Noise auditory impairment risk assessment method and system

Abstract

The invention discloses a noise auditory impairment risk assessment method and system, and particularly relates to the technical field of auditory impairment diagnosis, and the method comprises the steps: obtaining the auditory characterization data of I sample individuals, I being an integer greater than zero; the auditory characterization data comprises individual gene data, environmental exposure data, individual physiological data and behavior data; inputting the auditory representation data of the I sample individuals into a pre-constructed auditory analysis model to obtain a first screening result, the first screening result comprising Q extreme individuals; according to the comprehensive method based on multi-dimensional data, an accurate evaluation model and big data analysis, not only can the cause of current auditory impairment be analyzed in detail, but also the future risk can be predicted and customized protection suggestions can be provided for individuals, so that the auditory health management efficiency and effect are improved on the two aspects of prevention and treatment.

Description

Technical Field

[0001] The present invention relates to the technical field of hearing damage diagnosis, and more specifically, to a method and system for assessing the risk of noise-induced hearing damage. Background Art

[0002] Although traditional noise-induced hearing loss risk assessment methods can reflect an individual's hearing loss status, they have obvious limitations in predicting genetic susceptibility and quantifying damage risk.

[0003] Among the existing methods, for example, the Chinese patent application with publication number CN117297596A discloses an auditory pathway evaluation and analysis device and method thereof, including: an SSAEP test signal generation component, a tinnitus simulation signal generation component, a composite signal generation component, a stimulation component, an electroencephalogram recording component, a data processing and analysis module, an amplitude determination module, and a comparison module. It is suitable for evaluating the degree of damage to the auditory system of tinnitus subjects, by generating a simulated tinnitus signal that matches the tinnitus characteristics of the tinnitus subjects, and associating and combining the simulated tinnitus signal with the SSAEP test signal to form a real and comprehensive composite excitation signal, so as to more accurately evaluate the overall state of the auditory pathway. By generating a number of different SSAEP test signals, the SSAEP test signal is a sine wave with different frequencies, amplitudes or phases, and different SSAEP responses can be detected in a short time. Although the above method reduces the test time and can more quickly and comprehensively evaluate the function of the auditory pathway, it is found that the above method and the prior art have at least the following defects after studying and applying the above method and the prior art: Although evaluating the status of the auditory pathway based only on steady-state auditory evoked potential (SSAEP) signals and simulated tinnitus signals can capture the physiological characteristics of tinnitus patients, it ignores important information such as genetic factors and environmental exposure factors that may cause or aggravate hearing damage, which in turn leads to one-sided evaluation results.

[0004] To this end, the present invention provides a method and system for assessing the risk of noise-induced hearing loss. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides a noise-induced hearing loss risk assessment method and system to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for assessing the risk of noise-induced hearing loss, comprising: Step 1: Obtain auditory representation data of I sample individuals, where I is an integer greater than zero; the auditory representation data includes individual genetic data, environmental exposure data, individual physiological data, and behavioral data; Step 2: Input the auditory representation data of I sample individuals into a pre-built auditory analysis model to obtain a first screening result, wherein the first screening result includes Q extreme individuals; Step 3: Obtain Q extreme individuals output by the auditory analysis model, extract the qth extreme individual for whole exome sequencing, and obtain a set of gene variation sites; q = 1, 2, ..., Q; Step 4: Functionally annotate the set of gene variant sites, identify the set of suspected high-risk sites and the corresponding annotation results; Step 5: Based on the PRS model, evaluate the set of suspected high-risk sites to obtain the z-score value of the qth extreme individual; determine the injury risk level based on the z-score value; Step 6: Feedback the corresponding protection measures plan to the client for display based on the damage risk level.

[0007] Furthermore, the specific training process of the auditory analysis model includes: Pre-collect Y groups of auditory representation data, where Y is an integer greater than 1, set corresponding screening results for the auditory representation data, mark the digital labels of the screening results as prediction labels, and convert the auditory representation data and the corresponding prediction labels into a corresponding set of feature vectors; Each group of feature vectors is used as the input of the auditory analysis model. The auditory analysis model takes a group of prediction labels corresponding to each group of auditory representation data as output, and takes the actual prediction labels corresponding to each group of auditory representation data as prediction targets, where the actual prediction labels are pre-set prediction labels corresponding to the auditory representation data; minimizing the sum of prediction errors of all auditory representation data is used as the training target; the auditory analysis model is trained until the sum of prediction errors converges and the training is stopped; the auditory analysis model is a random forest, support vector machine or deep neural network model, and a first screening result is obtained.

[0008] Further, the method of extracting the qth extreme individual for whole exome sequencing to obtain a set of gene variation sites includes: Step a1: extracting and purifying DNA from the blood sample of the qth extreme individual using a specific reagent to obtain purified DNA; The specific reagent is a lysis buffer, a chelating agent or a precipitating agent; Step a2: mechanically or enzymatically decomposing the purified DNA to obtain DNA fragments, wherein the size of the DNA fragments is in the range of 100-500 bp; Step a3: using a specific exon probe pool to capture DNA fragments in the exon region, obtaining exon sequence fragments, and removing non-exon sequences; Step a4: connecting the adapter containing the sequencing adapter to both ends of the exon sequence fragment, and performing PCR amplification on the exon sequence fragment to obtain the exon sequence amplified fragment; Step a5: sequencing the exon sequence amplified fragments by a sequencer connected to the sequencing platform to obtain sequencing data; Step a6: Perform quality control on the sequencing data to obtain high-quality sequencing data, and compare the obtained high-quality sequencing data with the reference genome to determine the set of gene variation sites.

[0009] Furthermore, the method of performing quality control on the sequencing data to obtain high-quality sequencing data includes: Step b1: using a first tool to evaluate the sequencing data, removing the sequencing data with bases below the standard quality, and obtaining first sequencing data; Step b2: using a second tool to process the first sequencing data, remove low-quality bases at the ends of the read segments, and cut out adapter fragments to obtain second sequencing data; Step b3: setting every F bases of the sequencer as a sliding window, performing sliding inspection on the second sequencing data, obtaining the average quality score of the bases in each sliding window, removing the window segments below the average quality score threshold, and obtaining the third sequencing data; F is an integer greater than 1; Step b4: Use a third tool to mark the repetitive sequences generated by PCR amplification in the third sequencing data, and use a decontamination tool to clean up non-human sequences to obtain high-quality sequencing data.

