example, the detection of or binding to a specific antibody can indicate the presence of a specific antigen (e.g., a protein) in a sample. Here, a differentiating molecule or fragment is a molecule or fragment that, upon detection, indicates the presence or abundance of the aforementioned identified compound or cell. Markers can, for example, be isolated from the sample, measured directly in the sample, or detected or determined in the sample. Markers can, for example, be functional, partially functional, or non-functional. Markers may also be synonymous with "biomarker."

[0026] The term “subject” as used in this article can be understood as anyone involved in screening for aggressive socially harmful behaviors.

[0027] The terms "area under the curve" or "AUC" refer to the area under the receiver operating characteristic (ROC) curve, both of which are well-known in the field. AUC measurements are useful for comparing the accuracy of classifiers across the entire data range. Classifiers with higher AUCs have a greater ability to correctly classify unknowns between two target groups (e.g., groups with aggressive socially harmful behaviors and groups with non-aggressive socially harmful behaviors). ROC curves are useful for depicting the performance of specific features (e.g., any biomarkers and / or additional biomedical information described in this invention) when distinguishing between two groups (e.g., individuals in groups with aggressive socially harmful behaviors and groups with non-aggressive socially harmful behaviors).

[0028] In a first aspect, the present invention provides the application of a reagent for detecting microbial assemblages in the preparation of products for screening aggressive socially harmful behaviors, the reagent comprising: a reagent for detecting relative abundance data at the genus level of the microbial assemblages; The microbial assemblage includes Clostridium ( Clostridium ), genus *Trichophyton* Lachnospira ), Enterobacteriaceae (Enterocloster ), Fusobacterium genus ( Fusobacterium Mediterranean bacteria () Mediterraneibacter ), Broutbacterium spp. Blautia ), Flavonoids ( Flavonifractor ), symbiotic pectinobacterium ( Symbiopectobacterium ), Anaerobic rod-shaped bacteria ( Anaerostipes ), Dorperella spp. Dorea Bifidobacterium spp. Bifidobacterium ), Prevotella spp. Prevotella ) and Mycoplasma genus ( Mycoplasma ).

[0029] In a preferred embodiment of the present invention, the above-mentioned product is a reagent kit, a chip, or a screening device.

[0030] Furthermore, the kit may also include at least one of buffer, detection reagent, diluent, and washing solution, and is not limited thereto.

[0031] To improve the stability of reagents and extend their shelf life, those skilled in the art can add functional components such as stabilizers and protectants as needed. The reagents may be in various forms, including but not limited to solids, liquids, and semi-solids.

[0032] In a preferred embodiment of the present invention, the reagents are selected from primers and / or probes. Primers, for example, are selected from primers used to detect the 16S rDNA of these 13 microorganisms.

[0033] In one implementation, the abundance of the aforementioned microbial combination in the subject sample is obtained by metagenomic sequencing and analysis.

[0034] Secondly, this invention provides the application of biomarkers for aggressive socially harmful behaviors in the preparation of products for screening aggressive socially harmful behaviors. The biomarkers for aggressive socially harmful behaviors include: a microbial ensemble and sociodemographic information, wherein the microbial ensemble includes Clostridium species (…). Clostridium ), genus *Trichophyton* Lachnospira ), Enterobacteriaceae ( Enterocloster ), Fusobacterium genus ( Fusobacterium Mediterranean bacteria () Mediterraneibacter ), Broutbacterium spp. Blautia ), Flavonoids ( Flavonifractor ), symbiotic pectinobacterium ( Symbiopectobacterium ), Anaerobic rod-shaped bacteria ( Anaerostipes ), Dorperella spp. Dorea Bifidobacterium spp. Bifidobacterium ), Prevotella spp. Prevotella ) and Mycoplasma genus ( Mycoplasma ).

[0035] Sociodemographic information includes: age, body mass index (BMI), education level, marital status, employment status, smoking and drinking.

