Gene variation detection method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a gene variation detection method and device, electronic equipment and a storage medium, and belongs to the technical field of gene variation detection. The method comprises the following steps: fusing a long read length variation detection result and a short read length variation detection result of a to-be-detected gene segment; for the gene loci with inconsistent sequencing results, firstly detecting whether the gene loci are in a genome complex region or not, and if the gene loci are in the genome complex region, determining a variation detection result of the gene loci based on a long-read-long variation detection result; and if not, determining the variation detection result of the gene locus based on the short-read-long variation detection result. The method can improve the accuracy of genome variation detection of the to-be-detected gene segment.

Description

Technical Field

[0001] The present disclosure relates to the technical field of gene sequencing, and in particular to a gene variation detection method, device, electronic device and storage medium. Background Art

[0002] Genetic variation detection is an important area of ​​genomic research, which involves identifying and analyzing variations in an individual's genome that may cause disease, affect gene expression, or cause individual phenotypic differences. Genetic variation detection technology has a wide range of applications in many fields, including but not limited to genetic disease diagnosis, tumor gene detection, and drug development.

[0003] Genomic variation detection generally involves comparing the sequencing data of an individual's genome with a linear reference genome sequence to identify the differences between the individual's genome and the reference genome, thereby characterizing the variation information of the individual's genome.

[0004] In the prior art, the accuracy of detecting gene mutations in genes in the genome is not high. Summary of the invention

[0005] The main purpose of the embodiments of the present application is to propose a gene variation detection method, device, electronic device and storage medium, aiming to improve the accuracy of gene variation detection.

[0006] To achieve the above objectives, a first aspect of an embodiment of the present application provides a method for detecting gene variation, the method comprising: Comparing the long-read sequencing data of the gene fragment to be tested to the reference genome to obtain first sequencing result comparison data, and comparing the short-read sequencing data of the gene fragment to be tested to the reference genome to obtain second sequencing result comparison data; Determine a first variation detection result corresponding to each gene site in the gene segment to be tested based on the first sequencing result comparison data, and determine a second variation detection result corresponding to each gene site in the gene segment to be tested based on the second sequencing result comparison data; Identify a first target gene site where the first variation detection result is different from the second variation detection result, and when the first target gene site is in a complex genomic region of the gene fragment to be tested, determine a third variation detection result of the first target gene site according to the second variation detection result and variation detection parameters of each gene site; When the first target gene site is not in the complex region of the genome, determining a third variation detection result of the first target gene site according to the first variation detection result and the variation detection parameter; Based on the third variation detection result, multiple first variation sites where variations exist are determined, and according to the first variation sites and the second variation sites, a variation detection result of a target gene of the gene fragment to be tested is determined, wherein the second variation site is a variation site where there is an intersection between the first variation detection result and the second variation detection result.

[0007] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a gene mutation detection device, the device comprising: A comparison unit, used to compare the long-read sequencing data of the gene fragment to be tested to the reference genome to obtain first sequencing result comparison data, and to compare the short-read sequencing data of the gene fragment to be tested to the reference genome to obtain second sequencing result comparison data; a first determining unit, configured to determine a first variation detection result corresponding to each gene site in the gene segment to be tested based on the first sequencing result comparison data, and to determine a second variation detection result corresponding to each gene site in the gene segment to be tested based on the second sequencing result comparison data; an identification unit, configured to identify a first target gene site for which a first variation detection result is different from a second variation detection result, and when the first target gene site is located in a complex genomic region of the gene fragment to be tested, determine a third variation detection result of the first target gene site according to the second variation detection result and a variation detection parameter of each gene site; a second determining unit, configured to determine, when the first target gene site is not located in the complex region of the genome, a third variation detection result of the first target gene site according to the first variation detection result and the variation detection parameter; A third determination unit is used to determine multiple first variation sites where variations exist based on the third variation detection result, and to determine the target gene variation detection result of the gene fragment to be tested according to the first variation site and the second variation site, wherein the second variation site is a variation site where there is an intersection between the first variation detection result and the second variation detection result.

[0008] In some embodiments, the identification unit comprises: a first determining subunit, configured to determine a third variation detection result of the first target gene site according to the second variation detection result when the first target gene site is in the complex region of the genome and the second variation detection result corresponding to the first target gene site is a variation site; a first acquisition subunit, configured to acquire, when the first target gene site is in the complex region of the genome and the second variation detection result corresponding to the first target gene site is a non-variable site, a first variation detection parameter and a first reference variation parameter range corresponding to the first target gene site; The second determining subunit is used to determine a third variation detection result of the first target gene site according to the first variation detection parameter and the first reference variation parameter range.

[0009] Optionally, in some embodiments, the second determining subunit includes: A first determination module is configured to determine, when the variation detection parameter is within the reference variation parameter range, that the third variation detection result of the first target gene site is a variation site; The second determination module is used to determine that the third variation detection result of the first target gene site is a non-variation site when the variation detection parameter is not within the reference variation parameter range.

[0010] Optionally, in some embodiments, the second determining unit includes: a third determination subunit, configured to determine a third variation detection result of the first target gene site according to the first variation detection result when the first target gene site is not located in the complex region of the genome and the first variation detection result corresponding to the first target gene site is a variation site; a second acquisition subunit, configured to acquire, when the first target gene site is not located in the complex region of the genome and the first variation detection result corresponding to the first target gene site is not a variation site, a second variation detection parameter and a second reference variation parameter range corresponding to the first target gene site; The fourth determining subunit is used to determine a third variation detection result of the first target gene site according to the second variation detection parameter and the second reference variation parameter range.

[0011] Optionally, in some embodiments, the first determining unit includes: an evaluation subunit, configured to perform a quality evaluation on the variant sites determined based on the first sequencing result comparison data based on a first preset neural network model, and divide the variant sites into first candidate variant sites having a quality value higher than a first threshold and second candidate variant sites having a quality value not higher than the first threshold according to the quality values ​​obtained by the evaluation; a screening subunit, configured to screen the second candidate variant sites based on a second preset neural network model, and remove the second candidate variant sites determined by the model as non-variants; The fifth determination subunit is used to determine the first variation detection result corresponding to each gene site in the gene fragment to be tested according to the first candidate variation site and the screened second candidate variation site.

[0012] Optionally, in some embodiments, the evaluation subunit comprises: A third determination module, configured to determine a plurality of fourth candidate variation sites according to the first sequencing result comparison data; a first acquisition module, configured to acquire sequencing base ratio information of the fourth candidate variant site and its associated sites, and generate a first evaluation feature for each of the fourth candidate variant sites based on the sequencing base ratio information; The first calling module is used to call the first preset neural network model to perform variation quality assessment on the first assessment feature, and determine the quality value corresponding to each of the fourth candidate variation sites according to the assessment result.

[0013] Optionally, in some embodiments, the gene variation detection device further includes: The fourth determination module is used to determine the first threshold based on a preset proportional quantile value of the quality values ​​corresponding to all the fourth candidate variant sites.