[0010] Further, the method of comparing the obtained high-quality sequencing data with the reference genome to determine the set of gene variation sites includes: Step c1: Use the BWA-MEM algorithm to compare high-quality sequencing data with the target genome to obtain a SAM file; Step c2: using the fifth tool to convert the SAM file into a compressed BAM format, sorting by genomic position to obtain a BAM file, and using the sixth tool to mark duplicate sequences in the PCR process, correct alignment offsets, and filter out mismatched or low-quality alignment fragments; Step c3: Use the seventh tool to recalibrate the BAM file, extract the genome variation information, and generate a preliminary VCF file; Step c4: Use the VQSR module of the seventh tool to calibrate the detected variant sites and set the variant quality threshold to screen the matching gene variant site set.

[0011] Furthermore, the method of functionally annotating the gene variation site set, identifying the suspected high-risk site set and the corresponding annotation results includes: Step d1: Preliminarily screening the gene variation site set based on the known SNP site data to obtain a screened gene variation site set; the screened gene variation site set includes group E gene variation sites; Step d2: performing association analysis on the e-th group of gene variation sites, and calculating the association coefficient of the e-th group of gene variation sites, e=1, 2, ..., E; Step d3: Preset a correlation coefficient threshold, the correlation coefficient threshold includes and ,in ; The correlation coefficient of the gene variation site in group e Compare with the preset correlation coefficient threshold; like , then the gene variation sites in group e are marked as suspected high-risk sites; like , then the gene variation sites in group e are marked as medium-risk sites; like , then the gene variation sites in group e are marked as low-risk sites; Step d4: annotate the suspected high-risk sites using a functional annotation tool to obtain annotation results, set e=e+1, and return to step d2, the functional annotation tool is Annovar or VEP; the annotation results include gene positions and functional regions: the gene positions include coding regions, promoters or non-coding regions; Step d5: Repeat steps d2 to d4 until the cycle ends when e=E, and obtain a set of suspected high-risk sites and corresponding annotation results.

[0012] Furthermore, the sites of the PRS model include rs10783780, rs10794566, rs10794567, rs3887954, rs58011943, rs11836060, rs16973424 and rs74905175.

[0013] Furthermore, based on the PRS model, the suspected high-risk site set is evaluated, and the method for obtaining the z-score value of the qth extreme individual includes: Step s1: extracting the effect value of the hth suspected site in the suspected high-risk site set based on the association coefficient; h=1, 2, ..., H; H is the number of suspected sites in the suspected high-risk site set; Step s2: Calculate the PRS score of the qth extreme individual; Step s3: Repeat step s2 until q=Q, and the cycle ends, obtaining the PRS scores of Q extreme individuals; Step s4: Convert the PRS score of the qth extreme individual into a z-score value to obtain the z-score value of the qth extreme individual.

[0014] Furthermore, the injury risk levels include the first level, the second level, the third level, the fourth level and the fifth level; the first level> the second level> the third level> the fourth level> the fifth level.

[0015] Furthermore, the individual genetic data include KCNQ4, MYO6, MYO7A, MYO15A, OTOF, GJB2, SLC26A4, GRM7, TECTA, CDH23, ESRRB, HSP70, CAT and SOD2; the environmental exposure data include noise exposure intensity and noise exposure time; the individual physiological data include age, gender, genetic background, health status and historical history of hearing loss; the behavioral data include high-frequency sound sensitivity, tinnitus symptoms and tolerance to noise.

[0016] The second aspect is the noise-induced hearing loss risk assessment system, including: A data acquisition module, used to obtain auditory representation data of I sample individuals, where I is an integer greater than zero; the auditory representation data includes individual genetic data, environmental exposure data, individual physiological data and behavioral data; An analysis module, configured to input the auditory representation data of I sample individuals into a pre-built auditory analysis model to obtain a first screening result, wherein the first screening result includes Q extreme individuals; The sequencing module is used to obtain Q extreme individuals output by the auditory analysis model, extract the qth extreme individual for whole exome sequencing, and obtain a set of gene variation sites; q = 1, 2, ..., Q; The identification module is used to perform functional annotation on the gene variation site set, identify the suspected high-risk site set and the corresponding annotation results; The evaluation module evaluates the suspected high-risk site set based on the PRS model to obtain the z-score value of the qth extreme individual; and determines the injury risk level according to the z-score value; The display module is used to feed back the corresponding protection measures plan to the client for display according to the damage risk level.

[0017] Technical effects and advantages of the present invention: 1. The present invention constructs auditory characterization data by utilizing genetic data, environmental exposure data, individual physiological data, and behavioral data to achieve a multi-dimensional comprehensive analysis of hearing loss, reveal the multi-factor causes and interactions of hearing loss, and thus achieve precise analysis; utilizing the genetic data of extreme individuals, analyzing the gene variation sites known to be related to hearing function, standardizing them according to the PRS score (such as calculating the Z-score value), and corresponding them to the risk level of hearing loss, dividing them into multiple assessment levels, and formulating more stringent hearing protection measures for high-risk individuals.