[0036] Education level can also be equated with "educational level", and work status can also be equated with "work condition", including three types: full-time, part-time and unemployed.

[0037] Smoking information includes three types: smokers, quitters, and non-smokers.

[0038] Drinking information includes drinking frequency, such as being categorized as follows: (1) almost daily; (2) at least once a week; (3) at least once a month; (4) occasionally a year; (5) rarely or never drink alcohol.

[0039] The above information can be obtained through a questionnaire that includes the aforementioned sociodemographic information.

[0040] In one implementation, Body Mass Index (BMI) can also be equivalent to obtaining height and weight data.

[0041] BMI = weight (kg) / height² (m²).

[0042] Thirdly, the present invention provides a method for constructing a model for screening aggressive socially harmful behaviors, which includes the following steps: We acquire feature data representing markers of aggressive socially harmful behaviors in the training samples, and construct a model of aggressive socially harmful behaviors using a machine learning model.

[0043] In a preferred embodiment of the present invention, the machine learning model is selected from at least one of the following: random forest, K-nearest neighbors, Gaussian Bayes, Naive Bayes, support vector machine, AdaBoost, XGBoost, and DT decision tree; the feature data includes the abundance of microbial assemblages and sociodemographic information.

[0044] In a preferred embodiment of the present invention, the machine learning model is a random forest.

[0045] Fourthly, the present invention provides a method for screening aggressive socially harmful behaviors, the steps of which are performed by a computer; the method includes the following steps: Based on the characteristic data of the markers contained in the subjects, predictive results of the subjects' aggressive socially harmful behaviors are generated; The markers were selected from the markers of the aforementioned aggressive socially harmful behaviors.

[0046] In a preferred embodiment of the present invention, the feature data of the markers contained in the subject are input into a machine learning model; the predicted result of the subject's aggressive social harmful behavior is output; the machine learning model is selected from at least one of the following: random forest, K-nearest neighbors, Gaussian Bayes, Naive Bayes, support vector machine, AdaBoost, XGBoost and DT decision tree; the feature data includes the abundance of microbial assemblage and sociodemographic information.

[0047] Fifthly, the present invention provides an apparatus for screening aggressive socially harmful behaviors, comprising: an input module, a control module, and an output module; The input module is configured to: acquire feature data of markers contained in the subject, the markers being selected from the aforementioned markers of aggressive socially harmful behavior; The control module includes: a results analysis module, which is configured to: generate prediction results of the subject's aggressive socially harmful behavior based on the characteristic data of the markers contained in the subject; The output module is configured to output the predicted results of the subject's aggressive socially harmful behavior.

[0048] Specifically, the results analysis module is configured to input the feature data of the markers contained in the subjects into a pre-established machine learning model to generate prediction results of the subjects' aggressive socially harmful behaviors.

[0049] In a sixth aspect, the present invention provides a computer-readable storage medium storing an executable program instruction set, wherein at least one instruction or the executable program instruction set is loaded and executed by a processor to implement the method for screening aggressive socially harmful behaviors as described above.

[0050] The term "computer-readable storage medium" includes data storage media or cloud drives based on a single physical entity such as a CD, CD-ROM, hard disk, optical storage medium, or magnetic disk. Furthermore, the term also includes data storage media composed of physically separate entities that are effectively connected to each other, preferably in a query-and-search manner, to provide the aforementioned data set.

[0051] In a seventh aspect, the present invention provides an apparatus for screening aggressive socially harmful behaviors. The apparatus includes a processor and a memory, the memory storing a set of executable program instructions, which are loaded and executed by the processor to implement the above-described method for screening aggressive socially harmful behaviors.

[0052] Specifically, the electronic device may include a memory, a processor, a bus, and a communication interface, which are electrically connected directly or indirectly to each other to enable data transmission or interaction. For example, these components may be electrically connected to each other via one or more buses or signal lines. The processor may process information and / or data related to target identification to perform one or more functions described in this application.