[0014] Optionally, in some embodiments, the screening subunit comprises: a second acquisition module, configured to acquire, for each of the second candidate variation sites, variation assessment data within a preset assessment region of each of the second candidate variation sites, and determine a second assessment feature of each of the second candidate variation sites according to the variation assessment data; wherein the preset assessment region includes the second candidate variation site and its upstream and downstream regions; The second calling module is used to call the second preset neural network model to perform variation identification on the second evaluation feature, and determine the second candidate variation site determined by the model as a variation site as a third candidate variation site.

[0015] Optionally, in some embodiments, the second evaluation feature includes mutation evaluation information, base quality information and genotyping information.

[0016] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the gene variation detection method described in the first aspect is implemented.

[0017] To achieve the above-mentioned purpose, the fourth aspect of the embodiments of the present application proposes a storage medium, which stores a computer program. When the computer program is executed by a processor, the gene variation detection method described in the first aspect is implemented.

[0018] To achieve the above-mentioned purpose, the fifth aspect of an embodiment of the present application proposes a computer program product, which includes a computer program, and the computer program is read and executed by a processor of a computer device, so that the computer device executes the gene variation detection method described in the first aspect.

[0019] The gene variation detection method proposed in the embodiment of the present application is to obtain first sequencing result comparison data by aligning the long-read sequencing data of the gene fragment to be tested to the reference genome, and to obtain second sequencing result comparison data by aligning the short-read sequencing data of the gene fragment to be tested to the reference genome; determining the first variation detection result corresponding to each gene site in the gene fragment to be tested based on the first sequencing result comparison data, and determining the second variation detection result corresponding to each gene site in the gene fragment to be tested based on the second sequencing result comparison data; identifying a first target gene site whose first variation detection result is different from the second variation detection result, and when the first target gene site is in a complex genomic region of the gene fragment to be tested, determining a third variation detection result of the first target gene site according to the second variation detection result and a variation detection parameter of each gene site; when the first target gene site is not in a complex genomic region, determining a third variation detection result of the first target gene site according to the first variation detection result and the variation detection parameter; determining a plurality of first variation sites with variations based on the third variation detection result, and determining a target gene variation detection result of the gene fragment to be tested according to the first variation site and the second variation site, wherein the second variation site is a variation site having an intersection between the first variation detection result and the second variation detection result.

[0020] It can be seen that the gene variation detection method provided in the embodiment of the present application complements the advantages of variation detection based on short-read sequencing alignment with the advantages of variation detection based on nanopore sequencing alignment, so that in non-genomic complex regions, short-read sequencing alignment results are mainly used to obtain accurate variation detection results, and in areas with poor short-read sequencing alignment, nanopore sequencing alignment results are mainly used, thereby improving the accuracy of variation detection in the entire genome. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings are used to provide further understanding of the technical solution of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the technical solution of the present application and do not constitute a limitation on the technical solution of the present application.

[0022] Figure 1 Schematic diagram of inaccurate gene variation detection caused by inaccurate short-read sequencing alignment; Figure 2 is a schematic diagram of the process of the variation detection method provided in this application; Figure 3 yes Figure 2 Flow chart of step S202; Figure 4 yes Figure 3 A flowchart of step S301 in FIG. Figure 5 yes Figure 3 Another flow chart of step S301 in FIG. Figure 6 yes Figure 3 Flow chart of step S302 in FIG. Figure 7 yes Figure 6 Flow chart of step S601 in FIG. Figure 8 yes Figure 2 Flow chart of step S203 in FIG. Fig. 9 yes Figure 2 Flow chart of step S204 in FIG. Fig.10 is another schematic diagram of the process of gene variation detection provided by the present application; Fig.11 is a flow chart of a gene variation detection device provided in an embodiment of the present application; Fig.12 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] Before further describing the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations: Nanopore sequencing (NS): also known as the third-generation sequencing technology or long-read sequencing technology, is a single-molecule sequencing method based on nanopore technology. This method uses electrical signals to directly detect the base sequence of DNA or RNA molecules. Nanopore sequencing is a type of long-read sequencing technology.

[0025] Short Read Sequencing (SRS): It is a sequencing technology commonly used in genomic research, which mainly involves cutting DNA into shorter fragments (usually tens to hundreds of base pairs) and then performing high-throughput sequencing on these fragments. Among them, the second-generation sequencing technology, also known as the next-generation sequencing technology (Next-Generation Sequencing, NGS) is a short-read sequencing technology.

[0026] Gene mutation detection: A technique used to identify changes in the structure or composition of DNA or RNA sequences, which may be caused by genetic factors, environmental factors, or the interaction of both. Gene mutation detection can be divided into single gene mutation detection and genome mutation detection. In the embodiment of the present application, gene mutation detection can specifically be genome mutation detection.

[0027] In the related technology, due to the unique variation information existing in individuals and populations, it contains information related to the occurrence and development of diseases, clinical diagnosis and evolutionary mechanisms. Therefore, whole-genome variation detection is of great significance in exploring species evolution, understanding human health growth and disease treatment. The whole-genome variant calling process includes steps such as sequencing, alignment, variation detection and false positive site filtering. Among them, the sequencing read length, base recognition accuracy and alignment accuracy determine to a certain extent whether the variation can be correctly identified in a specific genomic background. Existing genomic variation detection schemes, whether based on traditional statistical models or emerging neural network-based variation detection tools, are designed to solve the background noise or errors caused by sequencing and alignment to a great extent, thereby improving the accuracy of variation detection.

[0028] In genomic regions where short reads (usually 150 to 250 base pairs) can be accurately aligned, the detection of gene variants using short-read sequencing platforms can already obtain relatively accurate detection results. However, short-read sequencing platforms still have great limitations in complex genomic regions (such as genomic repeat regions). Generally speaking, if the sequencing read length is shorter than the genomic repeat region, it is difficult to align the short read to its true position in the genome because it is impossible to determine which segment of the repeat it belongs to. Therefore, it will be difficult to obtain accurate variant information using short-read sequencing platforms to detect variants from such genomic repeat regions where short reads are difficult to align. Such repeat regions include segmental repeat regions, long interspersed nuclear repeat sequences, short tandem repeat sequences, variable number tandem repeat sequences, telomeres, and microsatellite repeat sequences (up to 30Mbp).

[0029] like Figure 1The figure shows an inaccurate short-read sequencing alignment leading to inaccurate gene variation detection. As shown in the figure, the short-read sequencing result XXXY can be aligned to the first segment 110, the second segment 120 and the third segment 130 in the reference gene sequence. If a gene variation is detected in a gene locus in the short read, it is impossible to determine which segment the variation locus is in the entire genome, resulting in inaccurate genome variation detection results.

[0030] The emergence of long-read sequencing technologies such as nanopore sequencing has overcome the limitation that short-read platforms cannot accurately align repetitive regions of the genome. Nanopore sequencing technology is a high-throughput sequencing method based on nanopores that uses electrochemical resistance in nanopores to capture and analyze DNA molecules. The core of this technology is that as single-stranded DNA passes through the nanopore, the DNA molecule can be captured and read through changes in current. The advantages of nanopore sequencing technology are its rapid, portable, real-time and direct sequencing and its powerful long read (LR) capabilities, making it an important sequencing technology for studying genome structure, genetic variation, disease diagnosis and other fields. However, the current sequencing accuracy of nanopore sequencing needs to be improved, and its performance in the homopolymer (DNA sequence composed of the same nucleotide units) and heteropolymer regions (DNA sequences composed of different nucleotide units) of the genome is still poor.

