Method, system and device for analyzing gene mutation based on phenotype and computer readable storage medium

Abstract

The invention discloses a phenotype-based gene mutation analysis method, system and device and a computer readable storage medium, and the method comprises the following steps: data acquisition: acquiring phenotype and gene mutation data of a subject; phenotype description: describing the phenotype of the subject as a standard entry; gene assignment: assigning each gene of gene mutation of the subject according to genetic heterogeneity, genetic pattern and phenotype comparison to obtain a score of each gene; and sorting: sorting the gene mutation of the subject based on the score. According to the application, the phenotype correlation between the phenotype of the subject and the gene where the mutation is located is analyzed, the gene mutation is assigned according to the correlation, and then the mutation is sorted according to the score, so that medical interpretation personnel can preferentially pay attention to the pathogenic gene mutation possibly related to the phenotype of the subject; and the working efficiency of medical interpretation personnel can be improved.

Description

Technical Field

[0001] The present application relates to the field of genetic analysis technology, and in particular to a method, system, device and computer-readable storage medium for phenotypic analysis of gene mutations. Background Art

[0002] Genetic diseases are diseases caused by genetic factors, usually involving mutations in one or more genes. Genetic testing is a technical means of analyzing specific genes or gene regions in an individual's genome to determine genetic information, gene mutations and their potential impact. Genetic testing is widely used in the diagnosis of genetic diseases. It can help confirm patients who have already shown symptoms, and can also be used for carrier detection and assessment of family members' risks.

[0003] Genetic testing of the subjects usually detects a large number of gene mutations. Although there are functional prediction tools such as protein function prediction software (such as SIFT, PolyPhen2, MutationTaster, DANN, CADD, primateAI and other software) and population frequency databases (such as 1000Genomes Project, Exome Variant Server, The Exome Aggregation Consortium and other databases) to predict the pathogenicity of gene mutations, these methods mainly consider the functionality of gene mutations and do not consider the association between patient phenotype and mutations. After functional prediction, there may still be a large number of mutations of unknown significance. How to find mutations that may be related to the disease suffered by the subject from these large number of gene mutations? Doctors or medical interpreters need to spend a lot of time to consult literature and databases, which requires a high level of experience for the interpreters and is time-consuming. Summary of the invention

[0004] Based on the above problems, the present application provides a method, system, device and computer-readable storage medium for analyzing gene mutations based on phenotype.

[0005] This application adopts the following technical solutions:

[0006] The first aspect of the present application discloses a method for phenotypic analysis of gene mutations, comprising: data acquisition, acquiring data on the phenotype and gene mutation of a subject; phenotype description, comprising describing the phenotype of the subject as a standardized entry; gene scoring, comprising scoring each gene of the subject's gene mutation according to genetic heterogeneity, inheritance pattern and phenotypic comparison to obtain a score for each gene; sorting, sorting the subject's gene mutations based on the score, wherein the gene with a higher score has a higher possibility of being the disease-causing gene of the subject; wherein the gene scoring includes the following situations: if the disease corresponding to the gene of the gene mutation has genetic heterogeneity, the gene is assigned a score of 0; if the disease corresponding to the gene does not have genetic heterogeneity and the inheritance pattern of the disease corresponding to the gene is dominant inheritance: the standardized entry of the subject's phenotype is fully contained in the standardized entry of the phenotype corresponding to the gene, the gene is assigned a score of 2; the standardized entry of the subject's phenotype is partially contained in the standardized entry of the gene if the standardized terms of the phenotype of the subject and the corresponding phenotype standardized terms of the gene do not overlap, the phenotype of the subject and the phenotype standardized terms corresponding to the gene where the gene mutation of the subject is located are compared for similarity, and if the similarity is greater than a predetermined value, the gene is assigned a point of 1; if the disease corresponding to the gene does not have genetic heterogeneity, and the inheritance pattern of the disease corresponding to the gene is recessive inheritance: the standardized terms of the phenotype of the subject are all included in the phenotype standardized terms corresponding to the gene, the gene is assigned a point of 1; the standardized terms of the phenotype of the subject are partially included in the phenotype standardized terms corresponding to the gene, the gene is assigned a point of 0.5; if the standardized terms of the phenotype of the subject and the corresponding phenotype standardized terms of the gene do not overlap, the phenotype of the subject and the phenotype standardized terms corresponding to the gene where the gene mutation of the subject is located are compared for similarity, and if the similarity is greater than a first predetermined value, the gene is assigned a point of 0.5.

[0007] It should be noted that in this application, by analyzing the phenotypic correlation between the subject's phenotype and the gene where the mutation is located, scoring the gene mutation according to the correlation, and then sorting the mutation according to the score, the medical interpretation personnel can give priority to the gene mutation that may be related to the subject's phenotype, and the mutation is very likely to be the pathogenic mutation that causes the subject's phenotype, thereby helping to improve the work efficiency of the medical interpretation personnel. This application takes into account the association between the patient's phenotype and the gene where the mutation is located, which can save manpower and time costs, and can reduce the impact of human subjective interpretation, making the results more standardized.