[0053] The memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc.

[0054] A processor can be an integrated circuit chip with signal processing capabilities. This processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0055] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0056] Example 1 This embodiment involves the screening of intestinal flora biomarkers.

[0057] 1. Sample collection The personnel in this embodiment of the invention all came from a certain prison. Based on whether the crimes of the male inmates involved actual acts of aggression against others (such as bodily harm, robbery, murder or attempted murder), fecal samples were collected from 121 male inmates with aggressive social harm and 113 male inmates with non-aggressive social harm (such as job-related crimes, fraud, operating a casino, drug trafficking).

[0058] Based on whether the crimes committed by male inmates involved actual acts of aggression against others (such as bodily harm, robbery, murder, or attempted murder), basic information questionnaires were collected from 121 male inmates with aggressive social harm and 113 male inmates with non-aggressive social harm. The questionnaires included information on age, height, weight, education, marital status, employment, smoking, and alcohol consumption.

[0059] (1) The questionnaire collection of sociodemographic information specifically includes the following: Age: years old; Height: cm; Weight: kg; Education level: □ Illiterate □ Primary school or below □ Junior high school, high school / technical secondary school □ Junior college, bachelor's degree or above; Marital status: □ Married □ Divorced □ Widowed □ Single; Employment status before imprisonment: □ Full-time □ Part-time □ Unemployed; Did you smoke before you went to prison? ① Smoked; ② Quit smoking; ③ Don't smoke; Before you went to prison, did you drink alcohol: ① almost every day; ② at least once a week; ③ at least once a month; ④ occasionally a few times a year; ⑤ hardly ever or never drink alcohol.

[0060] All 121 male inmates in the above cases of aggressive social harm were sentenced because they actually committed such acts. All 113 male inmates in the above cases of non-aggressive social harm were sentenced because they actually committed such acts.

[0061] (2) Based on the principle that gut microbiota measurement is not affected by factors such as some diseases, drugs, and substance addiction, and taking into account the principle of accuracy and objectivity in psychological characteristic measurement, the exclusion criteria for participants were formulated as follows: i. Inclusion criteria: a. Inmates who have been imprisoned for 3 months or more; b. The age range is 18-60 years old, and the offenders should be distributed as evenly as possible across all age groups; c. Has a certain level of education and is able to read and complete the questionnaire independently; d. The offender signs an informed consent form; ii: Exclusion criteria: a. Those who refuse to sign the informed consent form; b. Has a history of drug use or drug-related crimes; c. Alcohol or drug addiction; d. History of use of psychotropic drugs within the past month (excluding use of sleeping pills); e. History of head injury and unconsciousness for more than 10 minutes at the time of injury; f. Neurological disorders (previous diagnoses include cerebral hemorrhage, cerebral infarction, Alzheimer's disease, Parkinson's disease, and dementia). g. Mental disorders (previous diagnoses include schizophrenia, bipolar disorder, depression, intellectual disability, etc.); h. Cancer history (including all types of cancer); i. Those who have undergone major gastrointestinal surgery within the past 5 years (such as cholecystectomy, appendectomy, gastric / intestinal resection, etc.). j. Use of antibiotics or probiotic preparations within the past month; k. Those who are no longer able to serve their sentences in prison due to certain reasons.

[0062] (3) Sample grouping: The 121 male prisoners with aggressive social harm behaviors and the 113 male prisoners with non-aggressive social harm behaviors were grouped into training set and test set.

[0063] The data was divided into a training set and a test set at a ratio of 4:1, with 188 participants in the training set and 46 participants in the test set. The training set included 97 participants exhibiting aggressive and socially harmful behaviors, and 91 participants exhibiting non-aggressive and socially harmful behaviors. The test set included 24 participants exhibiting aggressive and socially harmful behaviors, and 22 participants exhibiting non-aggressive and socially harmful behaviors. The ratio of participants in the training set to those in the test set was approximately 1:1.