[0031] In summary, both the short-read sequencing platform and the long-read sequencing platform have certain defects that lead to low accuracy in detecting genomic variations.

[0032] Based on this, the embodiments of the present application provide a gene variation detection method, device, electronic device and storage medium, which improves the accuracy of genome variation detection by combining the advantages of long read sequencing technology with the advantages of long read sequencing technology. Below, the gene variation detection method provided in the embodiments of the present application is described.

[0033] Reference Figure 2 In some embodiments, the gene variation detection method provided in the embodiments of the present application includes but is not limited to steps S201 to S205.

[0034] Step S201, aligning the long-read sequencing data of the gene fragment to be tested to the reference genome to obtain first sequencing result comparison data, and aligning the short-read sequencing data of the gene fragment to be tested to the reference genome to obtain second sequencing result comparison data; Step S202, determining a first variation detection result corresponding to each gene site in the gene segment to be tested based on the first sequencing result comparison data, and determining a second variation detection result corresponding to each gene site in the gene segment to be tested based on the second sequencing result comparison data; Step S203, identifying a first target gene site for which the first variation detection result is different from the second variation detection result, and when the first target gene site is in a complex genomic region of the gene fragment to be tested, determining a third variation detection result of the first target gene site according to the second variation detection result and variation detection parameters of each gene site; Step S204, when the first target gene site is not in a complex region of the genome, determining a third variation detection result of the first target gene site according to the first variation detection result and the variation detection parameter; Step S205, determining a plurality of first variation sites where variations exist based on the third variation detection result, and determining a target gene variation detection result of the gene fragment to be tested based on the first variation sites and the second variation sites.

[0035] In the steps S201 to S205 shown in the embodiment of the present application, when the gene fragment to be tested is subjected to gene variation detection, the gene fragment to be tested can be subjected to long read sequencing and short read sequencing respectively, and the corresponding long read sequencing data and short read sequencing data are obtained respectively. Then, the sequencing data obtained by the two types of sequencing methods are aligned to the reference genome respectively to obtain long read sequencing alignment data and short read sequencing alignment data. Further, variation detection is performed based on the long read sequencing alignment data and the short read sequencing alignment data respectively to determine the long read variation detection result and the short read variation detection result. Since both the long-read variation detection results and the short-read variation detection results have their own defects, resulting in inaccurate variation detection results for some base sites, this scheme proposes that after obtaining the long-read variation detection results and the short-read variation detection results of the gene fragment to be tested respectively, the base sites that are judged as variations by both the long-read variation detection results and the short-read variation detection results are determined as accurate variation sites. For base sites where the long-read variation detection results and the short-read variation detection results are inconsistent, if the base site is in a complex region of the genome, the long-read variation detection result and the variation detection parameters of the base site are used to comprehensively determine whether it is a variation site; if the base site is not in a complex region of the genome, the short-read variation detection result and the variation detection parameters of the base site are used to comprehensively determine whether it is a variation site. In this way, in complex regions of the genome, the advantages of long-read sequencing can be used to avoid the problem of inaccurate variant detection results caused by the inability of short reads to determine the exact segment. In non-complex regions of the genome, the advantages of short-read sequencing can be used to avoid the problem of inaccurate variant detection results caused by the low accuracy of long-read sequencing results, thereby improving the accuracy of genetic testing for the genes to be tested.

[0036] In step S101 of some embodiments, the gene fragment to be tested can be any sampled gene fragment, such as a DNA fragment or an RNA fragment, or some gene fragment standards can be used to verify the effect of gene variation detection. For example, when the gene fragment to be tested is a DNA fragment, the gene variation detection method provided in the present application is used to perform gene variation detection on the DNA fragment, and the DNA fragment can be sequenced using nanopore sequencing technology and second-generation sequencing technology at the same time to obtain long read sequencing data and short read sequencing data of the DNA fragment. Among them, in some other embodiments, other sequencing technologies can also be used to measure the long read sequence and short read sequence of the DNA fragment. The long read sequencing data measured by nanopore sequencing and the short read sequencing data measured by the second-generation sequencing technology are only examples, and the specific sequencing means for the DNA fragment is not limited.

[0037] After obtaining the long-read sequencing data and short-read sequencing data of the DNA fragments, the long-read sequencing data and the short-read sequencing data can be further aligned to a preset reference genome using alignment software to obtain the corresponding alignment result data. The alignment result data can then be further formatted and converted into an alignment result file in a binary alignment (Binary Alignment / Map, BAM) format.

[0038] In step S202 of some embodiments, after the nanopore sequencing sequence and the second-generation sequencing sequence are aligned to the reference genome to obtain their respective corresponding alignment result files, the short-read sequencing platform and the long-read sequencing platform can be used to perform variation detection based on their respective corresponding alignment result files, and the variation detection results output by different sequencing platforms are obtained. Specifically, the long-read sequencing platform can output a first variation detection result, and the short-read sequencing platform can output a second variation detection result. Among them, in the disclosed embodiments, the variation detection result can specifically be whether each base site in the gene sequence is a variation site. When a certain base site is determined to be a variation site, the variation detection result can also include a specific type of variation. Variation types may include single nucleotide polymorphisms (SNP) and insertion / deletion (indel).

[0039] Among them, the more mature variant detection tool in the short read platform is the Genome Analysis Toolkit (GATK), which mainly uses Gaussian mixture models to recalibrate variant quality scores and Bayesian and other machine learning models to filter candidate variant sites. With the advancement of sequencing technology, a large amount of sequencing data and benchmarks that can be used for training and testing have been accumulated, covering millions of variants in a range of genomic environments, making it possible to use deep learning for small variant detection in the human genome. At present, the mainstream architecture of deep learning-based variant detection tools is convolutional neural network (CNN). Traditional machine learning-based variant detection tools use artificially designed sequence features (for example, whether it is a homologous multimer) and read features aligned to the reference genome (for example, whether the variant has chain bias). In order to reduce the need for feature engineering, the CNN architecture uses information from reads and the association of variant sites and their surrounding reference genome sequences. Through appropriate hyperparameter training, the CNN model can approximate a complex nonlinear function to classify potential variant sites into homologous variants, heterozygous variants, or non-variant sites. The representative tool for variant detection using the CNN framework is Deepvariant, which has achieved relatively excellent variant detection performance in both short-read and long-read sequencing platforms. In addition, some methods that use RNN combined with CNN frameworks for whole-genome small variant detection (such as clair3) have also achieved relatively excellent performance. This shows that deep learning-based methods have gradually replaced traditional statistical methods and become the mainstream method for genome variant detection.

[0040] Reference Figure 3 In some embodiments, step S202 includes but is not limited to steps S301 to S303.

