Establishment method of pathogenic gene database, database and risk prediction system

Abstract

The invention relates to a method for establishing a pathogenic gene database, a database and a risk prediction system, and the method comprises the steps: investigating the mutation conditions of mutation sites of people in other regions in specific people (i.e., people with specific residential regions, income, ages and health states) in a local region; mutation frequency and literature data are further combined for site filtering, and a set of pathogenic gene database suitable for pathogenic risk assessment of specific crowds in the region is developed. Based on the database, prediction of irrelevant diseases of the specific crowd can be greatly reduced by further optimizing a prediction algorithm, and the prediction efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of disease gene data processing, and specifically relates to a method for establishing a pathogenic gene database, a database and a risk prediction system, especially a method for establishing a pathogenic gene database for sub-healthy middle-aged and elderly high-income people living in Guangzhou, a database and a risk prediction system. Background Art

[0002] Human health is affected by both genetics and the environment. GWAS (Genome-wide association study) studies infer the correlation between genetic mutations and diseases in a specific population through large-scale population comparison screening studies. The theoretical basis of GWAS is the law of linkage disequilibrium (LD), which assumes that there is a strong LD between the observed SNP (single nucleotide polymorphism) and the true pathogenic mutation (causal variant). Through gene mutation detection, we can understand whether we are susceptible to certain common diseases (susceptibility risk), such as cardiovascular and cerebrovascular diseases, autoimmune diseases or tumors. In this way, we can try to avoid contact with harmful substances related to specific diseases, and conduct regular diagnostic monitoring in specific directions, so as to diagnose and treat the disease early at the onset, and minimize the damage caused by the disease.

[0003] However, the GWAS catalog (https: / / www.ebi.ac.uk / gwas / ) is a public database jointly developed and produced by the National Human Genome Research Institute (NHGRI) and the European Center for Bioinformatics (EMBL-EBI) to collect all published GWAS and related results. The database collects data from genome-wide association studies published since 2008. As of October 21, 2018, a total of 90,428 entries related to 61,613 unique SNP sites and 3,007 diseases were extracted from 411 different journals and 3,153 articles. Related studies must meet a strict standard to be included in the database. The GWAS catalog database provides a reliable basis for us to detect gene mutations.

[0004] On March 21, 2019, Cell magazine published a series of review and commentary articles on human genome sequencing research. The articles unanimously agreed that the ethnic diversity contained in the genomic data in the global human biobank is far less than people expected, with 68% coming from individuals of European descent; and disease susceptibility sites, especially polygenic disease susceptibility sites, are closely related to geography. High pathogenic sites in one region may not be in another region. To this end, in 2020, the Shanghai Institute of Nutrition and Health, Chinese Academy of Sciences, released the PGG.Han, a mutation frequency database of all SNP sites in 110,000 Northeast Asian populations: http: / / www.pgghan.org, which fills the gap in the large-scale available reference genome datasets for Northeast Asian populations. Based on this database, disease susceptibility sites in Northeast Asian populations can be screened.

[0005] Furthermore, considering that different clinical phenotype populations will have different gene mutation spectra, Ruijin Hospital affiliated to Shanghai Jiaotong University School of Medicine has released a database of all SNP site mutation frequencies of 10,588 Northeast Asians with various metabolic diseases, ChinaMAP: http: / / www.mbiobank.com / . Based on this database, various metabolic disease susceptibility sites in Northeast Asians can be screened.

[0006] Similarly, living in different regions, different economic classes, and long-term sub-health conditions will also affect the gene mutation spectrum. Therefore, it is very necessary to examine the mutation status of mutation sites in special populations and develop genetic data for different populations. Summary of the invention

[0007] Based on this, the present invention provides a method for establishing a pathogenic gene database, which not only takes into account the regional influence, but also evaluates the impact of different economic classes and long-term sub-health status on the population, and establishes a risk prediction system that can accurately and reliably evaluate the health risks of specific populations.

