A method and device for detecting the distribution of genome structural variation
Through the segmented constant fitting PCF algorithm and a combination detection device, the problem that genomic variant distribution detection is difficult to distinguish overlapping events in traditional methods is solved, efficient detection and visualization of genomic structural variant distribution is achieved, and the genome mutation hot spots are found.
Patent Information
- Application Number
- CN202111405165.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-11-24
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-11-24
AI Technical Summary
When traditionally visualizing the distribution of variants along the genome through rainfall maps, the size of the standard map is limited and it is impossible to distinguish overlapping events. Especially when multiple data sets are drawn in the same map, it will cause data congestion and it is difficult to effectively detect genome structural mutations. distribution.
The genome is segmented using the PCF algorithm to segment the genome, calculate the distance between adjacent variants, and detect and visualize the distribution of genome structural variants through a combination device of input module, calculation module, genome segmentation module and visualization module.
It provides convenient and efficient detection methods for genome structural variant distribution, which can effectively distinguish and visualize variant distribution, help find genome mutation hot spots, such as ‘mutant shower’ or ‘mutant fog’, suitable for VCF or MAF files of any genome.
Smart Images

Figure CN114141308B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of genome data analysis, and in particular to a genome structural variation distribution detection method and a detection device. Background Art
[0002] Structural variation in human genome sequences is the insertion, deletion or rearrangement of DNA sequence fragments, ranging in length from about 1000 to millions of base pairs. In the past few years, structural variation has been shown to be much more common in the human genome than previously thought, and it is not randomly distributed in the genome. The study of structural variation is of great significance for the study of genome evolution, population polymorphism analysis, and disease susceptibility. With the development of second-generation high-throughput sequencing technology, the map of structural variation on the human genome has been truly comprehensively and intensively studied. The traditional method of visualizing the distribution of variation along the genome through rainfall plots may not be able to distinguish overlapping events due to the limited size of standard plots, especially when multiple data sets are plotted in the same plot, which will lead to data congestion. Summary of the invention
[0003] In order to solve the deficiencies mentioned in the above background technology, the purpose of the present invention is to provide a method and a detection device for detecting the distribution of genomic structural variation, which can read high-throughput sequencing data and use the piecewise constant fitting PCF algorithm to detect the distribution of variation on chromosomes.
[0004] The purpose of the present invention can be achieved through the following technical solutions:
[0005] A method for detecting the distribution of genomic structural variations comprises the following steps:
[0006] S1. Acquire and filter genome sequencing data;
[0007] S2, calculate the distance between adjacent mutations;
[0008] S3, segmenting the genome using piecewise constant fitting PCF algorithm;
[0009] S4. Visualize the distribution of variants along the genome.
[0010] A device for detecting distribution of genome structural variation, comprising an input module, a calculation module, a genome segmentation module, and a visualization module;
[0011] Input module, which includes two file reading units, including VCF unit and MAF unit;
[0012] The calculation module sorts the variants according to the genome coordinates and calculates the distance between adjacent variants, and outputs new variant coordinates;
[0013] The genome segmentation module uses the piecewise constant fitting PCF algorithm to segment the genome and output the location of the segmented fragments and the number of mutations they contain;
[0014] A visualization module that displays the distribution of variants along the genome.
[0015] Beneficial effects of the present invention:
[0016] The present invention provides a convenient and efficient detection device for the distribution of genomic structural variation, and for the first time uses a piecewise constant fitting (PCF) algorithm to detect the distribution of genomic structural variation, which helps to find mutation hotspots in the genome, such as "mutation showers" or "mutation fog". The detection device and method for the distribution of genomic structural variation provided by the present invention are applicable to VCF or MAF files of any genome. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The present invention will be further described below in conjunction with the accompanying drawings.
[0018] Figure 1 Flow chart of the method for detecting the distribution of genomic structural variation in the present invention
[0019] Figure 2 The structure diagram of the detection device for genome structural variation distribution in the present invention is
[0020] Figure 3 It is a distribution diagram of all variations in the whole genome in a specific embodiment. DETAILED DESCRIPTION
[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0022] Example 1
[0023] like Figure 1 As shown, in this embodiment, a method for detecting the distribution of genomic structural variation is proposed, and the method includes the following steps: S1, obtaining and filtering genomic sequencing data; S2, calculating the distance between adjacent mutations; S3, using a piecewise constant fitting (PCF) algorithm to segment the genome; S4, visualizing the distribution of variations along the genome.
[0024] In some specific embodiments, S1, the specific steps of obtaining and filtering genome sequencing data are as follows: two formats of cancer genome sequencing files (VCF and MAF) can be obtained, including information of variants '#CHROM', 'POS', 'REF', 'ALT', and 'FILTER'; the variants are filtered according to the 'FILTER' column in the file, and the variant data corresponding to the 'FILTER' column as "PASS" are extracted.
