Screening method, system, device and readable storage medium for genetic disease related genes

Abstract

The application discloses a screening method, a screening system and a screening device based on a new mutation of a genetic disease related gene and a readable storage medium. The screening method comprises the following steps: obtaining a reference data set of a genetic disease; obtaining a sample data set; obtaining a p value of each gene; establishing a Poisson distribution model according to an expected mutation number of each gene in the reference data set; respectively calculating a probability that a mutation fraction of each gene under different mutation numbers is greater than an actually observed mutation fraction in the sample data set; and adding the probabilities under different mutation numbers to obtain the p value of the gene. The p value of each gene is compared with a significant level, and a gene with a p value less than the significant level is screened out, and the screened gene is a genetic disease related gene. According to the application, an efficient genetic disease related gene screening method, a screening system, a screening device and a readable storage medium can be provided.

Description

Technical Field

[0001] This application relates to the field of bioinformatics, and in particular to a method, system, apparatus and readable storage medium for screening genes related to genetic diseases. Background Technology

[0002] When interpreting genetic disease data using gene sequencing, the process typically begins with the phenotype. Medical interpreters, based on existing literature or database records, focus on the set of genes associated with the patient's phenotype to identify the pathogenic mutation. However, due to limitations in technology and scientific understanding, many genes related to genetic diseases remain unclear. Therefore, even if next-generation sequencing detects a pathogenic mutation, medical interpreters may overlook the impact of these genes on the patient's disease because the gene was not previously recorded, leading to an inability to determine the cause of the disease.

[0003] Currently, the gold standard for identifying disease-related genes remains biochemical molecular experimental methods, such as gene knockout and gene silencing. However, these methods are inefficient and costly, and cannot be used on a large scale to discover disease-related genes.

[0004] Therefore, there is an urgent need for a highly efficient method or device for screening genes related to genetic diseases. Summary of the Invention

[0005] The purpose of this application is to provide a method, system, device, and computer-readable storage medium for screening genes related to genetic diseases based on newly emerging mutations.

[0006] The following technical solution is adopted in this application:

[0007] The first aspect of this application discloses a method for screening genes related to genetic diseases based on novel mutations, comprising the following steps: obtaining a reference dataset of the genetic disease, the reference dataset including gene mutation information in a public database; obtaining a sample dataset, the sample dataset including information on novel mutations in the genetic disease pedigree; calculating the mutation score of each gene in the sample dataset, defining the number of mutations of each gene in the sample dataset as the observed mutation number, defining the number of mutations of each gene in the reference dataset as the expected mutation number, and the mutation score of a gene as the observed mutation number of that gene divided by the expected mutation number of that gene; obtaining the p-value of each gene, establishing a Poisson distribution model based on the expected mutation number of each gene in the reference dataset, calculating the probability that the mutation score of each gene under different mutation numbers is greater than the actual observed mutation score in the sample dataset, and summing the probabilities under different mutation numbers as the p-value of the gene; wherein, the p-value is obtained according to the following formula (I):

[0008]

[0009] In equation (III), P represents probability, S is the mutation score for each gene, score is the actual sample mutation score for each gene, k is the number of mutations, μ is the expected number of mutations, e is the natural constant, and S k The variable represents the mutation score of each gene under k mutations, and sum represents the summation; the p-value of each gene is compared with the significance level, and genes with p-values ​​less than the significance level are screened out. The screened genes are the genes related to the genetic disease.

[0010] In one implementation of this application, the novel mutation is a germline mutation, and the genetic disease is a single-gene genetic disease or a polygenic genetic disease.

[0011] In one implementation of this application, the gene mutation information in the reference dataset and the sample dataset has the same format.

[0012] In one implementation of this application, the gene mutation information in the reference dataset includes the gene name, chromosome number, genomic location, reference base, mutated base, mutation type, mutation probability, mutation harmfulness prediction result, whether it is located in a conserved gene, and whether it is located in a highly conserved gene.

[0013] In one implementation of this application, gene mutations are grouped according to the mutation type and / or the mutation harmfulness prediction result and / or whether they are located in a conserved gene and / or whether they are located in a highly conserved gene, the mutation score of each group is calculated, and then the mutation score of each gene is obtained.