[0041] Step S301, performing a quality assessment on the variant sites determined based on the first sequencing result comparison data based on a preset recurrent neural network model, and dividing the variant sites into first candidate variant sites with a quality value higher than a first threshold and second candidate variant sites with a quality value not higher than the first threshold according to the assessed quality values; Step S302, screening the second candidate variant site based on a preset convolutional neural network model to obtain a third candidate variant site determined as a variant site by the preset convolutional neural network model; Step S303, determining a first variation detection result corresponding to each gene site in the gene segment to be tested according to the first candidate variation site and the third candidate variation site.

[0042] In step S301 of some embodiments, the preset recurrent neural network model (RNN) can be specifically a 2-layer bidirectional long short-term memory network (bidirectional long short-term memory, biLSTM). The preset recurrent neural network model is used to perform quality assessment on the variant sites determined by the first sequencing result comparison data. Specifically, multiple potential variant sites in the gene sequence can be determined based on the first sequencing result comparison data, and then the corresponding RNN network input features can be obtained for these potential variant sites, and then the input features corresponding to each potential variant site are input into the RNN network, so that the genotype information and the variant quality value (mutation quality) corresponding to each potential variant site can be obtained, and then the potential variant sites corresponding to different quality values ​​can be divided based on a threshold, and specifically, the quality values ​​higher than the threshold can be divided into high-quality first candidate variant sites, and the quality values ​​not higher than the threshold can be divided into low-quality second candidate variant sites. The high-quality first candidate variant sites can be saved, and the low-quality second candidate variant sites can be further evaluated.

[0043] In step S302 of some embodiments, after RNN is used to divide the potential mutation sites into high-quality first candidate mutation sites and low-quality second candidate mutation sites according to mutation quality, a preset convolutional neural network model (CNN) can be further used to further screen the second candidate mutation sites determined to be low-quality. The CNN model can specifically be a CNN model including a residual network (ResNet). Specifically, for each second candidate mutation site, the mutation evaluation data corresponding to the candidate mutation site can be obtained first, and then the CNN model can be used to discriminate the mutation evaluation data to further filter the low-quality mutation sites in the second candidate mutation sites, and obtain the real mutation sites as the third candidate mutation sites.

[0044] The RNN proposed in step S301 and the CNN proposed in step S302 can both be models that have been pre-trained using sample data.

[0045] In step S203 of some embodiments, after two-level screening of potential variant sites using the RNN model and the CNN model, the respectively screened high-quality variant sites can be summarized as variant sites determined based on the first sequencing result comparison data, and then the first variant detection result corresponding to the first sequencing result comparison data can be further determined based on this.

[0046] Steps S301 to S303 use a deep learning method to perform gene variation detection on sequencing comparison data measured by different sequencing platforms, which can improve the detection efficiency and accuracy of gene variation detection based on sequencing comparison data. For the second sequencing comparison result, similar steps to steps S301 to S303 can also be used to perform variation detection to obtain a second variation detection result.

[0047] Please refer to Figure 4 In some embodiments, step S301 includes but is not limited to steps S401 to S403.

[0048] Step S401, determining a plurality of fourth candidate variant sites according to the first sequencing result comparison data; Step S402, obtaining sequencing base ratio information of the fourth candidate variant site and its associated sites, and generating a first evaluation feature for each fourth candidate variant site based on the sequencing base ratio information; Step S403, calling a preset recurrent neural network model to perform variation quality assessment on the first assessment feature, and determining a quality value corresponding to each fourth candidate variation site according to the assessment result.

[0049] Among them, in the disclosed embodiment, the sequencing base ratio information corresponding to each gene site in the whole genome of the DNA to be tested can be obtained first, wherein the sequencing base ratio information can be the ratio information of each base (A, G, C or T) in the multiple sequencing result data obtained by sequencing a certain site. For example, the site was sequenced 100 times, of which 20 times were A, 70 times were G, 5 times were C, and 5 times were T, then the sequencing base ratio information corresponding to the site is A: 20%, G: 70%, C: 5%, T: 5%. Generally speaking, if a site is a non-variable site, then the base composition of this site is single (that is, the proportion of a certain base in ATGC exceeds 90%), and the single base is consistent with the base of the site in the reference genome. Otherwise, if the base composition of the site is not single, or the base composition is single but inconsistent with the base of the site in the reference genome, then the site is determined as an abnormal site, or called the fourth candidate variant site.

[0050] Then, the sequencing base ratio information of each fourth candidate variant site and its associated sites can be obtained. Among them, the associated site can be specifically the site of the abnormal site and its flanking region, wherein the flanking region can be specifically the region including the abnormal site and its upstream and downstream regions, and the length of the flanking region here can be 128bp. After obtaining the sequencing base ratio information of each fourth candidate variant site and its associated sites, the sequencing base ratio information of the fourth candidate variant site and its associated sites can be summarized into a two-dimensional tensor as the first evaluation feature of the RNN model to evaluate the fourth candidate variant site.

[0051] Furthermore, the RNN model may be called to perform variation quality assessment on the first assessment feature corresponding to each fourth candidate variation site, thereby obtaining a quality value corresponding to each fourth candidate variation site.

[0052] Please refer to Figure 5 In some embodiments, step S301 may also include but is not limited to steps S501 to S503.

[0053] Step S501, determining a first threshold based on a preset ratio quantile value of quality values ​​corresponding to all fourth candidate variant sites; Step S502, determining the variation site whose quality value among the fourth candidate variation sites is higher than the first threshold as the first candidate variation site; Step S503: Determine the variation sites whose quality values ​​among the fourth candidate variation sites are not higher than the first threshold as the second candidate variation sites.

[0054] In the embodiment of the present disclosure, it is not necessary to determine the threshold for classifying the quality of gene variation in advance, because for different gene fragments to be tested, the threshold for classifying the quality of gene variation sites into high-quality variation sites and low-quality variation sites may be different. Therefore, in the embodiment of the present disclosure, after using RNN to evaluate and obtain the quality value corresponding to each fourth candidate variation site, the quality values ​​of all fourth candidate variation sites can be aggregated, and then the first threshold is determined based on the preset proportional quantile value of the aggregated quality value.

[0055] Specifically, for example, the 70th percentile of the quality value can be set as the first threshold, that is, the quality values ​​of the fourth candidate variant sites are sorted from high to low, and then the quality value sorted at the 70th percentile is determined as the first threshold. Further, the fourth candidate variant site with a quality value higher than the first threshold can be determined as the first candidate variant site, and then the fourth candidate variant site with a quality value lower than the fourth candidate variant site can be determined as the second candidate variant site.

[0056] Reference Figure 6 In some embodiments, step S302 includes but is not limited to steps S601 to S602.

[0057] Step S601, obtaining variation evaluation data in the vicinity of each second candidate variation site, and determining a second evaluation feature of each second candidate variation site according to the variation evaluation data; Step S602: Call a preset convolutional neural network model to perform variation identification on the second evaluation feature, and determine the second candidate variation site identified as a variation site by the preset convolutional neural network model as a third candidate variation site.