[0008] In one aspect, the present invention provides a method for establishing a pathogenic gene database, comprising the following steps:

[0009] (1) Obtain blood samples from a specific population, extract blood genomic DNA, and perform whole genome sequencing;

[0010] (2) Detect mutation sites in samples;

[0011] (3) Using the GWAS catalog database as a reference, filtering the GWAS catalog database according to the mutation sites of the test samples obtained in step (2) to obtain the pathogenic gene site database; the filtering includes primary filtering and secondary filtering, the primary filtering refers to screening the mutation frequencies of the mutation sites in the GWAS catalog database in the above-mentioned test samples, and retaining sites with mutation frequencies <30% and >0; the secondary filtering refers to filtering out pathogenic sites published in journals with IF <3.

[0012] The present invention collects blood from a specific population, extracts DNA, and performs whole genome sequencing to examine the mutation sites of different populations collected in the GWAScatalog database in the specific population to be examined in this study, and further combines the mutation frequency and literature data to filter the sites, wherein sites with a mutation frequency of <30% are likely to be pathogenic sites, and high-frequency sites often mean mutations that the public can tolerate; in addition, sites with a mutation frequency of 0 mean rare mutation sites in the population of this study, which contribute little to the risk of disease and increase background noise. Therefore, sites with a mutation frequency of <30% and >0 are retained in one filtering; and then, according to the authority of the papers published on the mutation sites, journals with IF<3 are filtered out for secondary screening, thereby developing a set of pathogenic gene site databases suitable for pathogenic risk assessment of specific populations, and improving the ability to accurately predict the risk of disease.

[0013] In one embodiment, the specific population is people aged 45-65, who have settled in Guangzhou for ≥10 years, have a monthly income of ≥20,000 yuan, and are in sub-health.

[0014] In one embodiment, in the whole genome sequencing, genomic DNA is prepared into PE150 DNA library for sequencing, and more than 1 μg of genomic DNA with good integrity is prepared for each sample. The criterion for judging good genome integrity is: the main DNA band in the 1% agarose gel electrophoresis diagram is intact and >23kb. In addition, slightly degraded genomic DNA can also be used for risk library construction.

[0015] In one embodiment, the requirements for constructing a successful DNA library are: the main band of the PCR product fragment is 450bp±50bp, there is no dimer or contamination, and the quantitative concentration of the product after digestion and purification is greater than 0.78ng / μL.

[0016] In one embodiment, the depth of whole genome sequencing is 10X-30X.

[0017] In one embodiment, in step (2), the mutation site is detected based on the dbsnp146 database.

[0018] Another aspect of the present invention also provides a pathogenic gene database obtained by the above-mentioned method for establishing a pathogenic gene database.

[0019] In another aspect of the present invention, there is provided a risk prediction system, comprising:

[0020] A data acquisition module is used to obtain the genetic testing data of the sample to be tested;

[0021] A data analysis module is used to analyze the above gene detection data through bioinformatics, and then substitute it into the above pathogenic gene database for comparison and analysis to obtain the risk value R;

[0022] The judgment output module is used to output the site mutation information according to the risk level.

[0023] In one embodiment, the risk value R is expressed as Calculation is performed, where n is the number of sites in a person's disease detection, and N is the total number of sites in the established database for the disease; OR value, the full name of which is Odds Ratio, is also called odds ratio and risk exposure ratio, which can be obtained from the GWAS catalog database. By constructing a risk value matrix, the number of behavioral samples is listed as the disease type, and the disease risk value corresponding to each sample is sorted. The risk value between the top 5%-1% is defined as medium risk; R within the top 1% is defined as high risk, and the rest is low risk.

[0024] Taking the OR value of all N risk sites of a disease as the base, take the logarithm of the OR value of n risk sites detected for a person. By taking the logarithm, the risk factor can be reduced in dimension to avoid a single risk factor dominating the risk assessment due to excessive value. Furthermore, the purpose of taking the total risk factor as the base is to standardize so that the risk values ​​of different individuals and different diseases are comparable. Only the mutation ratio of known risk sites is calculated to evaluate the risk level of the disease. In this way, both single-gene diseases and polygenic diseases have better prediction effects. Because the pathogenic genes of the disease are originally more or less, some are single-gene diseases (for example, thalassemia) and some are polygenic diseases (for example, cancer).