[0008] In one implementation of the present application, the standardized terms are HPO terms. It should be noted that HPO (Human Phenotype Ontology) is a standardized vocabulary for describing human phenotypic characteristics. HPO terms refer to specific phenotypic characteristics or symptoms defined in HPO. Each term has a unique identifier (ID), usually starting with "HP:", followed by a series of numbers. For example, "HP:0000010" refers to the term "Recurrent urinary tract infections". It should be noted that HPO terms can be Chinese terms downloaded from CHPO. CHPO refers to "Chinese Human Phenotype Ontology" (Chinese Human Phenotype Standard Terminology Alliance), which is a term set similar to HPO, used to translate and optimize existing HPO terms abroad. CHPO aims to provide a standardized terminology system for genetic disease diagnosis, gene function research and clinical data analysis in the Chinese population.

[0009] In one implementation of the present application, the gene mutation is further included: filtering the gene mutation before the data is acquired, or filtering the gene mutation after sorting, wherein the filtering includes excluding benign mutations and suspected benign mutations in the gene mutation. It should be noted that gene mutations can be filtered through gnomAD, clinvar, dbSNP, ExAC, 1000GenomesProject databases, etc., to remove benign and suspected benign mutations, so as to reduce the gene mutations that the interpreter needs to pay attention to and improve the efficiency of interpretation.

[0010] In one implementation of the present application, the gene mutation is a de novo mutation, which refers to a gene mutation that appears for the first time in the subject, and is not inherited from the parents. It should be noted that when a child has a genetic disease and the parents are healthy, the de novo mutation is usually the cause of the child's illness. Accurately assessing the pathogenicity of the de novo mutation is crucial for the diagnosis and treatment of genetic diseases. Therefore, analyzing the de novo mutation of the subject is conducive to analyzing the pathogenic gene of the subject.

[0011] In one implementation of the present application, the similarity is calculated by the python open source package pyhpo, and the first predetermined value is 0.1. It should be noted that pyhpo is a python library designed for processing human phenotypic term set (HPO), website: https: / / pyhpo.readthedocs.io / en / latest / tutorial / basics.html, github address: https: / / github.com / anergictcell / pyhpo, which supports the calculation of phenotypic similarity. In bioinformatics, the calculation of phenotypic similarity is very important for genetic disease diagnosis, gene function research and clinical data analysis. The calculate_similarity function provided by pyhpo can be used to calculate the phenotypic similarity between two cases: similarity_score = calculate_similarity(phenotypes_case_1, phenotypes_case_2, ontology) print(f"Similarity score between Case 1and Case 2:{similarity_score}").

[0012] In one implementation of the present application, it also includes: storing the data of multiple previously analyzed subjects in a local database, and after the gene is scored and before sorting, the similarity of the normalized terms of the phenotype of the current subject and the normalized terms of the phenotype of the past subjects is calculated by traversing the local database, and the score of the gene in the past subjects whose similarity is greater than the second predetermined value is added to the score of the same gene of the current subject to obtain the score of each gene of the current subject. It should be noted that by first scoring the mutations of a large number of subjects, storing the information in the database, and scoring the subsequent patients (current subjects), first search the database for similar diseases and cases with mutations or new mutations on the same gene. The more such cases there are, the more likely the mutation or new mutation is to be related to the disease. Adding the gene scores of similar cases to the current mutation to obtain the final gene score can further enhance the association between the subject's phenotype and the gene where the new mutation is located.

[0013] In one implementation of the present application, the similarity = 2*the number of identical entries in the normalized entries of the phenotype of the current subject and the normalized entries of the phenotype of the past subject / (the number of normalized entries of the phenotype of the current subject + the number of normalized entries of the phenotype of the past subject), and the second predetermined value is 0.5.

[0014] In one implementation of the present application, it also includes: converting the score of each gene into a gene pathogenicity level, if the score of the gene is less than 1, the pathogenicity level of the gene is extremely low pathogenicity; if the score of the gene is greater than or equal to 1 and less than 2, the pathogenicity level of the gene is relatively low pathogenicity; if the score of the gene is greater than or equal to 2 and less than 4, the pathogenicity level of the gene is relatively strong pathogenicity; if the score of the gene is greater than or equal to 4, the pathogenicity level of the gene is extremely strong pathogenicity. It should be noted that, in this way, the genetic mutations of the subjects can be classified, so that the interpreters can give priority to mutations with strong pathogenicity.