[0058] In steps S601 to S602 of some embodiments, after performing variation detection on the sequencing result comparison data based on RNN to determine the second candidate variation site with low quality, CNN can be further used to screen the second candidate variation sites with low quality to screen out the real variation sites with relatively high variation quality. In this process, variation evaluation data in the neighboring region of each second candidate variation site can be first obtained, and the second evaluation feature of each second candidate variation site can be determined based on the variation evaluation data. Among them, the neighboring region of the second candidate variation site can be understood as the flanking segment of the second candidate variation site, that is, the aforementioned second candidate variation site and the adjacent site on the read length where it is located; the variation evaluation data can include multiple feature data, and these feature data can support the evaluation of the variation quality of the second candidate variation site.

[0059] After obtaining the variation evaluation data corresponding to each second candidate variation site, the second evaluation feature of each second candidate variation site can be further generated based on the variation evaluation data. Specifically, the variation evaluation data can be embedded according to the input specification of the CNN model, so as to convert the variation evaluation data corresponding to the second candidate variation site into its corresponding feature vector, thereby obtaining the second evaluation feature corresponding to each second candidate variation site.

[0060] Furthermore, a preset CNN model can be called to process the second evaluation feature, thereby realizing variation identification of the second candidate variation site, and then the second candidate variation site identified as a variation site by the CNN model is determined as a third candidate variation site according to the variation identification result.

[0061] That is, in the embodiment of the present disclosure, the CNN model can be a binary classification model, the input of which is the second evaluation feature, and the input is yes or no, that is, the output is a variation detection conclusion for the second candidate variation site, which indicates that the second candidate variation site is a variation site or not a variation site.

[0062] Please refer to Figure 7 In some embodiments, step S601 includes but is not limited to steps S701 to S702.

[0063] Step S701, obtaining mutation assessment information, base quality information, and genotyping information corresponding to a preset number of base pairs upstream and downstream of the corresponding read length for each second candidate variant site; Step S702: Generate a second evaluation feature for each second candidate variant site based on mutation evaluation information, base quality information, and genotyping information corresponding to a preset number of base pairs.

[0064] In some embodiments, in step S701 to step S702, the neighboring region of the second candidate variation site may be a flanking segment within a preset sequence length range upstream and downstream of the read where the second candidate variation site is located. Specifically, the flanking segment here may be a base sequence 15 bp upstream and downstream of the second candidate variation site on the read. When the aforementioned CNN model is required to perform variation evaluation on the second candidate variation site, the mutation evaluation information, base quality information, and genotyping information contained in the flanking segment of the second candidate variation site may be obtained to generate a second evaluation feature of the second candidate variation site.

[0065] The mutation assessment information may include the mutation (variation) base sequence contained in the flanking segment, the mutation marker information, etc., and the assessment feature may also include the mutation marker, insertion sequence, and comparison quality information. In detail, the second assessment feature of the second candidate mutation site may be obtained in the following manner: First, for the second candidate variant site, its location on the chromosome and the chromosome segment containing 15bp upstream and downstream can be determined: A. Extract linear reference genome base sequence based on chromosome segment location B. At the same time, obtain the base quality of the read containing the target segment from fastq according to the positioning C. Count the base mutations of the reads in this segment D. Mark the target mutation (AC, AG, AT...) based on the statistics of reads in this segment E. Get the chain direction information from the bam containing the alignment information according to the read name F. Get the insertion sequence of the read in this segment from the bam containing the alignment information according to the read name G. Get the alignment quality of daily reads from the bam containing alignment information based on the read name H. Use typing tools to obtain typing information for each read.

[0066] Then, the acquired information can be embedded to obtain the corresponding feature vector, that is, the second evaluation feature. Then, the second evaluation feature can be input into the aforementioned CNN model for evaluation to determine whether the second candidate mutation site is a true mutation site.

[0067] In step S203 of some embodiments, after performing variation detection on the first sequencing result comparison data and the second sequencing result comparison data based on the above scheme to obtain the first variation detection result and the second variation detection result corresponding to each other, the embodiment of the present application can be used to fuse the first variation detection result and the second variation detection result. Specifically, when the results are fused, the first target gene site for which the first variation detection result and the second variation detection result are different can be distinguished first. That is, for sites with the same first variation detection result and the same second variation detection result, it can be determined that its variation detection result is accurate. At this time, when the variation detection result simultaneously indicates that a certain site is a variation site, it can be determined that it is a true variation site. Since the first variation detection result of the target gene site is different from the second variation detection result, that is, there is a result in the first variation detection result and the second variation detection result indicating that the target gene site is a variation site, at this time, it can be determined whether the first variation detection result or the second variation detection result is to be used according to whether the first target gene site is in a complex genomic region of the gene fragment to be tested.

[0068] Among them, complex genomic regions can specifically include regions where short reads are difficult to align, such as regions of segmental repeat sequences, tandem repeat sequences, and regions of variable number of tandem repeat sequences. In these regions, short read alignment is difficult to align to the exact region, resulting in inaccurate results for short read gene variation detection. Long read sequencing can avoid the above-mentioned problem of being unable to align to the exact region due to its advantage of longer read length, and thus can obtain more accurate variation detection results.

[0069] Therefore, when the first target gene site is in a complex genomic region of the gene fragment to be tested, the true variation detection result of the first target gene site can be determined mainly based on the long read variation detection result, that is, the second variation detection result, that is, the third variation detection result of the first target gene site can be determined.

[0070] Reference Figure 8 In some embodiments, step S203 includes but is not limited to steps S801 to S803.

[0071] Step S801, when the first target gene site is in a complex region of the genome and the second variation detection result corresponding to the first target gene site is a variation site, determining a third variation detection result of the first target gene site according to the second variation detection result; Step S802, when the first target gene site is in a complex region of the genome and the second variation detection result corresponding to the first target gene site is not a variation site, obtaining a first variation detection parameter and a first reference variation parameter range corresponding to the first target gene site; Step S803: determining a third variation detection result of the first target gene site according to the first variation detection parameter and the first reference variation parameter range.

[0072] In steps S801 to S803 of some embodiments, when it is detected that the first target gene site is in a complex region of the genome, it can be detected whether the second variation detection result corresponding to the first target gene site is used to determine that the first target gene site is a variation site. When the second variation detection result indicates that the first target gene site is a variation site, the third variation detection result of the first target gene site can be determined based on the second variation detection result, that is, the first target gene site can be determined to be a true variation site.

[0073] When the second variation detection result indicates that the first target gene site is not a variation site, the first variation detection parameter corresponding to the first target gene site and the corresponding first reference variation parameter range can be further obtained. Then, the third variation detection result of the first target gene site can be determined based on the first variation detection parameter and the first reference variation parameter range. Specifically, the first variation detection parameter can be compared with the first reference variation parameter range. When the first variation detection parameter is within the first reference variation parameter range, the first target gene site is determined to be a true variation site; conversely, if the first variation detection parameter is not within the first reference variation parameter range, the first target gene site is determined to be not a true variation site.

[0074] The first variation detection parameter may specifically include the following information of the first target gene locus: a. Mutation frequency (number of reads containing a specific mutation / total number of reads containing the site); b. Mutation type (single nucleotide base mutation, insertion and deletion); c. Mutation quality, according to the model output (can be used as an indicator to assess whether the mutation is reliable).