[0014] In one implementation of this application, the mutation type includes synonymous mutation, in-frame mutation, splicing mutation, frameshift mutation, missense mutation, and nonsense mutation, and the mutation harmfulness prediction result is a CADD standard score.

[0015] In one implementation of this application, gene mutations belonging to missense and nonsense mutations are grouped according to the following criteria: 0 < CADD standard score ≤ 6, 6 < CADD standard score ≤ 12, 12 < CADD standard score ≤ 18, 18 < CADD standard score ≤ 24, 24 < CADD standard score ≤ 30, and CADD standard score > 30. Further grouping is based on whether the mutation is located in a conserved gene, and then on whether it is located in a highly conserved gene. Gene mutations belonging to synonymous synchronization, in-frame mutations, splicing mutations, and frameshift mutations are grouped according to whether they are located in a conserved gene. A mutation score is calculated for each group, where the mutation score for each group is the observed number of mutations divided by the expected number of mutations in that group.

[0016] The second aspect of this application discloses a screening system for genetic disease-related genes based on novel mutations. The screening system includes the following modules: a reference dataset acquisition module, which includes gene mutation information from a public database; a sample dataset organization module, which includes information on novel mutations in the genetic disease pedigree; and a mutation score calculation module, used to calculate the mutation score of each gene in the sample dataset. The number of mutations in each gene in the sample dataset is defined as the observed mutation count, the number of mutations in each gene in the reference dataset is defined as the expected mutation count, and the mutation score of a gene is the observed mutation count of that gene divided by the expected mutation count of that gene.

[0017] The p-value calculation module establishes a Poisson distribution model based on the expected number of mutations for each gene in the reference dataset, calculates the probability that the mutation score of each gene is greater than the actual observed mutation score in the sample dataset under different mutation numbers, and sums the probabilities under different mutation numbers as the p-value of the gene; wherein, the p-value is obtained according to the following formula (I):

[0018]

[0019] In equation (III), P represents the probability, S is the mutation score for each gene, score is the actual sample mutation score for each gene, k is the number of mutations, μ is the expected number of mutations, e is the natural constant, and S k The variable represents the mutation score of each gene under k mutations, and sum represents the summation; the gene screening module compares the p-value of each gene with the significance level, and screens out genes with p-values ​​less than the significance level. The screened genes are the genes related to the genetic disease.

[0020] A third aspect of this application discloses a computer-readable storage medium storing a program that can be executed by a processor to implement the screening method disclosed in the first aspect of this application.

[0021] The fourth aspect of this application discloses a screening device based on newly mutated genetic disease-related genes, the screening device comprising a memory and a processor, wherein: the memory is used to store a program; and the processor is used to implement the screening method as disclosed in the first aspect of this application by executing the program stored in the memory.

[0022] This application also discloses an application of a screening method for genes related to genetic diseases based on novel mutations, wherein the genes related to the genetic disease are obtained by using the screening method disclosed in the first aspect of this application.

[0023] The beneficial effects of this application are as follows:

[0024] This application utilizes computer data processing and statistical methods to enrich newly emerging mutations into genes and then uses statistical methods to determine a list of genes significantly enriched by these mutations. This allows for the identification of a list of genes most relevant to genetic diseases, which is more efficient and less costly than biochemical molecular experimental methods. Furthermore, it employs rigorous statistical methods and effective statistical thresholds, enabling the effective detection of novel genes associated with genetic diseases. This means that genes not previously clearly recorded in literature or databases related to the genetic disease can be identified, thereby expanding the list of genes related to the genetic disease that medical interpreters are interested in, reducing the possibility of overlooking pathogenic mutations, and providing more accurate guidance for genetic counseling. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating the screening method for genetically related genes based on novel mutations involved in this application.

[0026] Figure 2 This is a Poisson distribution diagram of the screening method for genetically related genes based on novel mutations involved in this application. Detailed Implementation

[0027] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. In the following embodiments, many details are described to facilitate a better understanding of the present application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other materials or methods. In some cases, certain operations related to the present application are not shown or described in the specification. This is to avoid obscuring the core parts of the present application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; the relevant operations can be fully understood based on the description in the specification and general technical knowledge in the art.