[0018] 2. A comprehensive approach based on multi-dimensional data, precise assessment models and big data analysis can not only analyze the causes of current hearing loss in detail, but also predict future risks and provide individuals with tailored protection recommendations, thereby improving the efficiency and effectiveness of hearing health management at both the prevention and treatment levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flow chart of the noise-induced hearing loss risk assessment method of Example 1; Figure 2 A flow chart of the method for identifying a set of suspected high-risk sites and corresponding annotation results in Example 1; Figure 3 This is a flow chart of the method for obtaining the z-score value of the qth extreme individual in Example 1; Figure 4 This is the extremely susceptible and extremely resistant individual characteristic diagram of Example 1; Figure 5 This is the basic information table of the verification phase of Example 1; Figure 6 Manhattan plot for association analysis for risk assessment of noise-induced hearing loss; Figure 7 QQ plot of the results of whole exome association analysis; Figure 8 PRS distribution diagram of noise-induced hearing loss group and non-noise-induced hearing loss group; Fig. 9 The forest plot of PRS and risk of noise-induced hearing loss; Fig.10 This is a schematic diagram of the structure of the noise-induced hearing loss risk assessment system of Example 2. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0022] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and a similar second unit may be referred to as a first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0023] Example 1 See also Figure 1 As shown, this embodiment discloses a method for assessing the risk of noise-induced hearing loss, which is implemented based on a client and a cloud server, wherein the client is remotely connected to the cloud server for communication, and the method includes: Step 1: Obtain auditory representation data of I sample individuals, where I is an integer greater than zero; The auditory representation data includes individual genetic data, environmental exposure data, individual physiological data and behavioral data; It should be understood that the sample individuals include but are not limited to factory workers, construction workers, musicians, military personnel, music lovers who love high-decibel sounds, and other individuals who are in a high-noise environment for a long time.

[0024] The individual gene data include but are not limited to KCNQ4, MYO6, MYO7A, MYO15A, OTOF, GJB2, SLC26A4, GRM7, TECTA, CDH23, ESRRB, HSP70, CAT and SOD2.

[0025] It is explained that: KCNQ4, MYO6, MYO7A, MYO15A, OTOF are genes related to cochlear structure and function; GJB2, SLC26A4, GRM7 are genes related to neurotransmitter transmission; TECTA, CDH23, ESRRB are genes related to auditory conduction and inner ear protection; HSP70, CAT, SOD2 are genes for inner ear protection and repair. The individual gene data is obtained by extracting DNA from the blood or saliva of the sample individual and performing whole exon sequencing using high-throughput sequencing technology. The high-throughput sequencing technology is an existing technology, and any method of obtaining individual gene data is acceptable, and is not specifically limited here.

[0026] The environmental exposure data includes noise exposure intensity, noise exposure time and noise frequency; wherein the noise exposure intensity refers to the volume of the noise; the noise exposure time refers to the cumulative duration that the sample individual is exposed to the noise environment; the noise frequency refers to the pitch of the noise; the noise exposure intensity, noise exposure time and noise frequency are all collected by the sample individual through the noise monitoring equipment or smart watch carried by the sample individual.

[0027] The individual physiological data includes age, gender, genetic background, health status and history of hearing loss; the individual physiological data is collected and obtained in advance through questionnaires and other methods and pre-stored in the system database. Specifically, with age, the inner ear hair cells and auditory nerves gradually degenerate, and the risk of noise-induced hearing loss also increases; men are more likely to be exposed to high-noise environments due to the nature of their work (such as heavy industry, construction, military, etc.), so the incidence of noise-induced hearing loss is higher than that of women; genetic genes play a greater role in noise-induced hearing loss, and some gene mutations may cause individuals to be more sensitive to noise. For example, genes such as GJB2 and GRM7 are related to cell protection mechanisms, signal transmission and cell repair in the cochlea, and mutations may increase the risk of noise-induced hearing loss. Health problems, such as high blood pressure, diabetes and immune system diseases, can affect the blood circulation and cell health of the cochlea, thereby increasing the risk of noise-induced hearing loss. These chronic diseases weaken the resistance of inner ear cells to noise, making them more susceptible to damage. People with a history of hearing loss (such as hearing loss caused by occupational noise exposure, explosions, or ear diseases) are more likely to have their hearing loss aggravated by re-exposure to noise. Historical hearing loss causes damage to the hair cells and nerve connections in the cochlea, making individuals more sensitive to noise exposure.

[0028] The behavioral data includes high-frequency sound sensitivity, tinnitus symptoms, and noise tolerance. The high-frequency sound sensitivity refers to the individual's auditory response to high-frequency sounds (e.g., above 3000Hz). Individuals with higher high-frequency sensitivity can usually detect signs of noise damage more quickly in the early stages, because the high-frequency part of hearing is often the first to be damaged. Tinnitus symptoms are a common symptom of noise-induced hearing damage, which manifests as a continuous "buzzing" or "hissing" sound in the ear, even when there is no external noise. Tinnitus is usually the result of damage to the inner ear hair cells or auditory nerves, indicating that the auditory system has been damaged; noise tolerance refers to the ability of an individual to remain comfortable in a high-intensity or harsh noise environment. Low tolerance usually means that the individual is more sensitive to noise and feels uncomfortable or even painful at a lower noise intensity. The high-frequency sound sensitivity, tinnitus symptoms, and noise tolerance are obtained through hearing tests, questionnaires, and noise tolerance tests.

[0029] It should be noted that the auditory characterization data is pre-collected by a technician in this field, and after pre-processing the data of I sample individuals, it is entered into the system database through the client. The pre-processing method is data cleaning, standardization or normalization. The pre-processing method is a prior art and will not be explained in detail here.

[0030] Step 2: Input the auditory representation data of I sample individuals into a pre-built auditory analysis model to obtain a first screening result, which includes Q extreme individuals.

[0031] It should be noted that the extreme individuals are those who suffer severe hearing loss under low noise exposure and those who suffer no obvious damage under noise exposure.

[0032] In implementation, the specific training process of the auditory analysis model includes: Collect Y groups of auditory characterization data in advance, where Y is an integer greater than 1, set corresponding screening results for the auditory characterization data, collect Y groups of different auditory characterization data, and those skilled in the art set corresponding screening results for the Y groups of different auditory characterization data in turn according to actual experience, see Figure 8 shown.

[0033] Marking the digital labels of the screening results as prediction labels, and converting the auditory representation data and the corresponding prediction labels into a corresponding set of feature vectors; Each set of feature vectors is used as the input of the auditory analysis model. The auditory analysis model uses a set of prediction labels corresponding to each set of auditory representation data as output, and uses the actual prediction labels corresponding to each set of auditory representation data as the prediction target. The actual prediction labels are the pre-set prediction labels corresponding to the auditory representation data. The training goal is to minimize the sum of the prediction errors of all auditory representation data. The calculation formula of the prediction error is: ,in, is the prediction error, is the group number of the feature vector corresponding to the auditory representation data, For the The predicted labels corresponding to the group auditory representation data, For the The actual predicted label corresponding to the group auditory representation data is trained, and the training is stopped when the sum of the prediction errors reaches convergence.