[0075] In step S204 of some embodiments, when it is detected that the first target gene site is not in a complex region of the genome, the actual variation detection result of the first target gene site can be determined mainly based on the short read variation detection result, considering that the accuracy of short read sequencing is higher than that of long read sequencing.

[0076] Reference Fig. 9 In some embodiments, step S204 includes but is not limited to steps S901 to S903.

[0077] Step S901, when the first target gene site is not located in a complex region of the genome, and the first variation detection result corresponding to the first target gene site is a variation site, determining a third variation detection result of the first target gene site according to the first variation detection result; Step S902, when the first target gene site is not in a complex region of the genome, and the first variation detection result corresponding to the first target gene site is not a variation site, obtaining a second variation detection parameter and a second reference variation parameter range corresponding to the first target gene site; Step S903, determining a third variation detection result of the first target gene site according to the second variation detection parameter and the second reference variation parameter range.

[0078] In steps S901 to S903 of some embodiments, when it is detected that the first target gene site is not in a complex region of the genome, a first variation detection result corresponding to the first target gene site may be further obtained. If the first variation detection result corresponding to the first target gene site indicates that the first target gene site is a variation site, a third variation detection result of the first target gene site may be determined based on the first variation detection result, that is, the first target gene site is determined to be a true variation site. If the first variation detection result corresponding to the first target gene site indicates that the first target gene site is not a variation site, a second variation detection parameter corresponding to the first target gene site and a second reference variation parameter range may be further obtained.

[0079] Then, the third variation detection result of the first target gene site can be determined according to the second variation detection parameter and the second reference variation parameter range. Specifically, when the second variation detection parameter is within the second reference variation parameter range, the first target gene site can be determined to be a true variation site; conversely, when the second variation detection parameter is not within the second reference variation parameter range, the first target gene site can be determined not to be a true variation site.

[0080] Among them, for different types of variation, for example, when the first target gene site is determined to be a SNP variation type, the variation detection result of the first target gene site can be determined by using the second variation detection result as the main and the first variation detection result as the auxiliary to determine the third variation detection result; when the first target gene site is determined to be an insertion / deletion (indel) variation type, the variation detection result of the first target gene site can be determined by using the first variation detection result as the main and the second variation detection result as the auxiliary to determine the third variation detection result.

[0081] Therefore, based on the existing sequencing platforms, in the genomic regions where short reads can accurately align, using short read data for variant detection can already obtain very accurate variant detection results. However, regardless of the base accuracy, short reads will hinder the detection of variants in more complex regions of the genome, such as large tandem repeats and highly homologous regions (such as segmental duplications) and highly variable medically relevant human leukocyte antigens. In addition, short-read sequencing platforms have blind spots in high GC regions, and cannot provide effective sequencing data for these regions, which brings difficulties to variant detection. The direct sequencing and long read characteristics of nanopore sequencing can well solve the challenges encountered by short-read platforms. The longer reads make the alignment of reads in complex regions of the genome more accurate. At the same time, nanopore sequencing is not affected by high GC regions in the genome due to the advantages of the sequencing principle. Compared with short-read sequencing platforms, nanopore sequencing platforms have better variant detection performance in these regions. However, the sequencing error rate of nanopore long-read sequencing still needs to be further optimized. In addition, the problem of small indels (insertions / deletions) that are prevalent in nanopore sequencing data still makes nanopore sequencing platforms challenging in variant detection. The variation detection method provided in the present application can combine the excellent performance of the long-read sequencing platform in complex regions of the genome and the sequencing accuracy advantage of the short-read sequencing platform to greatly improve the accuracy of genome variation detection.

[0082] like Fig.10 As shown, it is another flow chart of the gene variation detection method provided by the present application. As shown in the figure, the gene variation detection method provided by the present application first uses a long read sequencing platform and a short read sequencing platform to sequence the gene fragment to be tested, and obtains long read sequencing data and short read sequencing data, and then aligns the long read sequencing data and the short read sequencing data to the reference genome, and obtains long read sequence alignment data and short read sequence alignment data. Further, a deep learning method is used to perform variation detection on the long read sequence alignment data and the short read sequence alignment data, respectively, to obtain the corresponding long read variation detection data and short read variation detection data.

[0083] When both the long-read variation detection data and the short-read variation detection data of a gene locus indicate that the gene locus is a variation site, the variation site is determined to be a true variation site; when both the long-read variation detection data and the short-read variation detection data of a gene locus do not indicate that the gene locus is a variation site, it is possible to further distinguish whether the gene locus is in a complex genomic region. When the gene locus is in a complex genomic region, the long-read variation detection result can be used to determine whether the gene locus is a true variation site; when the gene locus is not in a complex genomic region, the short-read variation detection result can be used to determine whether the gene locus is a true variation site.

[0084] The present application also discloses the following specific embodiments of the gene detection method provided by the present application: 1) Nanopore long-read sequencing and next-generation sequencing were performed on the HG002 standard simultaneously to obtain long-read sequence data and corresponding short-read sequencing data. Then, the long-read sequences and short-read sequences were aligned to the GRch38 genome using the alignment software minimap2 (v 2.24) and bwa (0.7.17-r1188), respectively, to obtain the corresponding alignment result files in BAM format.

[0085] 2) Obtain the pileup statistics of the sequencing data from the BAM file containing the alignment information obtained above, and then use a model of an RNN framework containing a two-layer bidirectional long short-term memory network that has been trained with a standard data set to perform preliminary variation detection on the target sites in the long read sequence to obtain potential variation sites in the entire genome of the HG002 sample.

[0086] 3) For the mutation sites output by the RNN model, the mutation sites with quality scores in the top 70% are marked as high-quality mutation sites, and the remaining 30% are marked as low-quality mutation sites.

[0087] 4) The variant sites marked as low-quality and their corresponding sequences in the region of the genome flanking chain (i.e., the region containing the variant site and its upstream and downstream with a total length of 128bp), base quality, base mutation sequence contained in the target interval, markers of the target mutation, positive and negative chain markers, insertion sequence, alignment quality, and typing information are used as input, and a CNN neural network framework including a residual neural network model is used to further perform model screening on these low-quality sites to obtain potential real variant sites.

[0088] 5) Obtain the mutation detection results corresponding to the long-read and short-read sequencing data respectively, and then use the iterator to screen these mutation sites: if the mutation site and mutation characteristics exist in both data, they are written into the final vcf file. Conversely, if the mutation detection site only exists in one type of data, first determine whether the site exists in the complex region of the genome (the region that is not friendly to short reads, the long read advantage region obtained through prior analysis, such as Figure 3 , which is known and provided here). If yes, the variant sites identified by the long read data are mainly used, and the variant sites in the short read data are filtered based on quality and mutation frequency, and the high-quality variant sites are retained; if no, the variant sites identified by the short read data are mainly used, and the variant sites in the long read data are filtered based on quality and mutation frequency, and the high-quality variant sites in the long read are also retained. Thus, the final whole genome variation detection result is obtained.