[0025] In some specific embodiments, the specific steps in S2, calculating the distance between adjacent variations, are as follows:
[0026] S201, sorting the extracted variant data in ascending order according to the genome coordinates '#CHROM' and 'POS';
[0027] S202, calculating the distance between adjacent mutations in the variation data according to the genome coordinates, the calculation method is:
[0028] p i = l i-1 +(l i -l i-1 ) / twenty one)
[0029] Among them, l i For each variant position on the chromosome, p i is the midpoint position between adjacent variations,
[0030] Then the distance between each observed adjacent variation is d i ,
[0031] d i = l i -l i-1 (2)
[0032] In some specific embodiments, S3, the specific steps of segmenting the genome using the piecewise constant fitting (PCF) algorithm are as follows:
[0033] p in step S202 i It will be used to mark the “jump point” / “break point” where the distance between variants changes on the genome, and d i It mainly consists of two parts
[0034] d i =c i +ε i (3)
[0035] c iis the distance between possible true variations at position i, ε i is the measurement noise, if c i ≠c i+1 Then it is considered that a breakpoint occurs between i and i+1. m , if the breakpoint set S = {I1 = {c1,...c m},...I j} represents a segment on a chromosome, such as I1 represents the distance between variants in the first segment, and to fit the model, minimize the penalty least squares
[0036]
[0037] Where |S| represents the number of segments in S, and γ is a constant greater than zero that controls the balance between good fit and the number of level shifts. If the best fit value for a given segment I is the average of the observed variance distances for that segment Then (4) can be written as:
[0038]
[0039] Where n I is the number of mutations contained in the segment I, and the first term in (5) does not depend on the segment, which can be simplified to the cost function:
[0040]
[0041] In order to obtain acceptable optimization efficiency and computational complexity, dynamic programming can be used to optimize (6). Assuming that the optimal segmentation of the distance from the first to the k-1th mutation is known, if 1...r segments and their corresponding errors e r It is known that r ≦ k, then the segment-related cost term can be simplified as:
[0042]
[0043] The overall error can be written as:
[0044]
[0045] γ is the penalty for the breakpoint, e0=0, and all parameters are obtained by simple recursion.
[0046] In some specific embodiments, according to any one of steps S1 to S3, a method for detecting the distribution of genomic structural variations based on high-throughput sequencing technology can be used to segment the genome, generate a segmented segment set S, and the number of variations n contained in each segment I. I and the average distance between variants
[0047] Example 2
[0048] like Figure 2 As shown, in this embodiment, a detection device for the distribution of genomic structural variations based on high-throughput sequencing technology is proposed, which is characterized by having an input module, a calculation module, a genome segmentation module, and a visualization module;
[0049] A: Input module, which contains two file reading units including VCF (Variant Call Format) unit and MAF (Mutation Annotation Format) unit;
[0050] B: Calculation module, which sorts the variants according to the genome coordinates and calculates the distance between adjacent variants, and outputs new variant coordinates;
[0051] C: Genome segmentation module, which uses the piecewise constant fitting (PCF) algorithm to segment the genome and output the location of the segmented fragments and the number of mutations contained;
[0052] D: Visualization module, which shows the distribution of variants along the genome.
[0053] In some specific embodiments, the input module in A includes:
[0054] Considering the computer memory capacity and the usual size of VCF / MAF files, this module only accepts files no larger than 2GB;
[0055] VCF unit: reads the VCF file format and returns a data set dat containing all variants that pass the filter, as shown in Table 1;
[0056] Table 1 VCF and MAF format file description
[0057]
[0058] MAF unit: reads the MAF file format, selects the "split" parameter to split multiple samples. When set to "TRUE", it splits by sample ID and returns a matrix list; when set to "FALSE", it does not split the matrix and returns a data set dat, whose structure is exactly the same as the data set returned by the VCF unit;
[0059] Data filtering, filter variants according to the 'FILTER' column. If it is "PASS", it means that this site can be considered as a variant. If it is other strings such as "q10; s50", the module will issue a warning and ask the user to decide whether to stop and filter or continue the analysis.
[0060] In some specific embodiments, the calculation module in B includes: an output data set dist, as shown in Table 2.
[0061] Table 2 Description of the dataset dist
[0062]
[0063]
[0064] In some specific embodiments, the genome segmentation module in C includes:
[0065] Set the parameter kmin, the minimum number of mutations in each segment, the default is 2;
[0066] Set the parameter gamma, the penalty for each discontinuity in the curve, the default value is 25;
[0067] Set the parameter len, the average distance between variants in the fragments;
[0068] Set the parameter nmut, the number of mutations contained in the fragment;
[0069] Set the parameter assembly, the reference assembly version used for analysis, the default is human hg19, and supported assemblies include hg38, hg16-19, and mm7-10;
[0070] The output data set seg includes the segment's "chr", "arm", "start.pos", "end.pos", "n.mut", and "dist.mean" information, as shown in Table 3.