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] The method for establishing a pathogenic gene database of the present invention examines the mutation sites of different populations collected in the GWAS catalog database and the mutation status of a specific population collected in the non-GWAS catalog database of this study, further combines the mutation frequency and literature data for site filtering, and population evaluation, to develop a pathogenic gene database suitable for pathogenic risk assessment of a specific population.

[0027] The pathogenic gene database of the present invention includes a large number of site records with high pathogenic risks in specific populations. By matching and analyzing gene detection sites with these site records, high-risk pathogenic sites can be quickly located, reducing the possibility of omissions and greatly improving the accuracy and efficiency of clinical interpretation.

[0028] The pathogenic gene database of the present invention can be used in the risk prediction system. By optimizing the risk assessment algorithm, it can be better applied to the prediction of various types of diseases, whether single-gene or multi-gene diseases. In addition, it also greatly reduces the prediction of non-risk diseases and avoids the deviation of risk level prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is an example of nucleic acid concentration detection in Example 1 of the present invention;

[0030] Figure 2 This is a mutation frequency change trend diagram of the gene database of the present invention. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention and the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0032] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0033] Unless otherwise specified, the reagents, materials, and equipment used in this example are all commercially available; the experimental methods are all conventional experimental methods in the art unless otherwise specified.

[0034] Example 1

[0035] A disease-causing gene database is established by the following method:

[0036] (1) Obtain blood samples from a specific population, extract blood genomic DNA, and perform whole genome sequencing.

[0037] 1. Sample collection

[0038] 198 blood samples were collected from Guangzhou Zhongxin Gene Medical Technology Co., Ltd. 2 ml of blood was collected from each sample. The age distribution of the samples ranged from 4 to 65 years old, and the gender was 100 males and 98 females.

[0039] 2. Blood genomic DNA extraction

[0040] Blood DNA extraction kit (Yisheng Biotechnology, Shanghai, Blood DNA Kit) was used to extract genomic DNA according to the instructions. The recovered DNA fragments were tested for concentration and purity by agarose gel electrophoresis and UV spectrophotometer. The DNA should have a significant absorption peak at OD260, and the OD260 / OD280 ratio should be 1.7-1.9. Figure 1 shown.

[0041] 3. Whole genome sequencing

[0042] The genomic DNA library preparation kit (purchased from MGI, catalog number: 940-200022-00) was used to prepare PE150 DNA library according to the instructions. More than 1 μg of genomic DNA with good integrity was prepared for each sample (samples with intact DNA main bands > 23 kb in the 1% agarose gel electrophoresis were considered complete genomic DNA samples). Slightly degraded genomic DNA can also be used for risk library construction.

[0043] The successfully constructed DNA needs to be quality controlled. After the PCR product is purified, agarose gel electrophoresis or Agilent2100Bioanalyzer can be used to detect the length distribution range of the PCR product. The main band of the PCR product fragment is required to be around 450bp, without dimer and other contamination. The digested and purified product can be selected by Qubit ssDNAAssay Kit or Quant-iT TM The product was quantified using a single-stranded DNA quantification kit such as OliGreen ssDNA Reagent. 1 μL of the digested and purified product was used for quantification. The product concentration should be greater than 0.78 ng / μL.

[0044] The constructed PE150 DNA library was sequenced on the BGI MGISEQ-T7 sequencer according to the instructions, with a sequencing depth of 10X-30X.

[0045] (2) Detect the mutation sites of the sample.

[0046] Specifically, fastp software (version 0.20.0, default parameters) was first used to filter the raw sequencing data. Then bwa software (version 0.7.15, default parameters) was used to align the filtered data to the human reference genome hg38. Samtools software (version 1.11, default parameters) was used to convert the file format. GATK software (version 4.1.2.0, database dbsnp146, default parameters) was then used to detect mutations in the samples.

[0047] (3) Site filtering.