[0064] The data samples in the training set are used for marker selection and model building.

[0065] 2. DNA extraction, library construction, and sequencing: (1) All genomic DNA in the fecal sample from step 1 was fragmented into segments of approximately 300 bp using enzyme digestion. The entire library was prepared through steps such as end repair, addition of A-tails, addition of sequencing adapters, purification, and PCR amplification. Sequencing was performed using paired-end sequencing based on Illumina high-throughput sequencing.

[0066] (2) The raw data obtained from sequencing is processed by removing adapters, low-quality sequences, and human (host) sequences to obtain clean data.

[0067] (3) Perform de novo metagenomic assembly and mix the unused reads from each sample together to discover information on low-abundance species in the samples.

[0068] (4) Combine the ORF prediction results of each sample and the mixed assembly, use CD-HIT software to remove redundancy, select the representative sequence as Unigenes, cluster by default with identity 95% and coverage 90%, and select the longest sequence as the representative sequence.

[0069] (5) Use Bowtie2 to align the CleanData of each sample to each Unigene sequence, calculate the number of reads aligned to each Unigene in each sample, and calculate the abundance information of each gene in each sample based on the number of gene reads aligned and the gene length.

[0070] (6) Using DIAMOND software, the Unigenes protein sequences were compared with the bacterial, fungal, archaea and virus sequences extracted from the NCBI NR (Version: 2022.05) database. The LCA algorithm (applied to the systematic classification of MEGAN software) was used to obtain the final species annotation information.

[0071] (7) Combine LCA annotation results with gene abundance information to obtain abundance information at each taxonomic level (kingdom, phylum, class, order, family, genus, species).

[0072] 3. LEfSe analysis for biomarker screening The relative abundance data at the genus level were analyzed using LEfSe software. The default LDA Score (Linear Discriminant Analysis Score) filter value was set to 2 (absolute value). The results are as follows: Figure 1 As shown, Figure 1 In this context, "Non-violent" refers to the group of non-aggressive socially harmful behaviors, while "violent" refers to the group of aggressive socially harmful behaviors.

[0073] Linear discriminant analysis (LDA) scoring is an analytical tool capable of mining and interpreting biomarkers from high-dimensional data, exhibiting significant statistical significance and biological correlation. Eleven microorganisms were identified as significantly reduced in the aggressive socially harmful behavior group: *Clostridium*, *Lachnospira*, *Enterocloster*, *Fusobacterium*, *Mediterraneibacter*, *Blautia*, *Flavonifractor*, *Symbiopectobacterium*, *Anaerostipes*, *Dorea*, and *Bifidobacterium*. Two microorganisms were identified as significantly increased in the aggressive socially harmful behavior group: *Prevotella* and *Mycoplasma*. Box scatter plots of Clostridium, Anaerostipes, Bifidobacterium, Blautia, Dorea, and Enterococcus are referenced. Figure 2 As shown, the box plots for Flavonoids, Fusobacterium, Lachnospira, Mediterranean bacteria, Prevotella, and Mycoplasma are referenced. Figure 3 As shown, the box plot of Symbiopectobacterium is referenced. Figure 4 As shown.

[0074] The results showed that Prevotella spp. ( ) were associated with individuals exhibiting aggressive and socially harmful behaviors. PrevotellaThe abundance of microorganisms in individuals with aggressive and harmful social behaviors was significantly higher than that in individuals with non-aggressive and harmful social behaviors. Among individuals with aggressive and harmful social behaviors, the abundance of 11 microorganisms other than *Prevotella* (clostridium, Lachnospira, Enterococcus, Fusobacterium, Mediterranean bacteria, Blautia, Flavonoids, Symbiopectobacterium, Anaerostipes, Dorea, and Bifidobacterium) was significantly lower than that in individuals with non-aggressive and harmful social behaviors.