[0034] The auditory analysis model is a random forest, support vector machine or deep neural network model, which includes an input layer, a hidden layer and an output layer; each hidden layer includes multiple neurons, each neuron is connected to the neurons in the next layer, and the connection contains weights, which determine the importance and influence of data transmission in the neural network; an activation function is applied to each neuron between the hidden layer and the output layer, and the activation function is mapped into nonlinearity, allowing the network to learn more complex patterns and features.

[0035] This step can improve the reliability of the screening results of extremely susceptible and resistant individuals by screening individuals through multi-dimensional auditory characterization data. The analysis of multi-level data reduces the bias caused by misjudgment of a single factor, making the screening results more accurate, thereby optimizing the reliability of noise-induced hearing loss risk assessment. Please refer to Fig. 9 Forest plot of PRS and risk of noise-induced hearing loss shown.

[0036] Step 3: Obtain Q extreme individuals output by the auditory analysis model, extract the qth extreme individual for whole exome sequencing, and obtain a set of gene variation sites; q=1,2,…,Q; In a specific embodiment, the method of extracting the qth extreme individual for whole exome sequencing to obtain a set of gene variation sites includes: Step a1: extracting and purifying DNA from the blood sample of the qth extreme individual using a specific reagent to obtain purified DNA; It should be noted that: the blood sample is obtained by pre-treating the saliva of extreme individuals collected by the sequencing institution, and the specific reagent is a lysis buffer, a chelating agent or a precipitant; the method of purifying DNA includes organic solvent extraction method, silica gel membrane column method, magnetic bead method or salting out method, and the pretreatment method is to detect the quality of saliva to ensure that the DNA is sufficient and free of contamination. Any method for purifying DNA in the existing technology can be used and will not be explained in detail here.

[0037] Step a2: mechanically or enzymatically decomposing the purified DNA to obtain DNA fragments, wherein the size of the DNA fragments is in the range of 100-500 bp; Step a3: using a specific exon probe pool to capture DNA fragments in the exon region, obtaining exon sequence fragments, and removing non-exon sequences; The purpose is to: remove the non-exon sequences that were not captured, ensure that the sequencing data is mainly concentrated in the exon region, and improve efficiency and accuracy; Step a4: connecting the adapter containing the sequencing adapter to both ends of the exon sequence fragment, and performing PCR amplification on the exon sequence fragment to obtain the exon sequence amplified fragment; Step a5: sequencing the exon sequence amplified fragments by a sequencer connected to the sequencing platform to obtain sequencing data; It should be noted that: the sequencing platform is Illumina or BGI, the sequencer uses optical or electrochemical signals to read the exon sequence amplification fragment and generate sequencing data, output sequencing data, and the sequencing data is stored in FASTQ format, which contains the sequence and its corresponding quality information. The quality information is encoded in ASCII characters. Each character corresponds to a Q value, which is obtained by subtracting a specific offset (usually 33 or 64) from the ASCII value. Each character position corresponds to the quality value of the base at that position in the sequence.

[0038] Step a6: Perform quality control on the sequencing data to obtain high-quality sequencing data, and compare the obtained high-quality sequencing data with the reference genome to determine the set of gene variation sites. The set of gene variation sites includes SNPs, insertions, and deletions in the exon region, and the reference gene is pre-stored in the system database.

[0039] In one embodiment, the method of performing quality control on sequencing data to obtain high-quality sequencing data includes: Step b1: Use the first tool to evaluate the sequencing data, remove the sequencing data with bases below the standard quality, and obtain the first sequencing data.

[0040] It should be noted that: the first tool is the FastQC tool, which is used to analyze the basic quality distribution, GC content, read length consistency and any possible contamination and sequencing errors of the data. The standard quality base is Q30, which is specifically set by a technician in this field and is not specifically limited here.

[0041] Step b2: Use a second tool to process the first sequencing data, remove low-quality bases at the ends of the read segments, and cut out adapter fragments to obtain second sequencing data.

[0042] It should be noted that: the second tool is the Trimmomatic or Cutadapt tool, and the low-quality bases at the ends of the fragments are bases below the preset quality threshold; the preset quality threshold is set by technicians in this field based on experience and is not limited here; the adapter fragment refers to a short sequence added to the end of the DNA fragment before sequencing. If it is not removed, it will lead to data errors and analysis deviations. Therefore, adapter removal is performed when processing sequencing data.

[0043] Step b3: Set every F bases of the sequencer as a sliding window, perform a sliding check on the second sequencing data, obtain the average quality score of the bases in each sliding window, remove the window segments below the average quality score threshold, and obtain the third sequencing data; F is an integer greater than 1; to ensure that each segment of the third sequencing data has high-quality bases.

[0044] The method for calculating the average quality score of the bases in each sliding window includes: ; In the formula, represents the average quality score of the bases in the sliding window, is the Phred quality score of the f-th base in the sliding window, represents the total number of bases in the sliding window, and P represents the probability of a base being incorrectly determined.

[0045] It should be noted that the Phred quality score represents an indicator of the accuracy of base calling in sequencing data, and is often used in the results of high-throughput sequencing. The higher the value of the Phred quality score, the higher the accuracy of the base determination.

[0046] For example: =10: The error probability is 0.1, that is, there is a 10% probability of sequencing error.

[0047] =20: The error probability is 0.01, that is, there is a 1% probability of sequencing error.

[0048] =30: The error probability is 0.001, that is, there is a 0.1% probability of sequencing error.

[0049] =40: The error probability is 0.0001, that is, there is a 0.01% probability of sequencing error.

[0050] Therefore, every 10-point increase in Phred quality score means a 10-fold decrease in the probability of error.

[0051] Step b4: Use a third tool to mark the repetitive sequences generated by PCR amplification in the third sequencing data, and use a decontamination tool to clean up non-human sequences to obtain high-quality sequencing data.