[0015] The second aspect of the present application discloses a system for phenotypic analysis of gene mutations, including: a data acquisition module for acquiring data on the phenotype and gene mutation of a subject; a phenotype description module for describing the phenotype of the subject as a standardized entry; a gene scoring module for scoring each gene of the subject's gene mutation according to genetic heterogeneity, inheritance pattern and phenotypic comparison to obtain a score for each gene; a sorting module for sorting the subject's gene mutations based on the score, wherein the gene with a higher score is more likely to be the disease-causing gene of the subject; wherein the gene scoring module includes: if the disease corresponding to the gene of the gene mutation has genetic heterogeneity, then the gene is assigned a score of 0; if the disease corresponding to the gene does not have genetic heterogeneity and the inheritance pattern of the disease corresponding to the gene is dominant inheritance: the standardized entry of the subject's phenotype is fully included in the standardized entry of the phenotype corresponding to the gene, then the gene is assigned a score of 2; the standardized entry of the subject's phenotype is partially included in the standardized entry of the phenotype corresponding to the gene If the phenotype of the gene is included in the phenotype standard entry, the gene is assigned 1 point; if the standardized entry of the phenotype of the subject and the phenotype standard entry corresponding to the gene do not overlap, the phenotype of the subject and the phenotype standard entry corresponding to the gene where the gene mutation of the subject is located are compared for similarity, and if the similarity is greater than a predetermined value, the gene is assigned 1 point; if the disease corresponding to the gene does not have genetic heterogeneity, and the inheritance pattern of the disease corresponding to the gene is recessive inheritance: the standardized entry of the phenotype of the subject is fully included in the phenotype standard entry corresponding to the gene, the gene is assigned 1 point; if the standardized entry of the phenotype of the subject is partially included in the phenotype standard entry corresponding to the gene, the gene is assigned 0.5 points; if the standardized entry of the phenotype of the subject and the phenotype standard entry corresponding to the gene do not overlap, the phenotype of the subject and the phenotype standard entry corresponding to the gene where the gene mutation of the subject is located are compared for similarity, and if the similarity is greater than a first predetermined value, the gene is assigned 0.5 points.

[0016] The third aspect of the present application discloses a device for phenotypic analysis of gene mutations, comprising a memory and a processor, wherein the memory is used to store programs, and the processor implements the method for phenotypic analysis of gene mutations as described in the first aspect of the present application by executing the programs stored in the memory.

[0017] The fourth aspect of the present application discloses a computer-readable storage medium, in which a program is stored. The program can be executed by a processor to implement the method for phenotypic analysis of gene mutations as described in the first aspect of the present application.

[0018] The beneficial effects of this application are:

[0019] In this application, by analyzing the phenotypic correlation between the subject's phenotype and the gene where the mutation is located, the gene mutation is scored according to the correlation, and then the mutation is sorted according to the score, so that medical interpretation personnel can give priority to gene mutations that may be related to the subject's phenotype, and the mutation is very likely to be a pathogenic mutation that causes the subject's phenotype, thereby helping to improve the work efficiency of medical interpretation personnel. This application takes into account the association between the patient's phenotype and the gene where the mutation is located, which can save manpower and time costs, and can reduce the impact of human subjective interpretation, making the results more standardized. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of the method for phenotypic analysis of gene mutation involved in the present application.

[0021] Figure 2 It is a schematic diagram of the principle of calculating the phenotypic similarity threshold involved in this application.

[0022] Figure 3 This is the Sanger verification result of the DYNC1H1 gene mutation in the subject sample in the example of this application.

[0023] Figure 4 This is the Sanger verification result of the DYNC1H1 gene mutation in the sample of the mother of the subject in the example of this application.

[0024] Figure 5 This is the Sanger verification result of the DYNC1H1 gene mutation in the sample of the father of the subject in the example of this application.

[0025] Figure 6 This is the Sanger verification result of the BICD2 gene mutation in the subject sample in the example of this application.

[0026] Figure 7 This is the Sanger verification result of the BICD2 gene mutation in the sample of the mother of the subject in the example of this application.

[0027] Figure 8 This is the Sanger verification result of the BICD2 gene mutation in the sample of the subject's father in the example of this application. DETAILED DESCRIPTION

[0028] The present invention is further described in detail below by specific embodiments in conjunction with the accompanying drawings. In the following embodiments, many detailed descriptions are intended to enable the present application to be better understood. However, those skilled in the art can easily recognize that some of the features can be omitted in different situations, or can be replaced by other materials and methods. In some cases, some operations related to the present application are not shown or described in the specification, in order to avoid the core part of the present application being overwhelmed by too much description, and for those skilled in the art, it is not necessary to describe these related operations in detail, and the related operations can be fully understood according to the description in the specification and the general technical knowledge in the art.

[0029] In addition, the features, operations or characteristics described in the specification can be combined in any appropriate manner to form various implementations. At the same time, the steps or actions in the method description can also be interchanged or adjusted in a manner that is obvious to those skilled in the art. Therefore, the various sequences in the specification and the drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a required sequence, unless otherwise specified that a certain sequence must be followed.