[0089] The benchmarking tool hap.py (v0.3.15) was used to evaluate the performance of the results, and the evaluation results are shown in Table 1. From the analysis results, it can be seen that compared with the existing variation detection solutions, the method proposed in the present invention has significantly improved the detection of small variations in the whole genome, whether it is SNP or INDEL identification (F1-score>= 0.9).

[0090] The gene variation detection device provided in the embodiments of the present application is described below.

[0091] As shown in Table 1, it is a comparison table of the effects of different variant detection schemes. From the data in the table, it can be seen that the scheme combining long read length and short read length provided in this application can achieve better variant detection accuracy.

[0092] Table 1 Comparison of mutation detection effects

[0093] Reference Fig.11 In some embodiments, the present application also provides a gene variation detection device 1100, which includes: The comparison unit 1101 is used to compare the long-read sequencing data of the gene fragment to be tested to the reference genome to obtain first sequencing result comparison data, and to compare the short-read sequencing data of the gene fragment to be tested to the reference genome to obtain second sequencing result comparison data; A first determining unit 1102 is used to determine a first variation detection result corresponding to each gene site in the gene segment to be tested based on the first sequencing result comparison data, and to determine a second variation detection result corresponding to each gene site in the gene segment to be tested based on the second sequencing result comparison data; The identification unit 1103 is used to identify the first target gene site for which the first variation detection result is different from the second variation detection result, and when the first target gene site is in a complex genomic region of the gene fragment to be tested, determine the third variation detection result of the first target gene site according to the second variation detection result and the variation detection parameter of each gene site; A second determining unit 1104 is configured to determine a third variation detection result of the first target gene site according to the first variation detection result and the variation detection parameter when the first target gene site is not located in a complex region of the genome; The third determination unit 1105 is used to determine multiple first variation sites where variations exist based on the third variation detection result, and determine the variation detection result of the target gene of the gene fragment to be tested according to the first variation site and the second variation site, wherein the second variation site is a variation site that has an intersection in the first variation detection result and the second variation detection result.

[0094] In some embodiments, the identification unit comprises: a first determining subunit, configured to determine a third variation detection result of the first target gene site according to the second variation detection result when the first target gene site is in the complex region of the genome and the second variation detection result corresponding to the first target gene site is a variation site; a first acquisition subunit, configured to acquire, when the first target gene site is in the complex region of the genome and the second variation detection result corresponding to the first target gene site is not a variation site, a first variation detection parameter and a first reference variation parameter range corresponding to the first target gene site; The second determining subunit is used to determine a third variation detection result of the first target gene site according to the first variation detection parameter and the first reference variation parameter range.

[0095] Optionally, in some embodiments, the second determining subunit includes: A first determination module is configured to determine that the third variation detection result of the first target gene site is a variation site when the variation detection parameter is within the reference variation parameter range; The second determination module is used to determine that the third variation detection result of the first target gene site is not a variation site when the variation detection parameter is not within the reference variation parameter range.

[0096] Optionally, in some embodiments, the second determining unit includes: a third determination subunit, configured to determine a third variation detection result of the first target gene site according to the first variation detection result when the first target gene site is not located in the complex region of the genome and the first variation detection result corresponding to the first target gene site is a variation site; a second acquisition subunit, configured to acquire, when the first target gene site is not located in the complex region of the genome and the first variation detection result corresponding to the first target gene site is not a variation site, a second variation detection parameter and a second reference variation parameter range corresponding to the first target gene site; The fourth determining subunit is used to determine a third variation detection result of the first target gene site according to the second variation detection parameter and the second reference variation parameter range.

[0097] Optionally, in some embodiments, the first determining unit includes: an evaluation subunit, configured to perform a quality evaluation on the variant sites determined based on the first sequencing result comparison data based on a preset recurrent neural network model, and divide the variant sites into first candidate variant sites having a quality value higher than a first threshold and second candidate variant sites having a quality value not higher than the first threshold according to the quality values ​​obtained by the evaluation; a screening subunit, configured to screen the second candidate variant site based on a preset convolutional neural network model to obtain a third candidate variant site determined as a variant site by the preset convolutional neural network model; The fifth determination subunit is used to determine the first variation detection result corresponding to each gene site in the gene fragment to be tested according to the first candidate variation site and the third candidate variation site.

[0098] Optionally, in some embodiments, the evaluation subunit comprises: A third determination module, configured to determine a plurality of fourth candidate variation sites according to the first sequencing result comparison data; a first acquisition module, configured to acquire sequencing base ratio information of the fourth candidate variant site and its associated sites, and generate a first evaluation feature for each of the fourth candidate variant sites based on the sequencing base ratio information; The first calling module is used to call a preset recurrent neural network model to perform variation quality assessment on the first assessment feature, and determine the quality value corresponding to each of the fourth candidate variation sites according to the assessment result.

[0099] Optionally, in some embodiments, the evaluation subunit further includes: A fourth determination module, configured to determine a first threshold based on a preset ratio quantile value of the quality values ​​corresponding to all the fourth candidate variant sites; a fifth determination module, configured to determine, among the fourth candidate variation sites, a variation site with a quality value higher than the first threshold as a first candidate variation site; The sixth determination module is configured to determine the variation sites among the fourth candidate variation sites whose quality values ​​are not higher than the first threshold as second candidate variation sites.

[0100] Optionally, in some embodiments, the screening subunit comprises: A second acquisition module, configured to acquire variation evaluation data in a neighboring region of each second candidate variation site, and determine a second evaluation feature of each second candidate variation site according to the variation evaluation data; The second calling module is used to call the preset convolutional neural network model to perform variation identification on the second evaluation feature, and determine the second candidate variation site identified as a variation site by the preset convolutional neural network model as a third candidate variation site.

[0101] Optionally, in some embodiments, the second acquisition module includes: An acquisition submodule, used to obtain mutation assessment information, base quality information, and genotyping information corresponding to a preset number of base pairs upstream and downstream of the corresponding read length for each second candidate variant site; A generating submodule is used to generate a second evaluation feature for each second candidate variant site based on the mutation evaluation information, base quality information and genotyping information corresponding to the preset number of base pairs.

[0102] Reference Fig.12 , Fig.12 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes: The processor 1201 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application; The memory 1202 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1202 can store an operating system and other application programs. When the technical solution provided in the embodiments of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1202, and the processor 1201 is called to execute the gene variation detection method of the embodiment of the present application; Input / output interface 1203, used to implement information input and output; The communication interface 1204 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.); A bus 1205 that transmits information between various components of the device (e.g., the processor 1201, the memory 1202, the input / output interface 1203, and the communication interface 1204); The processor 1201 , the memory 1202 , the input / output interface 1203 and the communication interface 1204 are connected to each other in communication within the device via the bus 1205 .

[0103] The embodiment of the present application further provides a computer program product, which includes a computer program. A processor of a computer device reads and executes the computer program, so that the computer device executes and implements the above-mentioned gene variation detection method.

[0104] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "comprises" and "comprising" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0105] It should be understood that in the present disclosure, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0106] It should be understood that in the description of the embodiments of the present application, the meaning of multiple (or multiple items) is more than two, greater than, less than, exceed, etc. are understood to not include the number, and above, below, within, etc. are understood to include the number.