[0028] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0029] Currently, gene sequencing screening for developmental disorders often starts with phenotype. Medical interpreters, based on existing literature or database records, focus on the set of genes associated with the patient's phenotype to identify the pathogenic mutation. However, due to technological and human limitations, many genes associated with hereditary diseases remain unclear. Therefore, even if next-generation sequencing detects a pathogenic mutation, medical interpreters may overlook the gene's impact on the patient's disease because it was not previously recorded, leading to uncertainty about the cause. The gold standard for identifying disease-related genes remains biochemical molecular experimental methods, such as gene knockout and gene silencing. However, these methods are inefficient and costly, and cannot be used on a large scale to discover disease-related genes. In addition, some computer data processing and statistical methods based on past data can be used to discover new genes associated with hereditary diseases. However, these methods often fail to consider the differences in the impact of different mutation types on gene function, nor do they consider the influence of the mutation's location in the genome on gene function, resulting in newly discovered genes often not being associated with hereditary diseases.

[0030] In cases where parents are healthy but their child suffers from a developmental disorder, de novo mutations in the child are often related to the condition. If the frequency of these de novo mutations in certain genes is significantly higher than their frequency in the general population, then these de novo mutations are often associated with the disease, indicating that the genes containing these de novo mutations are related to the disease. Therefore, by using statistical methods to identify genes with a significantly higher frequency of de novo mutations in three individuals from a family with a genetic disease than in the general population, and comparing these genes with gene lists in previous literature and databases, it is possible to discover some genes that are not yet universally recognized as being associated with genetic diseases.

[0031] This application collects sequencing results from three samples from families with a hereditary disease, enriches the proband's de novo mutations, and uses statistical methods to identify a list of genes significantly enriched by the de novo mutations, thereby determining the list of genes most relevant to the hereditary disease. This allows for the identification of genes associated with the hereditary disease that are not explicitly recorded in previous literature or databases, thus expanding the list of genes relevant to the hereditary disease that medical interpreters are interested in, reducing the possibility of overlooking pathogenic mutations, and providing more accurate guidance for genetic counseling.

[0032] This application uses computer data processing and statistical methods, which are more efficient and less costly than biochemical molecular experimental methods. It also takes into account the differences in the impact of mutation type and mutation location on gene function, and uses rigorous statistical methods and effective statistical thresholds, which can effectively detect new genes related to hereditary diseases.

[0033] Reference Figure 1 , Figure 1This is a flowchart illustrating the screening method for genes related to genetic diseases based on novel mutations, as described in this application. In one specific embodiment, the screening method of this application may include the following steps:

[0034] 1) Collect a reference dataset, annotate the sample dataset, and organize it into the same format as the reference dataset.

[0035] 2) Group the reference dataset and sample dataset according to mutation type, mutation scoring software scoring range, whether it is located in a conserved region of the genome, and conservatism score, calculate the mutation score for each group, and obtain the mutation score matrix for different groups.

[0036] 3) Obtain the mutation score of each gene mutation according to the corresponding group, and sum all the mutation scores of each gene in the sample dataset to obtain the mutation score of each gene.

[0037] 4) Calculate the expected number of mutations for each gene in the reference dataset, establish a Poisson distribution model, simulate and calculate the probability that the mutation score of each gene under different mutation numbers is greater than the actual observed mutation score in the sample dataset, and sum the probabilities under different mutation numbers as the p-value of the gene.

[0038] 5) Calculate the statistical significance level of the sample dataset, compare the p-value of each gene with the significance level, and genes with p-values ​​less than the significance level are associated with the disease.

[0039] In one specific embodiment, the genetic disease can be a single-gene genetic disease or a polygenic genetic disease.

[0040] In one specific embodiment, the reference dataset includes information on gene mutations from a public database.

[0041] In one specific embodiment, gene mutation information related to the corresponding genetic disease can be obtained from public databases as a reference dataset. For example, data from the DDD (Deciphering Developmental Disorders) project can be obtained from the Sanger Institute website (ftp: / / ftp.sanger.ac.uk / pub / project / ddd / rates / all_rates_cadd_maf_shet_2020_03_05.txt.gz) as a reference dataset. It is understood that if the genetic disease under study is not related to developmental disorders, data from other relevant projects can be obtained as a reference dataset.