[0030] Genetic testing of the subjects usually detects a large number of gene mutations. Although there are functional prediction tools such as protein function prediction software and population frequency databases to predict the pathogenicity of gene mutations, these methods mainly consider the functionality of gene mutations and do not consider the association between patient phenotypes and mutations. After functional prediction, there may still be a large number of mutations of unknown significance. How to find mutations that may be related to the disease suffered by the subject from these large number of gene mutations? Doctors or medical interpreters need to spend a lot of time to consult literature and databases, which requires a high level of experience for the interpreters and is time-consuming.

[0031] De novo mutations refer to mutations that are not inherited from parents. When a child has a genetic disease and the parents are healthy, de novo mutations are usually the cause of the child's illness. Accurately assessing the pathogenicity of de novo mutations is crucial for the diagnosis and treatment of genetic diseases. However, there are usually a large number of de novo mutations, which will cause doctors or medical interpreters to spend a lot of time consulting literature and databases to find pathogenic de novo mutations. This method is time-consuming and generally requires experienced personnel. There is currently no computer program that can specifically sort de novo mutations. Most common methods rely on protein function prediction software, population frequency data, etc. These methods have limitations and do not consider the association between patient phenotypes and de novo mutations.

[0032] This application first extracts standardized phenotypic information from the patient's medical records or disease descriptions, analyzes the association between these phenotypic information and the genes where the new mutations are located, and assigns scores to the new mutations in combination with similar cases in the past in the database, and uses database cases to enhance the association between the subject's phenotype and its gene mutation (find similar diseases in the database, and cases with new mutations on the same gene. The more such cases, the more likely the new mutation is to be related to the disease). By adding up these scores, the final score of each new mutation can be obtained, and all new mutations are sorted according to the final score, so that the new mutations most relevant to the patient's disease are ranked first, which can reduce the time for doctors and medical interpreters to find pathogenic mutations, and is more reasonable and reliable than existing computer methods. Focusing on the new mutations that are ranked first, and further verification (such as Sanger verification) on this basis, new mutations directly related to the disease can be obtained.

[0033] The present application provides a method, system, device and computer-readable storage medium for phenotypic analysis of gene mutations.

[0034] Figure 1 It is a flow chart of the method for phenotypic analysis of gene mutation involved in the present application.

[0035] like Figure 1 As shown, in a specific embodiment, a method for phenotypic analysis of gene mutations may include: data acquisition, acquiring data on the phenotype and gene mutation of the subject; phenotype description, describing the phenotype of the subject as a standard entry; gene scoring, assigning a score to each gene of the subject's gene mutation based on genetic heterogeneity, inheritance pattern and phenotypic comparison; assigning scores based on a local database, searching the local database for cases with similar phenotypes to the subject and mutations in the same gene, adding the score of the gene of the case to the score of the subject's gene to obtain a final score for each gene of the subject; and sorting, sorting the subject's gene mutations based on the score.

[0036] The present application is further described in detail below through specific examples. The following examples are only used to further illustrate the present application and should not be construed as limiting the present application.

[0037] It should be noted that this embodiment mainly analyzes new mutations, and can also be applied to other mutation geometries, such as scoring and ranking a set of mutations known to be associated with a certain disease. This embodiment does not filter the mutation list, and mutations can also be filtered according to actual research purposes.

[0038] Embodiment 1:

[0039] (1) Apply to download HPO Chinese entries from the CHPO official website (https: / / www.chinahpo.net / ). The file has 16493 lines and three columns. The first column is the HPO English entry, the second column is the HPO number, and the third column is the HPO Chinese entry, as shown in Table 1 below (due to copyright restrictions, only 10 entries are shown in this example):

[0040] Table 1

[0041] HPO English HPO Number HPO Chinese Recurrenturinarytractinfections HP:0000010 Recurrent urinary tract infections Neurogenicbladder HP:0000011 Neurogenic bladder dysfunction Urinary urgency HP:0000012 Urgent need to urinate Hypoplasia of theuterus HP:0000013 Uterine hypoplasia Abnormality of the bladder HP:0000014 Bladder abnormalities Bladderdiverticulum HP:0000015 Bladder diverticulum Urinary retention HP:0000016 Urinary retention Nocturia HP:0000017 Nocturia Urinary hesitancy HP:0000019 Hesitation in urination Urinary incontinence HP:0000020 Urinary incontinence Megacystis HP:0000021 Megacystis

[0042] Download the file corresponding to the gene and phenotype from the OMIM website (https: / / www.omim.org / ), and associate the gene with the HPO Chinese entry to generate a new file. This file contains 8962 records, the first column is the OMIM number of the phenotype, the second column is the English name of the disease, the third column is the inheritance mode of the disease, the fourth column is the gene name, the fifth column is the OMIM number of the gene, and the sixth column is the HPO number and Chinese entry, as shown in Table 2 below (due to copyright restrictions, this example only shows 5 entries):

[0043] Table 2

[0044]

[0045]

[0046]