[0107] In the several embodiments provided in the present disclosure, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0108] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0109] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0110] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present disclosure is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store program codes.

[0111] It should also be understood that the various implementation methods provided in the embodiments of the present application can be combined arbitrarily to achieve different technical effects.

[0112] The above is a specific description of the implementation methods of the present disclosure, but the present disclosure is not limited to the above implementation methods. Technical personnel familiar with the art can also make various equivalent modifications or substitutions without violating the spirit of the present disclosure. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present disclosure.

Claims

1. A method for detecting gene variation, characterized in that: The method comprises: Comparing the long-read sequencing data of the gene fragment to be tested to the reference genome to obtain first sequencing result comparison data, and comparing the short-read sequencing data of the gene fragment to be tested to the reference genome to obtain second sequencing result comparison data; Determine a first variation detection result corresponding to each gene site in the gene segment to be tested based on the first sequencing result comparison data, and determine a second variation detection result corresponding to each gene site in the gene segment to be tested based on the second sequencing result comparison data; Identify a first target gene site for which the first variation detection result is different from the second variation detection result, and when the first target gene site is in a complex genomic region of the gene fragment to be tested, determine a third variation detection result of the first target gene site according to the second variation detection result and variation detection parameters of each gene site; When the first target gene site is not in the complex region of the genome, determining a third variation detection result of the first target gene site according to the first variation detection result and the variation detection parameter; Based on the third variation detection result, multiple first variation sites where variations exist are determined, and according to the first variation sites and the second variation sites, a variation detection result of a target gene of the gene fragment to be tested is determined, wherein the second variation site is a variation site where there is an intersection between the first variation detection result and the second variation detection result.

2. The method according to claim 1, characterized in that When the first target gene site is located in the complex genomic region of the gene fragment to be tested, determining the third variation detection result of the first target gene site according to the second variation detection result and the variation detection parameter of each gene site includes: When the first target gene site is in the complex region of the genome, and the second variation detection result corresponding to the first target gene site is a variation site, determining a third variation detection result of the first target gene site according to the second variation detection result; When the first target gene site is in the complex region of the genome, and the second variation detection result corresponding to the first target gene site is a non-variable site, obtaining a first variation detection parameter and a first reference variation parameter range corresponding to the first target gene site; A third variation detection result of the first target gene site is determined according to the first variation detection parameter and the first reference variation parameter range.

3. The method according to claim 2, characterized in that The determining the third variation detection result of the first target gene site according to the first variation detection parameter and the first reference variation parameter range includes: When the variation detection parameter is within the reference variation parameter range, determining the third variation detection result of the first target gene site as a variation site; When the variation detection parameter is not within the reference variation parameter range, the third variation detection result of the first target gene site is determined to be a non-variation site.

4. The method according to any one of claims 1 to 3, characterized in that When the first target gene site is not located in the complex region of the genome, determining a third variation detection result of the first target gene site according to the first variation detection result and the variation detection parameter includes: When the first target gene site is not located in the complex region of the genome, and the first variation detection result corresponding to the first target gene site is a variation site, determining a third variation detection result of the first target gene site according to the first variation detection result; When the first target gene site is not in the complex region of the genome, and the first variation detection result corresponding to the first target gene site is not a variation site, obtaining a second variation detection parameter and a second reference variation parameter range corresponding to the first target gene site; A third variation detection result of the first target gene site is determined according to the second variation detection parameter and the second reference variation parameter range.

5. The method according to claim 1, characterized in that The determining of the first variation detection result corresponding to each gene site in the gene segment to be tested based on the first sequencing result comparison data includes: Performing a quality assessment on the variant sites determined based on the first sequencing result comparison data based on a first preset neural network model, and dividing the variant sites into first candidate variant sites with a quality value higher than a first threshold and second candidate variant sites with a quality value not higher than the first threshold according to the quality values ​​obtained by the assessment; Screening the second candidate variant sites based on the second preset neural network model, and removing the second candidate variant sites determined by the model as non-variable; The first variation detection result corresponding to each gene site in the gene fragment to be tested is determined according to the first candidate variation site and the screened second candidate variation site.

6. The method according to claim 5, characterized in that The method of performing quality assessment on the variant sites determined based on the first sequencing result comparison data based on a preset recurrent neural network model includes: Determine a plurality of fourth candidate variation sites based on the comparison data of the first sequencing result; Obtaining sequencing base ratio information of the fourth candidate variant site and its associated sites, and generating a first evaluation feature for each of the fourth candidate variant sites based on the sequencing base ratio information; The first preset neural network model is called to perform variation quality assessment on the first assessment feature, and a quality value corresponding to each of the fourth candidate variation sites is determined according to the assessment result.

7. The method according to claim 6, characterized in that Also includes: The first threshold is determined based on a preset proportional quantile value of the quality values ​​corresponding to all the fourth candidate variation sites.

8. The method according to claim 5, characterized in that The step of screening the second candidate variant site based on a preset convolutional neural network model to obtain a third candidate variant site determined as a variant site by the preset convolutional neural network model includes: For each of the second candidate variation sites, obtaining variation assessment data within a preset assessment region of each of the second candidate variation sites, and determining a second assessment feature of each of the second candidate variation sites according to the variation assessment data; wherein the preset assessment region includes the second candidate variation site and its upstream and downstream regions; The second preset neural network model is called to perform variation identification on the second evaluation feature, and the second candidate variation site determined by the model as a variation site is determined as a third candidate variation site.

9. The method according to claim 8, characterized in that The second evaluation feature includes mutation evaluation information, base quality information and genotyping information.

10. A gene mutation detection device, characterized in that: The gene variation detection device comprises: A comparison unit, used to compare the long-read sequencing data of the gene fragment to be tested to the reference genome to obtain first sequencing result comparison data, and to compare the short-read sequencing data of the gene fragment to be tested to the reference genome to obtain second sequencing result comparison data; a first determining unit, configured to determine a first variation detection result corresponding to each gene site in the gene segment to be tested based on the first sequencing result comparison data, and to determine a second variation detection result corresponding to each gene site in the gene segment to be tested based on the second sequencing result comparison data; an identification unit, configured to identify a first target gene site for which a first variation detection result is different from a second variation detection result, and when the first target gene site is located in a complex genomic region of the gene fragment to be tested, determine a third variation detection result of the first target gene site according to the second variation detection result and a variation detection parameter of each gene site; a second determining unit, configured to determine, when the first target gene site is not located in the complex region of the genome, a third variation detection result of the first target gene site according to the first variation detection result and the variation detection parameter; A third determination unit is used to determine multiple first variation sites where variations exist based on the third variation detection result, and to determine the target gene variation detection result of the gene fragment to be tested according to the first variation site and the second variation site, wherein the second variation site is a variation site where there is an intersection between the first variation detection result and the second variation detection result.

11. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the gene variation detection method according to any one of claims 1 to 9 is implemented.

12. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the gene variation detection method according to any one of claims 1 to 9 is implemented.

13. A computer program product, comprising a computer program, wherein the computer program is read and executed by a processor of a computer device, so that the computer device executes the gene variation detection method according to any one of claims 1 to 9.

Images