[0052] It should be noted that the third tool is the Picard tool, which marks duplicate reads by aligning the sequence and position information of each read segment to facilitate removal during subsequent analysis. By marking duplicate sequences, the deviation introduced during the amplification process can be effectively reduced, and the accuracy of the sequencing data can be improved. The decontamination tool is the DecontaMiner tool. The DecontaMiner tool is used to align the sequencing data with the standard reference database (such as RefSeq) to detect and remove non-human sequences, thereby ensuring the high quality and pollution-free of the sequencing data. Non-human sequences refer to DNA sequences detected in the sequencing data that do not match the human genome. These sequences are pre-stored in the standard reference database.

[0053] In one specific embodiment, the method of aligning the obtained high-quality sequencing data with a reference genome to determine a set of gene variation sites includes: Step c1: Use the BWA-MEM algorithm to compare the high-quality sequencing data with the target genome to obtain a SAM file.

[0054] It should be noted that the BWA-MEM algorithm can efficiently process sequencing data with short read lengths and some medium read lengths, ensuring high accuracy of alignment.

[0055] Step c2: Use the fifth tool to convert the SAM file into a compressed BAM format, sort by genomic position to obtain a BAM file, and use the sixth tool to mark repeated sequences in the PCR process, correct alignment offsets, and filter out mismatched or low-quality alignment fragments; to ensure the accuracy of the data.

[0056] The fifth tool is a Samtools tool, and the sixth tool is a Picard tool.

[0057] Step c3: Use the seventh tool to perform genome recalibration on the BAM file, extract genome variation information, and generate a preliminary VCF file; the seventh tool is the GATK tool.

[0058] Step c4: Use the VQSR module of the seventh tool to calibrate the detected variant sites and set the variant quality threshold to screen the matching gene variant site set.

[0059] It should be noted that the variation quality threshold is set by those skilled in the art based on experience and is not specifically limited here.

[0060] This step, through step-by-step filtering, calibration and optimization, makes the final set of variant sites highly reliable and reduces the interference of false positives and noise. These high-confidence gene variant site sets are more biologically interpretable and help identify potential functional gene variants and disease-related mutations.

[0061] Step 4: Functionally annotate the set of gene variant sites, identify the set of suspected high-risk sites and the corresponding annotation results; See also Figure 2 As shown, in a preferred embodiment, the method of functionally annotating a set of gene variation sites, identifying a set of suspected high-risk sites and corresponding annotation results includes: Step d1: Preliminarily screening the gene variation site set based on the known SNP site data to obtain a screened gene variation site set; the screened gene variation site set includes group E gene variation sites; It should be noted that the known SNP site data include SNP sites known to be related to auditory function, cochlear function, nerve conduction, etc. The preliminary screening methods include database support, functional enrichment, literature basis and batch tool screening to effectively identify gene variation sites related to hearing.

[0062] See also Figure 6 , Figure 7 As shown, step d2: performing association analysis on the gene variation sites in the e-th group, and calculating the association coefficient of the gene variation sites in the e-th group, e=1, 2, ..., E; The method for obtaining the association coefficient of the e-th group of gene variation sites includes: ; In the formula, represents the association coefficient of the gene variation site in group e, represents the p-value, represents the effect value, represents the sample size, , and is the preset weight, Represents the logarithmic function with base e.

[0063] The preset weights in the formula are obtained by a technician in this field to collect multiple sets of comprehensive parameters, and set corresponding weights for each set of comprehensive parameters. The preset weights and the collected comprehensive parameters are substituted into the formula. Any four formulas constitute a three-variable linear equation system. The calculated weights are screened and averaged to obtain , and The value of .

[0064] It should be noted that: the lower the p-value, the more significant the association; the larger the effect value, the stronger the impact of the variant site on the risk of hearing loss; the larger the sample size, the higher the statistical credibility of the association result; the p-value, effect value and sample size are all obtained using WES analysis software.

[0065] Step d3: Preset a correlation coefficient threshold, the correlation coefficient threshold includes and ,in ; Compare the association coefficient of the gene variation site in group e with the preset association coefficient threshold; like , then the gene variation sites in group e are marked as suspected high-risk sites, indicating that the SNPs are marked as significantly associated with the risk of hearing loss multiple times in WES or annotation tools; like , then the gene variation sites in group e are marked as medium-risk sites, indicating that the gene variation sites in group e have low correlation in the WES database and low statistical significance or effect; like , the gene variation sites in group e are marked as low-risk sites, indicating that there is no significant association or they are not enriched in the hearing-related pathway.

[0066] Step d4: annotate the suspected high-risk sites using a functional annotation tool to obtain annotation results, set e=e+1, and return to step d2, wherein the functional annotation tool is Annovar or VEP; It should be noted that: the annotation results include gene location and functional region: the gene location includes coding region, promoter or non-coding region.

[0067] Step d5: Repeat steps d2 to d4 until the cycle ends when e=E, and obtain a set of suspected high-risk sites and corresponding annotation results.

[0068] It should be noted that the method of identifying the suspected high-risk site set and the corresponding annotation results can also be through literature support, WES research or functional enrichment analysis. Specifically, the literature support is to review the published genetic studies on noise-induced hearing loss and identify SNP sites that have been reported to be associated with hearing loss. For example, certain specific sites of genes related to cochlear structure or neural signal transduction have been mentioned many times in the literature and are significantly associated with the risk of hearing loss; the exon part of the genome is captured and enriched by specific technical means, and then high-throughput sequencing is performed to obtain the base sequence information of the exon region. Compared with whole genome sequencing, its coverage is relatively more targeted, focusing on the exon region that accounts for about 1% to 2% of the human genome but has important functions, and can more efficiently detect disease-related or biologically significant variants; the functional enrichment analysis is to perform functional enrichment analysis on the annotated SNP sites to determine whether they are concentrated in known auditory system function-related pathways (such as cochlear development, auditory conduction pathways), thereby inferring their potential association with hearing loss.

[0069] Depend on Figure 6 As shown in the Manhattan plot of the association analysis for risk assessment of noise-induced hearing loss: The red line represents the genome-wide significance level threshold. When the -log10(p) value of a certain site is higher than this line, it means that the association between the site and noise-induced hearing loss is highly statistically significant, that is, the site is likely to be a genetic site that is truly associated with noise-induced hearing loss.