[0047] (2) The subject is a fetus, and the fetal ultrasound test is abnormal. The fetal amniotic fluid sample is taken, and the parents' blood is drawn to perform family whole exome second-generation sequencing. The sequencing data are compared to obtain the VCF files of the three family members. After the family VCF is merged, the list of new mutations of the subject is analyzed. The first column format is the chromosome number of the mutation / hg19 coordinate / reference base / mutated base, and the second column is the gene name where the mutation is located. The list of new mutations is shown in Table 3 below:

[0048] Table 3

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055] (3) Extract standardized HPO entries from the patient's medical records or disease descriptions:

[0056] 1) Use the self-built synonym dictionary to match the original text. Each synonym corresponds to an HPO entry. That is, when any synonym appears in the original text, its corresponding HPO entry will be extracted. For example, 'NT thickening' corresponds to

[0057] 'HP:0010880 Increased thickness of nuchal translucency', if 'NT thickening' appears in the patient's medical record, then 'HP:0010880 Increased thickness of nuchal translucency' will be one of the HPO entries for that patient.

[0058] 2) First, the patient's medical records are divided into different sentences according to Chinese punctuation marks (such as periods and commas), and the HPO Chinese entries are segmented using open source software. If each word after segmentation can be completely matched in any sentence of the original text, then the HPO entry will be used as one of the patient's HPO entries. For example, 'HP:0007340 Lower limb muscle weakness' will become 'lower limb', 'muscle' and 'weakness' after segmentation, and the patient's medical record has 'the patient reported that the lower limb muscles were slightly weak', then 'HP:0007340 Lower limb muscle weakness' will be used as one of the patient's HPO entries.

[0059] 3) In some cases, neither of the above two methods can match HPO entries. In this case, fuzzy matching is used to compare the word similarity between each sentence and the HPO Chinese entry, and the one with the highest similarity is selected. For example, if the patient's medical record only has a simple description of "oligospermia and asthenozoospermia", all HPO Chinese entries are taken at this time, and each Chinese entry is compared with "oligospermia and asthenozoospermia". Finally, "HP:0000798 Oligospermia" has the highest similarity with the original text, so "HP:0000798 Oligospermia" is selected as one of the patient's HPO entries.

[0060] The phenotype of the subject in this example and the results of the standardized description into HPO terms are shown in Table 4 below:

[0061] Table 4

[0062]

[0063]

[0064] (4) Genetic scoring, local database comparison and sorting of the newly detected mutation list of the subject:

[0065] 1) Record relevant information according to the OMIM file (i.e., the file shown in Table 2): the inheritance mode of each gene, that is, dominant or recessive inheritance; analyze the genetic heterogeneity of the diseases corresponding to the genes: record all the diseases corresponding to each gene, and record all the genes corresponding to each disease. When a disease corresponds to multiple genes, the disease has genetic heterogeneity. When a disease corresponds to only one gene and is unrelated to other genes, there is no genetic heterogeneity.

[0066] 2) The phenotypic HPO entry combination corresponding to each newly-occurring mutation gene is compared with the phenotypic HPO entry combination of the patient, and the genetic heterogeneity is scored according to the inheritance mode:

[0067] i. The disease corresponding to the gene is dominantly inherited: a) If all the phenotypic HPO entries of the patient are included in the phenotypic HPO corresponding to the gene or all the phenotypic HPO corresponding to the gene are included in the phenotypic HPO entries of the patient, the phenotypic similarity is judged to be high and a score of 2 is assigned; b) If more than one (not all) of the phenotypic HPO entries of the patient appear in the phenotypic HPO corresponding to the gene, the phenotypic similarity is judged to be average and a score of 1 is assigned; c) If the phenotypic HPO entries of the patient and the phenotypic HPO corresponding to the gene do not overlap at all, the python open source package pyhpo is used to calculate the similarity of the two groups of HPOs. If the similarity score is greater than 0.1 (the threshold value is described later in the text), a score of 1 is assigned;

[0068] ii. The disease corresponding to the gene is recessive: a) If all the phenotypic HPO entries of the patient are included in the phenotypic HPO corresponding to the gene or all the phenotypic HPO corresponding to the gene are included in the phenotypic HPO entries of the patient, the phenotypic similarity is judged to be high and a score of 1 is assigned; b) If more than one (not all) of the phenotypic HPO entries of the patient appear in the phenotypic HPO corresponding to the gene, the phenotypic similarity is judged to be average and a score of 0.5 is assigned; c) If the phenotypic HPO entries of the patient and the phenotypic HPO corresponding to the gene do not overlap at all, the python open source package pyhpo is used to calculate the similarity of the two groups of HPOs. If the similarity score is greater than 0.1, a score of 0.5 is assigned;

[0069] iii. If the disease corresponding to the gene has genetic heterogeneity, the mutations in the gene will be assigned a score of 0.