[0048] Taking the GWAS catalog database as a reference, the GWAS catalog database is filtered according to the mutation sites of the test samples obtained in step (2) to obtain the pathogenic gene database; the filtering includes primary filtering and secondary filtering, the primary filtering refers to screening the mutation frequencies of the mutation sites in the GWAS catalog database in the test samples, and retaining sites with mutation frequencies within a certain range; the secondary filtering refers to filtering out the pathogenic sites published in journals with IF<3. Thus, the gene library of the specific population to be studied in this study is established.

[0049] 1-1 Evaluation of sample size

[0050] In order to ensure that the sample size can represent all specific populations to be studied, the sample size is determined by evaluating the database saturation to ensure the stability of the site mutation frequency. The details are as follows:

[0051] From the 198 subjects, we randomly selected 20, 40, 60, 80, 100, 120, 140, 160 and 180 individuals and all 198 individuals to calculate the mutation frequency of 4693 sites in the GWAS catalog database. We evaluated the proportion of sites with mutation frequencies greater than 20% and changes greater than 10% for every 20 additional samples, and drew a trend graph ( Figure 2 ).like Figure 2 As shown in the figure, as the number of samples increases, the percentage of sites with mutation frequencies greater than 20% and changes greater than 10% gradually becomes 0. This indicates that the site mutation frequency tends to be stable. The mutation frequency calculation based on 198 samples is reliable for sites with mutation frequencies greater than 20%. As for sites with mutation frequencies less than 20%, their mutation frequencies are no longer important. This is because once a mutation is detected, it will be included in our prediction model and cannot be filtered out.

[0052] pass Figure 2 It can be seen from the mutation frequency change trend that the 198 samples in this application basically guarantee the stability of the database.

[0053] 1-2. Step (3) Site filtering and screening condition selection

[0054] After determining that 198 samples could well meet the database stability requirements, three screening methods were further set up for comparison to determine the most recent screening parameters.

[0055] (A) In the first filtering, sites with mutation frequencies <20% and >0 were retained; in the second filtering, pathogenic sites published in journals with IF <3 were filtered out.

[0056] (B) In the first filtering, sites with mutation frequencies <30% and >0 were retained; in the second filtering, pathogenic sites published in journals with IF <3 were filtered out.

[0057] (C) In the first filtering, sites with mutation frequencies <40% and >0 were retained; in the second filtering, pathogenic sites published in journals with IF <3 were filtered out.

[0058] Among them, under the (A) scheme, a total of 1256 sites with mutation frequencies <20% and >0 were retained. Furthermore, pathogenic sites published in journals with IF <3 were filtered out, resulting in 1167 sites remaining, corresponding to 303 diseases (projects).

[0059] Under the (B) scheme, 2078 sites with mutation frequencies <30% and >0 were retained. Furthermore, pathogenic sites published in journals with IF <3 were filtered out, resulting in 1886 sites remaining, corresponding to 376 diseases (projects).

[0060] Under the (C) scheme, 3853 sites with mutation frequencies <40% and >0 were retained. Furthermore, pathogenic sites published in journals with IF <3 were filtered out, resulting in 3622 sites remaining, corresponding to 425 diseases (projects).

[0061] Specifically, for scheme (B), a total of 4693 mutation sites were collected from the GWAS catalog database as of May 21, 2021, and the mutation frequencies of these sites in 198 specific population samples were calculated. A total of 2078 sites with mutation frequencies <30% and >0 were screened and retained. We believe that sites with mutation frequencies <30% are likely to be pathogenic sites, and high-frequency sites often mean mutations that the public can tolerate; in addition, sites with mutation frequencies = 0 mean rare mutation sites for the population in this study, which contribute little to the risk of disease and increase background noise. For example, the mutation sites rs4773144 and rs2237895 in Table 1 have a specific population frequency in this study > 0.3, so they are removed; the mutation sites rs688034 and rs12734338 are equal to 0 and are removed. Furthermore, we conducted a secondary screening based on the authority of the papers published on the mutation sites, filtering out pathogenic sites published in journals with IF < 3. For example, the reported mutation sites rs12734338, rs3782889, and rs7560163 in Table 1 were disease-related sites with a journal IF < 3, and were removed, resulting in 1886 sites remaining, corresponding to 376 diseases (projects), such as the coronary heart disease-related sites rs1994016 and rs6922269, and the type 2 diabetes-related site rs17036101.