[0042] In one specific embodiment, gene mutation information obtained from public databases can be organized into a predetermined format. In another specific embodiment, the gene mutation information can be organized according to a predetermined format, including gene name, chromosome number, genomic location, reference base, mutated base, mutation type, mutation probability, mutation harmfulness prediction result (which may refer to the score from mutation scoring software), whether it is located in a conserved gene, and whether it is located in a highly conserved gene.

[0043] In one specific embodiment, the sample dataset includes information on de novo mutations in the genetic disease pedigree. It should be noted that a de novo mutation (or newly occurring mutation) is a gene mutation that newly occurs in the offspring's genome; the parents do not carry the mutation, meaning the mutation is not inherited from them, but rather arises spontaneously during the formation of individual germ cells (such as during gamete formation) or during the individual's own embryonic development.

[0044] In one specific embodiment, the gene mutation is a germline mutation. The forms of gene mutation include, but are not limited to: substitutions, deletions, duplications, insertions, inversions, and deletions / insertions. The locations of gene mutations include, but are not limited to: intronic regions within exons, 5'-UTR regions, and 3'-UTR regions. The types of gene mutations include: synonymous mutations, in-frame mutations (in-frame mutations refer to gene mutations resulting from the insertion or deletion of 3 or multiples of 3 bases), splicing mutations, frameshift mutations, missense mutations, and nonsense mutations.

[0045] In one specific embodiment, the sample dataset can be formatted to be the same as the gene mutation information in the reference dataset, for example, both being in a predetermined format.

[0046] In one specific embodiment, gene mutations in the reference dataset and sample dataset can be grouped, and a mutation score can be calculated for each group. For example, the reference dataset and sample dataset can be grouped according to mutation type, mutation scoring software scoring range, whether it is located in a conserved region of the genome, and conservatism score, and a mutation score can be calculated for each group to obtain a mutation score matrix for different groups.

[0047] In one specific embodiment, the mutation score of each mutation of each gene can be obtained according to the corresponding group, and the mutation score of each gene can be obtained by summing them.

[0048] In one specific embodiment, the expected number of mutations for each gene in the reference dataset can be calculated, a Poisson distribution model can be established, and the probability that the mutation score of each gene is greater than the actual observed sample mutation score under different mutation numbers can be simulated and calculated respectively. The probabilities under different mutation numbers are summed as the p-value of the gene. Figure 2 This is a Poisson distribution diagram of the screening method for genes related to genetic diseases based on novel mutations involved in this application, such as... Figure 2 As shown, the mutation score of each gene approximately follows a Poisson distribution. The area enclosed by the Poisson distribution curve, the dashed line of the mutation score of the sample genes, and the horizontal axis is the p-value of the gene.

[0049] Specifically, the value of p can be obtained according to the following formula (I):

[0050] in,

[0051] P(S>score|k)=sum(S k >score) / sum(S k )

[0052] In equation (III), P represents the probability, S is the mutation score for each gene, score is the actual sample mutation score for each gene, k is the number of mutations, μ is the expected number of mutations, e is the natural constant, and S k The variable represents the mutation score of each gene under the condition of k mutations, and sum represents the summation.

[0053] The present application will be further described in detail below through specific embodiments. The following embodiments are only for further illustration of the present application and should not be construed as limiting the present application.

[0054] Example:

[0055] In this embodiment, developmental disorders are used as the research object, and genes related to developmental disorders are screened based on novel mutations. The specific steps include:

[0056] (1) Data from the DDD (Deciphering Developmental Disorders) project was obtained from a public database—the Sanger Institute website (website address: ftp: / / ftp.sanger.ac.uk / pub / project / ddd / rates / all_rates_cadd_maf_shet_2020_03_05.txt.gz). This file was used as a reference dataset. The file is annotated according to the format in Table 1 (Table 1 is an example of the file format of the reference dataset and sample dataset), that is, the first column is the gene name, the second column is the chromosome number, the third column is the starting coordinate, the fourth column is the reference base, the fifth column is the mutated base, the sixth column is the mutation type, the seventh column is the mutation probability, the eighth column is the original score of the mutation scoring software CADD (Combined Annotation Dependent Depletion), the ninth column is the standard score of the mutation scoring software CADD, the tenth column is the maximum population frequency of the mutation, the eleventh column is whether the mutation is located in a conserved gene, and the twelfth column is whether the mutation is located in a highly conserved gene. There are a total of 111,414,306 mutation records.