[0070] The similarity threshold is obtained by the following method: Figure 2 This is a schematic diagram of the phenotypic similarity threshold calculation involved in this application, combined with Figure 2As described, a number of patient samples that have undergone whole exome sequencing of three people in a family are selected. The phenotypes of these patients have been standardized by experienced medical personnel into HPO term combinations, and this HPO term combination is recorded as patient_HPO. And each patient has at least one pathogenic gene identified based on factors such as the patient's phenotype and the pathogenic effect of the new mutation. Each pathogenic gene can find a corresponding disease syndrome in the OMIM database, and these disease syndromes correspond to several HPO terms. The HPO term combination corresponding to each disease is recorded as gene_HPO1, gene_HPO2, ..., gene_HPOn. The similarity score S between each gene_HPO and patient_HPO is calculated by pypho (the calculate_similarity function provided by pyhpo is used to calculate the phenotypic similarity between two cases:

[0071] similarity_score = calculate_similarity(phenotypes_case_1, phenotypes_case_2, ontology) print(f"Similarity score between Case 1and Case 2:{similarity_score}")), and finally each patient takes the maximum similarity score Smax. Collect the Smax of all patients, assuming that these scores follow a normal distribution, and the expected value plus or minus three times the standard deviation includes the scores of 99.7% of the patients, then take the expected value minus three times the standard deviation as the score threshold Scutoff. In this embodiment, by calculating the distribution of the scores of 100 samples, the calculated score threshold Scutoff is 0.1.

[0072] 3) Search for cases in the local database that have any HPO in the HPO entry combination that is the same as the current patient's HPO entry combination, and narrow the search scope first. Then, calculate the similarity between the current patient's HPO combination and these HPO combinations by traversing, and the similarity calculation formula is: 2*the number of patient HPO combinations that are the same as the HPO combination of the current database case / the number of patient HPOs + the number of HPOs of the current database case. For example, the HPO combination of the current patient is HP:0001155: hand abnormality HP:0001838: rocking chair foot HP:0002533: posture abnormality HP:0006101: finger syndactyly, and there is a case in the database with the HPO combination of HP:0002533: posture abnormality HP:0006101: finger syndactyly, so the similarity is 2*2 / (4+2)=0.6. Select the records with phenotypic similarity greater than 0.5. If there is no record greater than 0.5, do not proceed to the following steps. Then select the record with the highest similarity. If there are multiple records with the same similarity, select the one with the highest mutation score, and then add its mutation score to the mutation of the current patient to get the final score. After the scoring is completed, the current patient sample number, the information of the mutation, including the chromosome number, genome coordinates (hg19 version), as well as the reference base, the mutant base, the gene where it is located, the patient's HPO entry and Chinese, and the pathogenicity score of the new mutation are stored in the database.

[0073] (5) The final scores and intermediate results are output to the following Table 5, which is sorted in descending order according to the gene scores:

[0074] Table 5

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086] In Table 5, the first column is the mutation information (chromosome number of mutation / hg19 coordinate / reference base / mutated base); the second column is the name of the mutated gene; the third column is the inheritance mode of the gene, AD is dominant inheritance, and AR is recessive inheritance; the fourth column is whether the disease corresponding to the gene is genetic heterogeneity, if so, it is 1, otherwise it is 0; the fifth to seventh columns are gene score correlation, the fifth column is the number of patients' phenotypic HPOs that are the same as the HPO corresponding to the gene, and the sixth column is the similarity between the patient's HPO term combination calculated by the pyhpo package and the HPO combination corresponding to the gene; the seventh column is the score given to the current mutation based on phenotypic comparison, gene inheritance mode, and genetic heterogeneity; the eighth to ninth columns are local database search correlation, the eighth column is the medical record number and mutation information of similar cases in the database, and the ninth column is the score of the gene in the case; the tenth column is the final score of the mutation, and the list of new mutations is sorted in descending order according to the final score.

[0087] The following is a schematic analysis of the above table using the score calculation process of the new mutation of the DYNC1H1 gene as an example:

[0088] 1) The patient's HPO combination is HP:0001155: hand abnormality, HP:0001838: rocker foot, HP:0002533: posture abnormality, HP:0006101: finger syndactyly. There are three diseases related to DYNC1H1 in the OMIM database, and their OMIM IDs are #614228, #614563 and #158600 respectively.

[0089] 2) The number of HPO terms corresponding to these three OMIM IDs that are completely consistent with the patient's HPO combination is 0, so the value of the fifth column is 0.