[0075] The above results indicate that 13 genera are associated with aggressive socially harmful behaviors in the human gut microbiota.

[0076] Example 2 This embodiment establishes a random forest model based on the relative abundance data of 13 microbial biomarkers in the training set of Embodiment 1.

[0077] The relative abundance data of 13 microbial biomarkers were subjected to CLR central log ratio transformation.

[0078] Within the training set, the optimal model parameters under 5-CV are calculated using the random forest algorithm and Bayesian parameter tuning, and a random forest classification model for screening aggressive social harmful behaviors is established.

[0079] ROC curves for the training and test sets of Example 1 were plotted using a random forest classification model, and the AUC values ​​were obtained. The results are referenced... Figure 8 As shown, the results indicate that the 13 microbial biomarkers have a certain degree of accuracy in screening for aggressive socially harmful behaviors. The AUC value of the test set is 0.6636, the cutoff value is 0.5, the F1 value is 0.6531, the recall value is 0.6400, and the accuracy is 0.6383.

[0080] Example 3 This embodiment establishes a random forest model based on the relative abundance data of 13 microbial biomarkers in the training set of Embodiment 1 and sociodemographic information.

[0081] Within the training set, the optimal model parameters under 5-CV are calculated using the random forest algorithm and Bayesian parameter tuning, and a random forest classification model for screening aggressive social harmful behaviors is established.

[0082] The ROC curve for the test set of Example 1 was plotted using a random forest classification model, and the AUC value was obtained. The results are referenced... Figure 5 As shown, the results indicate that the screening of aggressive socially harmful behaviors using 13 microbial biomarkers combined with sociodemographic information has extremely high accuracy. The AUC value of the ROC curve plotted based on the test set for this screening and scoring method, with an optimal cutoff value of 0.5, F1 score of 0.8148, recall of 0.8800, and ACC (accuracy) of 0.7872, is 0.9019.

[0083] Shap analysis results are all for Figure 5 The random forest model was used, and the SHAP interpretability analysis ranking graph was referenced. Figure 6 and Figure 7 As shown.

[0084] Figure 7 In the scatter plot, the further to the right a point is, the higher the risk of aggressive socially harmful behavior; the redder the color, the higher the SHAP eigenvalue, indicating a higher risk of aggressive socially harmful behavior.

[0085] Example 4 This embodiment establishes an XGBoost model based on the relative abundance data of 13 microbial biomarkers and sociodemographic information from the training set of Example 1. The ROC curve of the test set of Example 1 is plotted using the XGBoost model, and the AUC value is obtained. Figure 9 As shown, the AUC value of the ROC curve plotted based on the training set is 0.9072, the AUC value of the ROC curve plotted based on the test set is 0.8618, the optimal cutoff value is 0.5, the F1 score is 0.7692, the recall is 0.8000, and the ACC (accuracy) is 0.7447.

[0086] In summary, this invention is the first to discover that eight genera of microorganisms—Prevotella, Clostridium, Lachnospira, Phascolarctobacterium, Ruminococcus, Enterococcus, Mediterranean bacteria, and Blautia—are associated with individuals exhibiting aggressive and socially harmful behaviors. Specifically, the abundance of seven of these microorganisms is significantly lower than that of individuals exhibiting non-aggressive and socially harmful behaviors, while the abundance of one microorganism is significantly higher. When these eight microorganisms are combined with sociodemographic information as screening factors for aggressive and socially harmful behaviors, the accuracy is high, and the screening method is non-invasive and objective. This invention provides a new approach for the screening and intervention of aggressive and socially harmful behaviors.