[0057] Table 1

[0058]

[0059] (2) VCF files from 2500 families sequenced at Shenzhen Anji Kang'er Laboratory Co., Ltd. from 2019 to 2023 were selected. Novel mutations in the proband samples were annotated and compiled. The file format was consistent with that of the reference mutation file to obtain the sample dataset. Among them, the mutation type, the original score of the mutation software CADD, the standard score of the mutation software CADD, and the maximum population frequency of the mutation were obtained by annotation from VEP (Variant Effect Predictor) software; whether the mutation is located in a conserved gene and whether the mutation is located in a highly conserved gene were obtained from the same gene in the reference mutation set; the mutation probability column was empty, with a total of 45221 mutation records.

[0060] (3) Based on mutation type, the reference dataset and sample dataset were divided into synonymous mutations, in-frame mutations, splicing mutations, frameshift mutations, missense mutations, and nonsense mutations. Missense and nonsense mutations were further grouped according to the following criteria: six groups based on CADD standard score ranges: 0-6, 6-12, 12-18, 18-24, 24-30, and >30; two more groups based on whether they are located in conserved genes; and two more groups based on whether they are located in highly conserved genes. The remaining synonymous, in-frame, splicing, and frameshift mutations were divided into two groups based on whether they are located in conserved genes.

[0061] (4) Calculate the expected number of mutations in each group of the reference mutation: sum the mutation probabilities in the seventh column of the same group, and multiply the sum by the number of families in the sample; calculate the observed number of mutations in each group of the sample mutation: directly count the mutations in the same group. Divide the observed number of mutations in each group of the sample mutation by the expected number of mutations in the same group of the reference mutation to obtain the mutation score for each group.

[0062] (5) Calculate the sum of mutation scores for each gene in the sample dataset: For each mutation of each gene, obtain the mutation score from the corresponding group, and sum the mutation scores of each gene; calculate the expected number of mutations for each gene in the reference dataset, establish a Poisson distribution based on the expected number of mutations, and calculate the probability that the mutation score of a gene is greater than the sum of the actual observed mutation scores in the sample when the number of mutations is between 0 and 250. Then sum these probabilities to obtain the p-value of the gene, i.e.:

[0063]

[0064] in, P(S>score|k)=sum(S k >score) / sum(S k );

[0065] In the three formulas above: P represents probability, S is the sum of mutation scores for each gene, score is the sum of actual observed mutation scores for each gene, k is the number of mutations, μ is the expected number of mutations, e is the natural constant, and S... k The variable represents the sum of mutation scores for each gene under the condition of k mutations, where sum means summation.

[0066] The first formula represents the probability that the gene mutation score is greater than the sum of the actual observed mutation scores in the sample. The smaller this probability, the less likely the sum of the actual observed mutation scores of the gene will occur. This indicates that the sum of the actual observed mutation scores of the gene is significantly higher than the expected value in the general population, and therefore the novel mutations in the sample set are enriched in this gene.

[0067] The second formula represents the probability of a mutation number of k.

[0068] The meaning of the third formula is that when the number of mutations is k, the number of mutations in the reference dataset whose mutation scores are greater than the actual observed mutation scores of the sample is divided by the number of mutations in the reference dataset.

[0069] (6) Calculate the significance level α: The significance level used in statistics is generally 0.05. According to the Bonferroni correction principle, if n independent hypotheses are tested simultaneously on the same dataset, the statistical significance level used for each hypothesis should be 1 / n of the significance level when only one hypothesis is tested. Therefore, the significance level α is 0.05 divided by the number of families in the sample, that is, the significance level α is: 0.05 / 2500 = 2.0e-5.