[0090] 3) The HPO combination corresponding to #614563 is "HP:0002365 Brainstem dysplasia, HP:0001249 Intellectual disability, HP:0001302 Paleocysia, HP:0001250 Epileptic seizures, HP:0000252 Microcephaly, HP:0003477 Axonal peripheral neuropathy, HP:0002079 Corpus callosum dysplasia, HP:0007359 Focal epileptic seizures, HP:0200055 Small hands, HP:0001288 Unsteady gait, HP:0000494 Downward oblique palpebral fissure, HP:0000006 Autosomal dominant inheritance, HP:0002510 spastic quadriplegia, HP:0001357 plagiocephaly, HP:0001760 foot abnormality, HP:0001321 cerebellar hypoplasia, HP:0011220 frontal bossing, HP:0001252 hypotonia, HP:0001265 decreased tendon reflexes". This combination has a similarity of 0.26 with the patient's HPO combination of "HP:0001155: hand abnormality, HP:0001838: rocker foot, HP:0002533: abnormal posture, HP:0006101: syndactyly" calculated by pyhpo, while the similarity scores of the other two OMIM IDs are 0.11 and 0.004 respectively. The maximum value of 0.26 is taken as the value of the sixth column.

[0091] 4) There is no genetic heterogeneity in diseases related to the DYNC1H1 gene. The inheritance mode of the DYNC1H1 gene is autosomal dominant. If the similarity score between the HPO combination of the disease related to the DYNC1H1 gene and the HPO combination of the patient is 0.26 and greater than 0.1, then according to the scoring rule, the DYNC1H114 / 102476743 / C / T mutation is assigned 1 point, which is the value of the seventh column.

[0092] 5) A similar disease is searched in the local database, and it is a new mutation case that also occurs in the DYNC1H1 gene: AS23184784:chr14:102467292. The score of the DYNC1H1 gene in this case is 4 (the value in the ninth column). This score is added to the current case, and the final score of 14 / 102476743 / C / T is 5, which is the value in the tenth column.

[0093] For the mutations of DYNC1H1 ranked first and BICD2 ranked second in the table above, the samples of the three family members have been verified by Sanger, confirming that these two mutations are indeed new mutations. Figures 3 to 8 As shown, Figure 3It is the Sanger verification result of the DYNC1H1 gene mutation of the subject sample in the embodiment of the present application, which shows that the subject carries the DYNC1H1 gene mutation: 14 / 102476743 / C / T; Figure 4 It is the Sanger verification result of the DYNC1H1 gene mutation of the sample of the subject's mother in the embodiment of the present application, which shows that there is no mutation at 14 / 102476743 of the subject's mother; Figure 5 It is the Sanger verification result of the DYNC1H1 gene mutation of the sample of the father of the subject in the embodiment of the present application, which shows that there is no mutation at 14 / 102476743 of the father of the subject; Figure 6 It is the Sanger verification result of the BICD2 gene mutation of the subject sample in the embodiment of the present application, which shows that the subject carries the BICD2 gene mutation: 9 / 95484981 / T / G; Figure 7 It is the Sanger verification result of the BICD2 gene mutation of the sample of the subject's mother in the embodiment of the present application, which shows that there is no mutation at 9 / 95484981 of the subject's mother; Figure 8 This is the Sanger verification result of the BICD2 gene mutation in the sample of the subject's father in the example of this application, showing that there is no mutation at 9 / 95484981 in the subject's father.

[0094] Mutations in the DYNC1H1 gene are usually associated with complex cortical dysplasia with other brain malformations type 13, axonal Charcot-Marie-Tooth disease type 2O, and spinal muscular atrophy type 1 (lower limb involvement type); mutations in the BICD2 gene are usually associated with autosomal dominant lower limb involvement type spinal muscular atrophy type 2B and autosomal dominant lower limb involvement type spinal muscular atrophy type 2A. Experienced medical interpreters determined that these two mutations may be pathogenic and are closely related to the phenotype of the fetus. This proves that the present application can effectively rank new mutations related to the patient's phenotype, and can be quickly noticed by doctors and medical interpreters.

[0095] The above contents are further detailed descriptions of the present application in combination with specific implementation methods, and it cannot be determined that the specific implementation of the present application is limited to these descriptions. For ordinary technicians in the technical field to which the present application belongs, several simple deductions or substitutions can be made without departing from the concept of the present application.

Claims

1. A method for analyzing gene mutation based on phenotype, characterized in that: include: Data acquisition: obtaining the phenotype and gene mutation data of the subjects; Phenotypic description, including describing the phenotype of the subject as a standard term; Gene scoring, including assigning a score to each gene of the subject's gene mutation based on genetic heterogeneity, inheritance pattern and phenotypic comparison to obtain a score for each gene; Sorting, ranking the gene mutations of the subject based on the scores, wherein the gene with a higher score has a higher possibility of being the disease-causing gene of the subject; Wherein, the gene scoring includes the following situations: If the disease corresponding to the gene with the gene mutation is genetically heterogeneous, the gene is assigned a score of 0; If the disease corresponding to the gene does not have genetic heterogeneity, and the inheritance pattern of the disease corresponding to the gene is dominant inheritance: if the standardized terms of the phenotype of the subject are all included in the phenotype standardized terms corresponding to the gene, then the gene is assigned 2 points; if the standardized terms of the phenotype of the subject are partially included in the phenotype standardized terms corresponding to the gene, then the gene is assigned 1 point; if the standardized terms of the phenotype of the subject and the phenotype standardized terms corresponding to the gene do not overlap, then the phenotype of the subject and the phenotype standardized terms corresponding to the gene where the gene mutation of the subject is located are compared for similarity, and if the similarity is greater than a predetermined value, then the gene is assigned 1 point; If the disease corresponding to the gene does not have genetic heterogeneity, and the inheritance pattern of the disease corresponding to the gene is recessive inheritance: if the standardized entries of the phenotype of the subject are all included in the phenotype standardized entries corresponding to the gene, then the gene is assigned 1 point; if the standardized entries of the phenotype of the subject are partially included in the phenotype standardized entries corresponding to the gene, then the gene is assigned 0.5 points; if there is no overlap between the standardized entries of the phenotype of the subject and the phenotype standardized entries corresponding to the gene, then the phenotype of the subject and the phenotype standardized entries corresponding to the gene where the gene mutation of the subject is located are compared for similarity, and if the similarity is greater than a first predetermined value, then the gene is assigned 0.5 points.