[0070] The blue line represents the suggestive association threshold. If the -log10(p) value of a site exceeds the blue line but is lower than the red line, it indicates that there may be an association between the site and noise-induced hearing loss. However, its significance level has not yet reached the genome-wide significance level and requires further research and verification. These thresholds help technicians in this field quickly identify which sites on chromosomes have a meaningful association with the trait being studied (here, noise-induced hearing loss).

[0071] Depend on Figure 7 As shown in the Q-Qplot diagram of the whole exon association analysis results, it can be seen that: The vertical axis of observation represents the actual observed p-value of the association analysis after -log10 transformation, and the horizontal axis of expectation represents the expected -log10(p) value under the null hypothesis (that is, there is no real genetic association and the observed association is caused by random factors). The QQ plot is used to test whether the results of the association analysis meet the null hypothesis. If the data points are closely clustered in Figure 7 The red reference line in the figure (representing the expected distribution under the null hypothesis) indicates that the observed results are consistent with the expected results under the null hypothesis, that is, no significant genetic loci associated with the studied trait were found.

[0072] Step 5: Based on the PRS model, evaluate the set of suspected high-risk sites to obtain the z-score value of the qth extreme individual; determine the injury risk level based on the z-score value.

[0073] It should be noted that the PRS model is a risk score constructed based on multiple genetic loci, which indicates an individual's genetic susceptibility to a disease or specific phenotype.

[0074] The loci of the PRS model include rs10783780 (located at position 56704152 of human chromosome 12, base A / G), rs10794566 (located at position 125521533 of human chromosome 10, base G / A), rs10794567 (located at position 125521590 of human chromosome 10, base G / A), rs3887954 (located at position 53186088 of human chromosome 12 , base is G / C), rs58011943 (located at position 18684527 on human chromosome 17, base is G / C), rs11836060 (located at position 80750659 on human chromosome 12, base is G / A), rs16973424 (located at position 66904001 on human chromosome 17, base is A / C), rs74905175 (located at position 31815326 on human chromosome 7, base is C / T).

[0075] It is worth noting that the construction of PRS is based on 7 SNP sites with p<0.001 and consistent effect direction in the validation stage, and PRS is obtained by weighting each site. The validation method is as follows: The method consists of two phases: the discovery phase includes 150 extremely susceptible and 150 extremely resistant individuals, and the validation phase includes 2028 non-noise-induced hearing-impaired workers and 3836 noise-induced hearing-impaired workers. In addition, this application collects basic information including gender and age of the research subjects, please refer to Figure 4 and Figure 5 shown. Figure 5 In , n is the number of verification groups.

[0076] Among them, continuous variables are expressed as median (quartile), categorical variables are expressed as number of people (percentage), and CNE is the cumulative noise exposure.

[0077] Whole exome sequencing was performed on the Illumina HiSeq X10 platform. In the constructed DNA small fragment library, each insert fragment was sequenced at both ends, with 150 bp at each end. The average sequencing coverage depth was set to 50X.

[0078] After sequencing, the original sequence is analyzed. First, the data quality needs to be evaluated for quality control to determine whether it meets the standards. If it meets the standards, the samples are tested for mutations, including SNPs and InDels, and annotated. If it does not meet the standards, additional tests or a new library must be built based on the actual situation.

[0079] See also Figure 3 As shown, specifically, based on the PRS model, the suspected high-risk site set is evaluated, and the method for obtaining the z-score value of the qth extreme individual includes: Step s1: extracting the effect value of the hth suspected site in the suspected high-risk site set based on the association coefficient; h=1, 2, ..., H; It should be noted that the set of suspected high-risk sites includes H suspected sites, and the method for obtaining the association coefficient has been described above and will not be repeated here. It should be noted that the size of the effect value of each suspected site is related to the risk increase multiple of hearing loss associated with the site. For example, if the effect value of the suspected site is 1.5, it means that the risk increases by 1.5 times for each increase in the allele of the suspected site.

[0080] Step s2: Calculate the PRS score of the qth extreme individual, and the calculation formula is: ; In the formula, represents the PRS score of the qth extreme individual, is the effect value of the hth suspected site, Represents the genotype of the hth suspected site.

[0081] What needs to be said is: the genotype is coded as 0, 1, or 2 (usually 0 is no risk allele, 1 is one risk allele, and 2 is two risk alleles).

[0082] Step s3: Repeat step s2 until q=Q, and the cycle ends, obtaining the PRS scores of Q extreme individuals; Step s4: convert the PRS score of the qth extreme individual into a z-score value to obtain the z-score value of the qth extreme individual; ; ; In the formula, represents the z-score value of the qth extreme individual, represents the average PRS score of Q extreme individuals, Represents the standard deviation of the PRS score.

[0083] It should be noted that compared with the complex scores of the original PRS, the z-score value is more intuitive. It is difficult for non-technical personnel to understand the meaning of the original score, but the z-score shows the risk by the degree of deviation from the mean. A positive number represents a higher risk than the average, and a negative number represents the opposite. It is convenient for individuals to quickly know the relative risk status of extreme individuals.

[0084] The injury risk levels include level one, level two, level three, level four and level five; level one > level two > level three > level four > level five.

[0085] In implementation, based on the correspondence between the z-score value and the injury risk level, the method for determining the injury risk level of the Qth extreme individual includes: As shown in Table 1, an injury risk level table is set, and the corresponding injury risk level is obtained according to the z-score value; the z-score value is represented by ZS.

[0086] Table 1 Injury risk level table

[0087] It should be noted that K is a preset threshold value. K is obtained by collecting corresponding z-score values ​​during the historical hearing loss risk assessment process by technical personnel in this field. The average of the z-score values ​​collected multiple times is calculated and the average is used as the preset threshold value.

[0088] The z-score value of the qth extreme individual is matched to the injury risk level table to obtain the corresponding injury risk level.

[0089] Step 6: Feedback the corresponding protection measures plan to the client for display based on the damage risk level.

[0090] In implementation, the corresponding protection measures include: Option 1: The first level of recommended measures include: Completely avoid high noise exposure (such as avoiding long stays in places such as concerts and high-noise factories).

[0091] Customized protective equipment, such as special noise-cancelling earplugs or custom eardrum protectors.