[0070] (7) The p-value of each gene was compared with the significance level of 2.0e-5, and the genes were sorted in ascending order of p-value (see Table 2). Genes with p-values ​​less than the significance level were screened out. A total of 345 genes were finally screened out, and Table 2 contains these 345 genes:

[0071] Table 2

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078] (8) The selected genes were sorted in ascending order of p-value. In OMIM (Online Mendelian Inheritance in...) The website OMIM (Man's Online Mendelian Genetics Database) https: / / www.omim.org / was used to search for the 20 genes with the smallest p-values ​​(i.e., the top 20 genes in the gene list sorted from smallest to largest p-value). It was found that these genes are all associated with developmental disorders. For example, the CTCF gene is associated with autosomal dominant type 21 intellectual disability (omim ID: #615502); the SOX11 gene is associated with intellectual disability with microcephaly, with or without eye malformations or hypogonadism (omim ID: #615866); and the KIF1A gene is associated with NESCAV syndrome (omim ID: #614255). NESCAV syndrome (NESCAVS) is a neurodegenerative disease characterized by onset in infancy or early childhood. Affected individuals exhibit overall developmental delay, accompanied by delayed or difficult walking, and varying degrees of intellectual disability, speech delay, learning disabilities, and / or behavioral abnormalities. Genes with p-values ​​in the middle of the ranking, such as ARID2 (rank 106), were found to be associated with Coffin-Siris syndrome (omim ID: #617808), and IRF2BPL (rank 119) was associated with neurodevelopmental disorders, accompanied by motor regression and abnormalities, language loss, and epilepsy (omim ID: #618088). Genes with p-values ​​at the end of the ranking, such as NALCN (rank 334), were found to be associated with congenital limb and facial contractures, hypotonia, and developmental delay (omim ID: #616266), and ZMYM2 (rank 344) was associated with neurodevelopmental-craniofacial syndrome with variable kidney and heart abnormalities (omim ID: #619522). In summary, the gene list obtained using the screening method of this embodiment has a very strong association with developmental disorders.

[0079] (9) This embodiment identified 18 genes: AP2S1, ARF1, KCNK3, UPF1, RAB14, MMGT1, DDX23, ZFHX4, GIGYF1, PSMC5, HNRNPD, TCF7L2, MIB1, DSP, PRPF8, MSL2, MFN2, and ARHGAP35. Most of these genes are ranked after 200 in the list. However, the de novo mutations in these genes have been verified as true mutations using first-generation sequencing, and no de novo mutations have been found in known genes associated with developmental disorders in probands with these genes. Among them, some literature has also reported the discovery of a heterozygous de novo mutation of TCF7L2 (ranked 252) (PMID: 34003604) in some subjects of developmental disorders, and another literature has reported the discovery of a de novo mutation of DSP (ranked 152) (PMID: 26073755) in an Irish boy with mild global developmental delay. This indicates that the novel mutations in these genes are strongly correlated with the developmental disorders of patients, and therefore these genes are newly discovered genes related to developmental disorders in this embodiment.

[0080] The above description, in conjunction with specific embodiments, provides a further detailed explanation of this application and should not be construed as limiting the specific implementation of this application to these descriptions. Those skilled in the art to which this application pertains can make several simple deductions or substitutions without departing from the concept of this application.

Claims

1. A method for screening genes related to genetic diseases based on novel mutations, characterized in that, Includes the following steps: Obtain a reference dataset for the genetic disease, which includes information on gene mutations in a public database; Obtain a sample dataset, which includes information on de novo mutations in the genetic disease pedigree; Calculate the mutation score of each gene in the sample dataset, define the number of mutations of each gene in the sample dataset as the observed mutation number, define the number of mutations of each gene in the reference dataset as the expected mutation number, and the mutation score of a gene is the observed mutation number of the gene divided by the expected mutation number of the gene. Obtain the p-value for each gene. Based on the expected number of mutations for each gene in the reference dataset, establish a Poisson distribution model. Calculate the probability that the mutation score of each gene at different mutation numbers is greater than the actual observed mutation score in the sample dataset. Sum these probabilities at different mutation numbers as the p-value for the gene. The p-value is obtained according to the following formula (I): P(S>score|k) = sum(S k > score) / sum(Sk) Equation (III), In equations (I), (II), and (III), P represents the probability, S is the mutation score for each gene, score is the actual sample mutation score for each gene, k is the number of mutations, μ is the expected number of mutations, e is the natural constant, and S k The variable represents the mutation score of each gene under the condition of k mutations, and sum represents the summation; The p-value of each gene is compared with the significance level, and genes with p-values ​​less than the significance level are screened out. The screened genes are those related to the genetic disease.