[0087] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. The application of reagents for detecting microbial assemblages in the preparation of products for screening aggressive socially harmful behaviors, characterized in that, The reagents include: reagents for detecting relative abundance data at the genus level of microbial assemblages; The microbial assemblage includes Clostridium species ( Clostridium ), genus *Trichophyton* Lachnospira ), Enterobacteriaceae ( Enterocloster ), Fusobacterium genus ( Fusobacterium Mediterranean bacteria () Mediterraneibacter ), Broutbacterium spp. Blautia ), Flavonoids ( Flavonifractor ), symbiotic pectinobacterium ( Symbiopectobacterium ), Anaerobic rod-shaped bacteria ( Anaerostipes ), Dorperella spp. Dorea Bifidobacterium spp. Bifidobacterium ), Prevotella spp. Prevotella ) and Mycoplasma genus ( Mycoplasma ).

2. The application according to claim 1, characterized in that, The products mentioned are reagent kits, chips, or screening devices.

3. The application according to claim 1, characterized in that, The reagents are selected from primers and / or probes.

4. The application of biomarkers for aggressive socially harmful behavior in the preparation of products for screening aggressive socially harmful behavior, characterized in that, The markers of aggressive socially harmful behavior include: microbial assemblages and sociodemographic information, wherein the microbial assemblages include Clostridium species (…). Clostridium ), genus *Trichophyton* Lachnospira ), Enterobacteriaceae ( Enterocloster ), Fusobacterium genus ( Fusobacterium Mediterranean bacteria () Mediterraneibacter ), Broutbacterium spp. Blautia ), Flavonoids ( Flavonifractor ), symbiotic pectinobacterium ( Symbiopectobacterium ), Anaerobic rod-shaped bacteria ( Anaerostipes ), Dorperella spp. Dorea Bifidobacterium spp. Bifidobacterium ), Prevotella spp. Prevotella ) and Mycoplasma genus ( Mycoplasma ); The sociodemographic information includes: age, body mass index (BMI), education level, marital status, employment status, smoking, and alcohol consumption.

5. A method for constructing a model for screening aggressive socially harmful behaviors, characterized in that, It includes the following steps: Obtain feature data representing the markers of aggressive socially harmful behavior as described in claim 4 from the training samples, and construct a model of aggressive socially harmful behavior using a machine learning model.

6. The construction method according to claim 5, characterized in that, The machine learning model is selected from at least one of the following: random forest, K-nearest neighbors, Gaussian Bayes, Naive Bayes, support vector machine, AdaBoost, XGBoost, and DT decision tree; the feature data includes the abundance of microbial assemblages and sociodemographic information.

7. A method for screening aggressive socially harmful behaviors, characterized in that, The steps of the method are performed by a computer; the method includes the following steps: Based on the characteristic data of the markers contained in the subjects, predictive results of the subjects' aggressive socially harmful behaviors are generated; The marker is selected from the markers of aggressive socially harmful behavior described in claim 4.

8. The method for screening aggressive socially harmful behaviors according to claim 7, characterized in that, The feature data of the markers contained in the subjects are input into the machine learning model; the output is the prediction result of the subjects' aggressive social harmful behavior; the machine learning model is selected from at least one of the following: random forest, K-nearest neighbors, Gaussian Bayes, Naive Bayes, support vector machine, AdaBoost, XGBoost and DT decision tree; the feature data includes the abundance of microbial assemblage and sociodemographic information.

9. A device for screening aggressive socially harmful behaviors, characterized in that, It includes: Input module, control module, and output module; The input module is configured to: acquire feature data of markers contained in the subject, wherein the markers are selected from the markers of aggressive socially harmful behavior as described in claim 4; The control module includes: a result analysis module, which is configured to: generate a prediction result of the subject's aggressive socially harmful behavior based on the characteristic data of the markers contained in the subject; The output module is configured to output the predicted results of the subject's aggressive socially harmful behavior.

10. A computer-readable storage medium, characterized in that, The storage medium stores a set of executable program instructions, wherein at least one instruction and the set of executable program instructions are loaded and executed by a processor to implement the method for screening aggressive socially harmful behaviors as described in any one of claims 7-8.

Images