[0071] Table 3 Dataset seg description
[0072]
[0073] In some specific embodiments, the visualization module in D includes: displaying a rainfall map of the distribution of variations along the genome based on information of the data sets dat, dist, and seg;
[0074] In some specific embodiments, "mutation showers" and "mutation fog" are two mutation hotspots that have been recently discovered, which seriously damage genetic material and may cause cancer. According to the digital information output by the genome segmentation module and combined with the image information output by the visualization module, it helps to find these two mutation hotspots. The VCF file from mouse pancreatic cancer GSM5241877_KPC (https: / / www.ncbi.nlm.nih.gov / geo / query / acc.cgi?acc=GSM5241877) downloaded from GEO, filters the mutations according to the 'FILTER' column in the file. If it is "PASS", it means that this site can be considered as a mutation. Set the parameters (kim=2, gamma=25, assembly=mm10) to segment the whole genome, and output a rainfall map of the distribution of all mutations in the entire genome along the genome, such as Figure 3 As shown in the figure, a particularly dense vertical line may be the location where the mutation hotspot occurs. Here we set the parameters len=1000 and nmut=2 to filter out the regions that meet the "mutation shower" and "mutation fog". For example, in Table 4, "mutation fog" is a fragment with 2 to 4 mutations within 1000bp, and "mutation shower" is a fragment with more than 5 mutations within 1000bp.
[0075] Table 4 Segmentation information
[0076]
[0077] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.
[0078] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.
Claims
1. A method for detecting the distribution of genomic structural variations, It is characterized in that The following steps are involved: S1. Acquire and filter genome sequencing data; S2, calculate the distance between adjacent mutations; S3, segmenting the genome using piecewise constant fitting PCF algorithm; S4, Visualizing the distribution of variants along the genome; The step S3 uses the piecewise constant fitting PCF algorithm to segment the genome to generate a segment set S. The number of variants contained in each segment I is and the average distance between variants , as follows: In step S202 will be used to mark jump points or breakpoints where the distance between variants changes on the genome, while It consists of two parts: (3), is the distance between possible true variations at position i, To measure noise, if It is considered that a breakpoint occurs between i and i+1; For the sequence ,..., , if the breakpoint set Represents a segment on a chromosome. Represents the distance between variants in the first segment, minimizing penalized least squares for fitting the model: (4) Where |S| represents the number of segments in S, and γ is a constant greater than zero, which is used to control the balance between good fit and the number of horizontal shifts; If the best fit value for a given segment I is the average of the observed variation distances for that segment , then formula (4) can be written as: (5) Where nI is the number of mutations contained in the segment I, and the first term in equation (5) does not depend on the segment, it can be simplified to the cost function: (6) In order to obtain acceptable optimization efficiency and computational complexity, the dynamic programming method can be used to optimize equation (6). Assuming that the optimal segmentation of the distance from the first to the k-1th mutation is known, if 1...r segments and their corresponding errors It is known that r ≦ k, then the segment-related cost term can be simplified as: (7) The overall error can be written as: (8) γ is the penalty for the breakpoint, e0=0, and all parameters are obtained by simple recursion.
2. The method for detecting genomic structural variation distribution according to claim 1, It is characterized in that The genome sequencing files that can be obtained in step S1 include VCF and MAF format files. The variation is filtered according to the FILTER column in the file, and the variation data corresponding to the PASS in the FILTER column is extracted.
3. The method for detecting genomic structural variation distribution according to claim 1, It is characterized in that The specific steps of calculating the distance between adjacent variations in step S2 are as follows: S201, sorting the extracted variant data in ascending order according to the genome coordinates #CHROM and POS; S202, calculating the distance between adjacent mutations in the mutation data according to the genome coordinates, the calculation method is: (1), in, For each variant's position on the chromosome, is the midpoint position between adjacent variations, The distance between each observed adjacent variation is : (2)。 4. A detection device based on the method for detecting the distribution of genomic structural variation according to any one of claims 1 to 3, It is characterized in that It includes input module, calculation module, genome segmentation module and visualization module; The input module comprises two file reading units, including a VCF unit and a MAF unit; The calculation module sorts the variants according to the genome coordinates and calculates the distances between adjacent variants, and outputs new variant coordinates; The genome segmentation module segments the genome using the piecewise constant fitting PCF algorithm, and outputs the location of the segmented fragments and the number of mutations contained therein; The visualization module displays the distribution of variations along the genome.
Citation Information
Patent Citations
Method for detecting gene mutation detection results with relatively high credibility
CN108256291A
Gene sequencing data simulation system and method for simulating crowd background information
CN110491441A
Method for detecting genomic structure variation based on nanopore sequencing
CN110600078A