2. The screening method according to claim 1, characterized in that, The newly discovered mutation is a germline mutation, and the genetic disease is a single-gene genetic disease or a polygenic genetic disease.

3. The screening method according to claim 1, characterized in that, The format of gene mutation information in the reference dataset and the sample dataset is the same.

4. The screening method according to claim 3, characterized in that, The information on gene mutations in the reference dataset includes the gene name, chromosome number, genomic location, reference base, mutated base, mutation type, mutation probability, mutation harmfulness prediction result, whether it is located in a conserved gene, and whether it is located in a highly conserved gene.

5. The screening method according to claim 4, characterized in that, Gene mutations are grouped according to the mutation type and / or the mutation harmfulness prediction result and / or whether they are located in a conserved gene and / or whether they are located in a highly conserved gene. The mutation score of each group is calculated, and then the mutation score of each gene is obtained.

6. The screening method according to claim 5, characterized in that, The mutation types include synonymous mutations, in-frame mutations, splicing mutations, frameshift mutations, missense mutations, and nonsense mutations. The mutation harmfulness prediction result is a CADD standard score.

7. The screening method according to claim 6, characterized in that, Gene mutations belonging to missense and nonsense mutations were grouped according to the following criteria: 0 < CADD score ≤ 6, 6 < CADD score ≤ 12, 12 < CADD score ≤ 18, 18 < CADD score ≤ 24, 24 < CADD score ≤ 30, and CADD score > 30. Further grouping was based on whether the mutation was located in a conserved gene, and then on whether it was located in a highly conserved gene. Gene mutations belonging to synonymous synchronization, in-frame mutations, splicing mutations, and frameshift mutations were grouped according to whether they were located in a conserved gene. A mutation score was calculated for each group, where the mutation score was the observed number of mutations divided by the expected number of mutations in the group.

8. A screening system for genetic disease-related genes based on novel mutations, characterized in that, The screening system includes the following modules: A module for obtaining a reference dataset, wherein the reference dataset includes information on gene mutations in a public database; A module for organizing sample datasets, wherein the sample datasets include information on de novo mutations in the genetic disease pedigrees; The mutation score calculation module is used to calculate the mutation score of each gene in the sample dataset. The number of mutations of each gene in the sample dataset is defined as the observed mutation number, the number of mutations of each gene in the reference dataset is defined as the expected mutation number, and the mutation score of a gene is the observed mutation number of the gene divided by the expected mutation number of the gene. The p-value calculation module establishes a Poisson distribution model based on the expected number of mutations for each gene in the reference dataset, calculates the probability that the mutation score of each gene is greater than the actual observed mutation score in the sample dataset under different mutation numbers, and sums the probabilities under different mutation numbers as the p-value of the gene; wherein, the p-value is obtained according to the following formula (I): P(S>score|k)=sum(S k >score) / sum(S k Equation (III), in Equations (I), (II), and (III), P represents the probability, S is the mutation score for each gene, score is the actual sample mutation score for each gene, k is the number of mutations, μ is the expected number of mutations, e is the natural constant, and S k The variable represents the mutation score of each gene under the condition of k mutations, and sum represents the summation; The gene screening module compares the p-value of each gene with a significance level, and filters out genes with p-values ​​less than the significance level. The filtered genes are those related to the genetic disease.

9. A screening device based on newly discovered mutations in genetic disease-related genes, characterized in that, It includes a memory and a processor, the memory being used to store a program, and the processor executing the program stored in the memory to implement the screening method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a program that can be executed by a processor to implement the screening method according to any one of claims 1 to 7.

Images