[0062] Combined with Table 1, it can be seen that after the above-mentioned site filtering, the obtained database better retains the sites with high-risk pathogenic risk for the population in this study in the original GWAScatalog database, and is more suitable for the population included in the non-GWAScatalog database of this study.

[0063] Table 1. Examples of disease-related mutation site screening

[0064]

[0065] Example 2

[0066] A risk prediction system, comprising:

[0067] A data acquisition module is used to obtain the whole genome data of the sample to be tested;

[0068] A data analysis module is used to substitute the above-mentioned whole genome data into the pathogenic gene database established in Example 1 for comparison and analysis to obtain a risk value R after bioinformatics analysis;

[0069] The determination output module is used to output the above site mutation information according to the risk level.

[0070] The traditional risk assessment algorithm is as follows:

[0071] R=log2(OR1*OR2*OR3*..*OR n )*e, the OR values ​​of n risk sites related to the disease are multiplied, the product is log2 logarithm, and then multiplied by the risk factor e, where OR is the odds ratio corresponding to different sites, e = n / N (the number of sites n detected for a certain disease in a person / the total number of sites for the disease N). The purpose of taking log2 is to reduce the dimension of the total risk factor to make it comparable. The risk factor eliminates the possibility of a certain disease occurring randomly due to the large number of related sites.

[0072] Screen each sample for items whose risk value is in the top 20% of the 198 samples. Then, define the risk value R between 0.5-1 as medium risk; R>1 as high risk. In this way, among the 198 samples, each sample has an average of 75 risk diseases. From the perspective of predicting the number of diseases, it is easy to increase the psychological burden on the subjects and make them feel at a loss.

[0073] Moreover, this algorithm is more inclined to polygenic diseases because the more loci there are, the larger the R value will be. This will result in some diseases being medium risk rather than high risk even if all loci are mutated.

[0074] Therefore, further optimization of the risk prediction algorithm is carried out.

[0075] According to the formula To calculate the risk value, we use logarithms to reduce the dimension of the risk factors to avoid a single risk factor dominating the risk assessment due to its large value. Furthermore, the purpose of using the total risk factor as the base is to standardize so that the risk values ​​of different individuals and different diseases are comparable.

[0076] Then, a risk value matrix is ​​constructed: 198 samples are listed as 376 diseases, with a total of 74,448 risk values. Then, these 74,448 risk values ​​are sorted. Then, the risk value R between the top 5%-1% is defined as medium risk; R in the top 1% is defined as high risk.

[0077] Thus, among the 198 samples, each sample has an average of 15 medium-risk diseases and 4 high-risk diseases. This greatly reduces the number of predicted risk diseases and is more targeted. In addition, regardless of the number of disease risk sites, the algorithm only calculates the mutation ratio of known risk sites and assesses the risk level of the disease based on this. This algorithm is applicable to both single-gene diseases and multi-gene diseases.

[0078] Example 3

[0079] The risk prediction system of Example 2 is used to analyze and predict the sample of a male patient in Case 1.

[0080] Case 1: Male, 62 years old, suffered from periodontitis, vasculitis, poor sleep, and prone to coughing; his brother died of leukemia and his bone marrow transplant was ineffective; his sister had symptoms of dementia when she was young.

[0081] The optimized algorithm can indicate all known risks (Table 2, No. 1-6), while the traditional algorithm cannot find the risk of insomnia. Insomnia is a medium risk in the traditional algorithm because its R = 0.627 < 1, but it is a high risk in the new algorithm because its R value is ranked 588th (0.79%) in the ranking.

[0082] In addition, as shown in Table 2, the optimized algorithm only found 8 high-risk diseases, while the algorithm before optimization found 17 high-risk diseases. It can be seen that the optimized algorithm greatly reduces the prediction of irrelevant diseases.