2. The method according to claim 1, characterized in that The standard terms are HPO terms.

3. The method according to claim 1, characterized in that Also includes: The gene mutations are filtered before the data are acquired, or the gene mutations are filtered after sorting, wherein the filtering includes excluding benign mutations and suspected benign mutations in the gene mutations.

4. The method according to claim 1, characterized in that: The gene mutation is a de novo mutation, which refers to a gene mutation that occurs for the first time in the subject and is not inherited from parents.

5. The method according to claim 1, characterized in that The similarity is calculated by using the python open source package pyhpo, and the first predetermined value is 0.

1.

6. The method according to any one of claims 1 to 5, characterized in that: Also includes: The data of multiple subjects analyzed in the past are stored in a local database. After the gene is scored and before sorting, the similarity between the normalized entries of the phenotype of the current subject and the normalized entries of the phenotype of the past subjects is calculated by traversing the local database, and the scores of genes in the past subjects whose similarity is greater than a second predetermined value are added to the scores of the same genes of the current subject to obtain the scores of each gene of the current subject.

7. The method according to claim 6, characterized in that The similarity = 2*the number of identical entries in the normalized entries of the current subject's phenotype and the normalized entries of the past subject's phenotype / (the number of normalized entries of the current subject's phenotype + the number of normalized entries of the past subject's phenotype), and the second predetermined value is 0.

5.

8. A system for phenotypic analysis of gene mutations, characterized in that: include: A data acquisition module, used to obtain the phenotype and gene mutation data of the subject; A phenotype description module, used to describe the phenotype of the subject as a standard term; A gene scoring module is used to score each gene of the subject's gene mutation according to genetic heterogeneity, inheritance pattern and phenotypic comparison to obtain a score for each gene; A ranking module is used to rank the gene mutations of the subject based on the scores, wherein the gene with a higher score is more likely to be the disease-causing gene of the subject; Wherein, the gene scoring module includes: If the disease corresponding to the gene with the gene mutation is genetically heterogeneous, the gene is assigned a score of 0; If the disease corresponding to the gene does not have genetic heterogeneity, and the inheritance pattern of the disease corresponding to the gene is dominant inheritance: if the standardized terms of the phenotype of the subject are all included in the phenotype standardized terms corresponding to the gene, then the gene is assigned 2 points; if the standardized terms of the phenotype of the subject are partially included in the phenotype standardized terms corresponding to the gene, then the gene is assigned 1 point; if the standardized terms of the phenotype of the subject and the phenotype standardized terms corresponding to the gene do not overlap, then the phenotype of the subject and the phenotype standardized terms corresponding to the gene where the gene mutation of the subject is located are compared for similarity, and if the similarity is greater than a predetermined value, then the gene is assigned 1 point; If the disease corresponding to the gene does not have genetic heterogeneity, and the inheritance pattern of the disease corresponding to the gene is recessive inheritance: if the standardized entries of the phenotype of the subject are all included in the phenotype standardized entries corresponding to the gene, then the gene is assigned 1 point; if the standardized entries of the phenotype of the subject are partially included in the phenotype standardized entries corresponding to the gene, then the gene is assigned 0.5 points; if there is no overlap between the standardized entries of the phenotype of the subject and the phenotype standardized entries corresponding to the gene, then the phenotype of the subject and the phenotype standardized entries corresponding to the gene where the gene mutation of the subject is located are compared for similarity, and if the similarity is greater than a first predetermined value, then the gene is assigned 0.5 points.

9. A device for analyzing gene mutation based on phenotype, characterized in that: The method comprises a memory and a processor, wherein the memory is used to store a program, and the processor implements the method for analyzing gene mutation based on phenotype according to any one of claims 1 to 7 by executing the program stored in the memory.

10. A computer-readable storage medium, characterized in that: The storage medium stores a program, and the program can be executed by a processor to implement the method for phenotypic analysis of gene mutations as described in any one of claims 1 to 7.

Images