[0092] Get a hearing health check-up every six months to monitor for any signs of hearing loss or early noise damage.

[0093] If there is a risk of injury, it is recommended to consult an ENT or audiology specialist in advance.

[0094] Option 2: The second level of recommended measures include: Strictly follow noise protection regulations in the workplace or in life.

[0095] Wear high-quality personal hearing protection and make sure it fits properly and is replaced regularly.

[0096] Avoid exposure to high-intensity noise environments (such as above 100dB).

[0097] Strengthen health monitoring and establish personalized hearing health records.

[0098] Option 3: The third level of recommended measures include: Always use protective equipment (such as earmuffs or high-efficiency noise-reducing earplugs) when working or engaging in activities in a noisy environment.

[0099] Limit noise exposure time (e.g. no more than 2 hours at a time).

[0100] Schedule regular hearing assessments (e.g., annually).

[0101] Option 4: The fourth level of recommended measures include: Avoid prolonged exposure to moderate noise (such as general machine noise in an industrial environment).

[0102] Use noise-canceling earplugs or earmuffs as appropriate.

[0103] Increase education on hearing care knowledge.

[0104] Option 5: The recommended measures for the fifth level include: No special protection is required, but healthy hearing protection habits should be developed (such as avoiding long-term exposure to noisy environments).

[0105] Have regular hearing checks (e.g. every two years).

[0106] This embodiment uses genetic data, environmental exposure data, individual physiological data and behavioral data to construct auditory representation data, realize multi-dimensional comprehensive analysis of hearing loss, reveal the multi-factor causes and interactions of hearing loss, and thus achieve accurate analysis; use the genetic data of extreme individuals to analyze gene variation sites known to be related to hearing function, standardize them according to the PRS score (such as calculating the Z-score value), and correspond them to the risk level of hearing loss, divide them into multiple assessment levels, and formulate more stringent hearing protection measures for high-risk individuals.

[0107] This embodiment is based on a comprehensive approach of multi-dimensional data, precise assessment models and big data analysis. It can not only analyze the causes of current hearing damage in detail, but also predict future risks and provide individuals with tailored protection recommendations, thereby improving the efficiency and effectiveness of hearing health management at both the prevention and treatment levels.

[0108] Example 2 See also Fig.10As shown, this embodiment provides a noise-induced hearing loss risk assessment system, a data acquisition module, an analysis module, a sequencing module, an identification module, an assessment module and a display module; each module is connected by wired and / or wireless means to achieve data transmission between modules; the system includes: A data acquisition module, used to obtain auditory representation data of I sample individuals, where I is an integer greater than zero; the auditory representation data includes individual genetic data, environmental exposure data, individual physiological data and behavioral data; An analysis module, configured to input the auditory representation data of I sample individuals into a pre-built auditory analysis model to obtain a first screening result, wherein the first screening result includes Q extreme individuals; The sequencing module is used to obtain Q extreme individuals output by the auditory analysis model, extract the qth extreme individual for whole exome sequencing, and obtain a set of gene variation sites; q = 1, 2, ..., Q; The identification module is used to perform functional annotation on the gene variation site set, identify the suspected high-risk site set and the corresponding annotation results; The evaluation module evaluates the suspected high-risk site set based on the PRS model to obtain the z-score value of the qth extreme individual; and determines the injury risk level according to the z-score value; The display module is used to feed back the corresponding protection measures plan to the client for display according to the damage risk level.

[0109] The formulas involved in the above are all calculated by removing dimensions and taking their numerical values. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The weight factors in the formula and the preset thresholds in the analysis process are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data; the size of the weight factor is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the weight factor depends on the amount of sample data and the corresponding processing coefficient initially set by technicians in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0110] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0111] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for assessing the risk of noise-induced hearing loss, characterized in that: include: Step 1: Obtain auditory representation data of I sample individuals, where I is an integer greater than zero; the auditory representation data includes individual genetic data, environmental exposure data, individual physiological data, and behavioral data; Step 2: Input the auditory representation data of I sample individuals into a pre-built auditory analysis model to obtain a first screening result, wherein the first screening result includes Q extreme individuals; Step 3: Obtain Q extreme individuals output by the auditory analysis model, extract the qth extreme individual for whole exome sequencing, and obtain a set of gene variation sites; q = 1, 2, ..., Q; Step 4: Functionally annotate the set of gene variant sites, identify the set of suspected high-risk sites and the corresponding annotation results; Step 5: Based on the PRS model, evaluate the set of suspected high-risk sites and obtain the z-score value of the qth extreme individual; Determine the injury risk level based on the z-score value; Step 6: Feedback the corresponding protection measures plan to the client for display based on the damage risk level.

2. The noise-induced hearing loss risk assessment method according to claim 1, characterized in that: The specific training process of the auditory analysis model includes: Pre-collect Y groups of auditory representation data, where Y is an integer greater than 1, set corresponding screening results for the auditory representation data, mark the digital labels of the screening results as prediction labels, and convert the auditory representation data and the corresponding prediction labels into a corresponding set of feature vectors; Each group of feature vectors is used as the input of the auditory analysis model. The auditory analysis model takes a group of prediction labels corresponding to each group of auditory representation data as output, and takes the actual prediction labels corresponding to each group of auditory representation data as prediction targets, where the actual prediction labels are pre-set prediction labels corresponding to the auditory representation data; minimizing the sum of prediction errors of all auditory representation data is used as the training target; the auditory analysis model is trained until the sum of prediction errors converges and the training is stopped; the auditory analysis model is a random forest, support vector machine or deep neural network model, and a first screening result is obtained.

3. The noise-induced hearing loss risk assessment method according to claim 2, characterized in that: The method of extracting the qth extreme individual for whole exome sequencing and obtaining a set of gene variation sites includes: Step a1: extracting and purifying DNA from the blood sample of the qth extreme individual using a specific reagent to obtain purified DNA; The specific reagent is a lysis buffer, a chelating agent or a precipitating agent; Step a2: mechanically or enzymatically decomposing the purified DNA to obtain DNA fragments, wherein the size of the DNA fragments is in the range of 100-500 bp; Step a3: using a specific exon probe pool to capture DNA fragments in the exon region, obtaining exon sequence fragments, and removing non-exon sequences; Step a4: connecting the adapter containing the sequencing adapter to both ends of the exon sequence fragment, and performing PCR amplification on the exon sequence fragment to obtain the exon sequence amplified fragment; Step a5: sequencing the exon sequence amplified fragments by a sequencer connected to the sequencing platform to obtain sequencing data; Step a6: Perform quality control on the sequencing data to obtain high-quality sequencing data, and compare the obtained high-quality sequencing data with the reference genome to determine the set of gene variation sites.