[0083] Table 2. Comparison of predictions of different algorithms for case 1

[0084]

[0085]

[0086] Example 4 Comparison of different thresholds

[0087] Using the method of step (2) in Example 1, the database under the three screening conditions (A), (B) and (C) is applied to Case 1, and the prediction system of Example 3 is used to predict the results. The specific prediction results are shown in Table 3. It can be seen that the database of the present invention uses 20% or 40% as a threshold to screen sites, and its health prediction cannot find the risk of periodontitis or dementia; while using 30% as a threshold can indicate all known risks (Table 3), indicating that the 30% threshold is the best and has an unexpected technical effect. Therefore, the gene database of the embodiment of the present invention adopts a screening condition of 30%.

[0088] Table 3. Comparison of predictions for case 1 at different screening thresholds

[0089] serial number High-risk diseases 20% Threshold 30% Threshold 40% Threshold 1 Childhood acute lymphoblastic leukemia √ √ √ 2 Immunoglobulin A-associated vasculitis √ √ √ 3 Lewy body dementia × √ × 4 Periodontitis √ √ × 5 Insomnia √ √ √ 6 Diisocyanate-induced asthma √ √ √ 7 Silicosis × √ √ 8 Alzheimer's disease × √ √ 9 Primary biliary cirrhosis √ × √ 10 Nontyphoidal Salmonella bacteremia × × √ 11 Refractory atopic dermatitis √ × × 12 Multiple sclerosis relapse × × √ 13 Esophageal cancer × × √ 14 Biliary atresia √ × ×

[0090] Example 5 Comparison of different populations

[0091] Since different regions, different economic classes and long-term sub-health status can also affect the gene mutation spectrum, further, by statistically analyzing factors such as age, income, and health status of the 198 subjects in the embodiment of the present application, the impact of these conditions on the database was examined.

[0092] According to statistics, all 198 subjects in Example 1 of the present application have settled in Guangzhou for a long time for ≥10 years. According to statistics, 195 are between 40 and 70 years old, and the other 3 are between 4 and 15 years old and are their descendants; and their income is ≥20,000 yuan, and they are in sub-health status. These 198 subjects are defined as high-income sub-healthy middle-aged and elderly people in Guangzhou.

[0093] In order to evaluate the specificity of site mutation frequency, we used the mutation frequency of 4693 sites in the GWAS catalog database in the gene database determined by the screening conditions determined in the above Example 3, and compared it with the mutation frequency in the PGG.Han database of 110,000 people. The results are shown in Table 4. Among the 1886 sites with a mutation frequency of less than 30% in the Guangzhou high-income sub-healthy middle-aged and elderly gene database in the embodiment of the present invention, 123 sites have a mutation frequency of greater than 30% and a variation of more than 10% in the PGG.Han database. On the contrary, among the 4693 sites in the GWAS catalog database, 178 sites have a mutation frequency of less than 30% in the PGG.Han database, while the mutation frequency in the gene database of the embodiment of the present invention is greater than 30% and the variation is more than 10%.

[0094] This suggests that the high-income, sub-healthy middle-aged and elderly people in Guangzhou may have their own special mutation spectrum, which may be affected by factors such as diet, living environment, mental stress and age, changing the acquired gene mutation spectrum in the blood, thereby causing special background noise for the congenital gene mutation spectrum we are studying.

[0095] Furthermore, for comparison purposes, 200 volunteers (age: 20-30 years old, settled in Guangzhou for ≥10 years, healthy, with a monthly income of less than 6,000 yuan) were recruited for whole genome sequencing with a sequencing depth of 30X, and a genetic database of low-income healthy young people in Guangzhou was established according to the method of the above embodiment.

[0096] The mutation frequencies of 4693 loci in the GWAS catalog database in the Guangzhou high-income sub-healthy middle-aged and elderly gene database were compared with those in the Guangzhou low-income healthy young people gene database. The results are shown in Table 4. Among the 1886 loci with a mutation frequency of less than 30% in the Guangzhou high-income sub-healthy middle-aged and elderly gene database, 157 loci had a mutation frequency of more than 30% and a variation of more than 10% in the Guangzhou low-income healthy young people gene database. On the contrary, among the 4693 loci in the GWAS catalog database, 334 loci had a mutation frequency of less than 30% in the Guangzhou low-income healthy young people gene database, while a mutation frequency of more than 30% and a variation of more than 10% in the Guangzhou high-income sub-healthy middle-aged and elderly gene database.

[0097] In contrast, the mutation frequencies of the Guangzhou Low-income Healthy Young People Gene Database and the PGG.Han Database are similar (Table 4). Among the 2435 sites with a mutation frequency of less than 30% in the Guangzhou Low-income Healthy Young People Gene Database, only 47 sites have a mutation frequency of more than 30% and a variation of more than 10% in the PGG.Han Database. Similarly, among the 4693 sites in the GWAS catalog database, only 59 sites have a mutation frequency of less than 30% in the PGG.Han Database, while they have a mutation frequency of more than 30% and a variation of more than 10% in the Guangzhou Low-income Healthy Young People Gene Database.

[0098] This shows that the database constructed in the embodiment of the present application is more suitable for reflecting the high-income sub-healthy middle-aged and elderly people in Guangzhou.

[0099] Table 4. Mutation frequency of disease-related mutation sites in our own database and PGG.Han database

[0100]

[0101] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for establishing a pathogenic gene database, characterized in that: The steps include: (1) Obtain blood samples from a specific population, extract blood genomic DNA, and perform whole genome sequencing; (2) Detect mutation sites in samples; (3) Using the GWAS catalog database as a reference, filtering the GWAS catalog database according to the mutation sites of the test samples obtained in step (2) to obtain the pathogenic gene database; the filtering includes primary filtering and secondary filtering, the primary filtering refers to screening the mutation frequencies of the mutation sites in the GWAS catalog database in the above-mentioned test samples, and retaining sites with mutation frequencies <30% and >0; the secondary filtering refers to filtering out pathogenic sites published in journals with IF <3.

2. The method for establishing a pathogenic gene database according to claim 1, characterized in that: The specific population is people aged 45-65, who have settled in Guangzhou for ≥10 years, have a monthly income of ≥20,000 yuan, and are in sub-health.

3. The method for establishing a pathogenic gene database according to claim 1, characterized in that: In the whole genome sequencing, genomic DNA was prepared into PE150 DNA library for sequencing, and more than 1 μg of genomic DNA with good integrity was prepared for each sample. The criterion for judging the good integrity of the genome was that the main DNA band in the 1% agarose gel electrophoresis diagram was intact and >23 kb.

4. The method for establishing a pathogenic gene database according to claim 3, characterized in that: The requirements for constructing a successful DNA library are: the main band of the PCR product fragment is 450bp±50bp, there is no dimer or contamination, and the quantitative concentration of the product after digestion and purification is greater than 0.78ng / μL.

5. The method for establishing a pathogenic gene database according to claim 4, characterized in that: The depth of the whole genome sequencing is 10X-30X.

6. The method for establishing a pathogenic gene database according to claim 1, characterized in that: In the step (2), the mutation site is detected based on the dbsnp146 database.

7. A pathogenic gene database obtained by the method for establishing a pathogenic gene database according to any one of claims 1 to 6.

8. A risk prediction system, characterized in that: include: A data acquisition module is used to obtain the genetic testing data of the sample to be tested; A data analysis module, for analyzing the above gene detection data by bioinformatics, and then substituting the data into the pathogenic gene database of claim 6 for comparison and analysis to obtain a risk value R; The judgment output module is used to output the site mutation information according to the risk level.

9. The risk prediction system according to claim 8, characterized in that: The risk value R is expressed as follows: Calculation is performed, wherein OR is the odds ratio corresponding to different sites; n is the number of sites detected for a certain disease in a person, and N is the total number of sites of the disease in the pathogenic gene database described in claim 6; By constructing a risk value matrix, the number of behavioral samples is listed as disease types, and the disease risk values ​​corresponding to each sample are sorted. Risk values ​​between the top 5%-1% are defined as medium risk; R within the top 1% is defined as high risk, and the rest are low risk.

Images