4. The noise-induced hearing loss risk assessment method according to claim 3, characterized in that: Methods for quality control of sequencing data and obtaining high-quality sequencing data include: Step b1: using a first tool to evaluate sequencing data, removing sequencing data with bases below the standard quality, and obtaining first sequencing data; Step b2: using a second tool to process the first sequencing data, remove low-quality bases at the ends of the read segments, and cut out adapter fragments to obtain second sequencing data; Step b3: setting the sequencer to have every F bases as a sliding window, performing a sliding check on the second sequencing data, obtaining an average quality score of the bases in each sliding window, removing window segments below the average quality score threshold, and obtaining third sequencing data; F is an integer greater than 1; Step b4: Use a third tool to mark the repetitive sequences generated by PCR amplification in the third sequencing data, and use a decontamination tool to clean up non-human sequences to obtain high-quality sequencing data.

5. The noise-induced hearing loss risk assessment method according to claim 4, characterized in that: Methods for aligning the obtained high-quality sequencing data with the reference genome to determine the set of gene variation sites include: Step c1: Use the BWA-MEM algorithm to compare high-quality sequencing data with the target genome to obtain a SAM file; Step c2: using the fifth tool to convert the SAM file into a compressed BAM format, sorting by genomic position to obtain a BAM file, and using the sixth tool to mark repeated sequences in the PCR process, correct alignment offsets, and filter out unmatched alignment fragments; Step c3: Use the seventh tool to recalibrate the BAM file, extract the genome variation information, and generate a preliminary VCF file; Step c4: Use the VQSR module of the seventh tool to calibrate the detected variant sites and set the variant quality threshold to screen the matching gene variant site set.

6. The method for assessing risk of noise-induced hearing loss according to claim 5, characterized in that: The method of functionally annotating a set of gene variant sites, identifying a set of suspected high-risk sites and corresponding annotation results includes: Step d1: preliminarily screening the gene variation site set based on the known SNP site data to obtain a screened gene variation site set; the screened gene variation site set includes group E gene variation sites; Step d2: performing association analysis on the e-th group of gene variation sites, and calculating the association coefficient of the e-th group of gene variation sites, e=1, 2, ..., E; Step d3: Preset a correlation coefficient threshold, the correlation coefficient threshold includes and ,in ; The correlation coefficient of the gene variation site in group e Compare with the preset correlation coefficient threshold; like , then the gene variation sites in group e are marked as suspected high-risk sites; like , then the gene variation sites in group e are marked as medium-risk sites; like , then the gene variation sites in group e are marked as low-risk sites; Step d4: annotate the suspected high-risk sites using a functional annotation tool to obtain annotation results, set e=e+1, and return to step d2; the annotation results include gene positions and functional regions: the gene positions include coding regions, promoters or non-coding regions; Step d5: Repeat steps d2 to d4 until the cycle ends when e=E, and obtain a set of suspected high-risk sites and corresponding annotation results.

7. The method for assessing risk of noise-induced hearing loss according to claim 6, characterized in that: The sites of the PRS model include rs10783780, rs10794566, rs10794567, rs3887954, rs58011943, rs11836060, rs16973424 and rs74905175.

8. The method for assessing risk of noise-induced hearing loss according to claim 7, characterized in that: Based on the PRS model, the method of evaluating the suspected high-risk site set and obtaining the z-score value of the qth extreme individual includes: Step s1: extracting the effect value of the hth suspected site in the suspected high-risk site set based on the association coefficient; h=1, 2, ..., H; H is the number of suspected sites in the suspected high-risk site set; Step s2: Calculate the PRS score of the qth extreme individual; Step s3: Repeat step s2 until q=Q, and the cycle ends, obtaining the PRS scores of Q extreme individuals; Step s4: Convert the PRS score of the qth extreme individual into a z-score value to obtain the z-score value of the qth extreme individual.

9. The method for assessing risk of noise-induced hearing loss according to claim 8, characterized in that: The injury risk levels include level one, level two, level three, level four and level five; level one > level two > level three > level four > level five.

10. The method for assessing risk of noise-induced hearing loss according to claim 9, characterized in that: The individual gene data include KCNQ4, MYO6, MYO7A, MYO15A, OTOF, GJB2, SLC26A4, GRM7, TECTA, CDH23, ESRRB, HSP70, CAT and SOD2; the environmental exposure data include noise exposure intensity and noise exposure time; The individual physiological data includes age, gender, genetic background, health status and history of hearing loss; the behavioral data includes high-frequency sound sensitivity, tinnitus symptoms and tolerance to noise.

11. A noise-induced hearing loss risk assessment system, used to implement the noise-induced hearing loss risk assessment method according to any one of claims 1 to 10, characterized in that: include: A data acquisition module, used to obtain auditory representation data of I sample individuals, where I is an integer greater than zero; the auditory representation data includes individual genetic data, environmental exposure data, individual physiological data and behavioral data; An analysis module, configured to input the auditory representation data of I sample individuals into a pre-built auditory analysis model to obtain a first screening result, wherein the first screening result includes Q extreme individuals; The sequencing module is used to obtain Q extreme individuals output by the auditory analysis model, extract the qth extreme individual for whole exome sequencing, and obtain a set of gene variation sites; q = 1, 2, ..., Q; The identification module is used to perform functional annotation on the gene variation site set, identify the suspected high-risk site set and the corresponding annotation results; The evaluation module evaluates the set of suspected high-risk sites based on the PRS model and obtains the z-score value of the qth extreme individual; Determine the injury risk level based on the z-score value; The display module is used to feed back the corresponding protection measures plan to the client for display according to the damage risk level.

Images