Bioinformatics analysis acceleration method
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TIANJIN MEDICAL LAB BGI
- Filing Date
- 2024-12-05
- Publication Date
- 2026-06-05
Smart Images
Figure CN122157802A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of bioinformatics, specifically to a method for accelerating bioinformatics analysis, a bioinformatics analysis device, a computer storage medium, and an electronic device. Background Technology
[0002] Bioinformatics analysis is now widely used to gain a deeper understanding of the molecular mechanisms underlying tumor biology and its development; to assist clinicians in more accurate tumor diagnosis and classification by analyzing the molecular characteristics of tumor cells; to provide patients with personalized treatment plans, including targeted therapy and immunotherapy; and to conduct prognostic assessments and drug resistance studies. Tumor information analysis relies on the massive amounts of data generated by high-throughput technologies, resulting in large data volumes and long analysis times. For patients with intermediate or advanced-stage tumors, providing tumor information analysis results as quickly as possible to formulate treatment plans is particularly important.
[0003] Detecting low-frequency mutations in tumors is crucial for clinical diagnosis, treatment, and prognostic assessment. Introducing unique molecular tags (UMIs) to correct for systematic errors in sequencing and improve the detection sensitivity of low-frequency mutations is a commonly used method. This method can reduce the probability of false positive mutation detection and improve mutation detection sensitivity. However, there is currently a lack of workflow acceleration algorithms specifically for tumor data with UMI tags.
[0004] Therefore, there is a need for a method to accelerate bioinformatics analysis of tumor data with UMI tags. Summary of the Invention
[0005] The present invention aims to at least partially solve one of the technical problems in the related art.
[0006] To this end, an embodiment of the first aspect of the present invention proposes a method for accelerating bioinformatics analysis, comprising: providing raw sequencing data of one or more samples, wherein the raw sequencing data includes a first sequencing file derived from one or more sequencing libraries and bearing one or more UMI tags; processing the first sequencing file into a second data file such that the second data file has a first classification of non-repeating UMI tags; splitting the second data file into a plurality of third data files based on the second classification of the UMI tags; providing the plurality of third data files to a bioinformatics analysis component to obtain a plurality of fourth data files; and processing the fourth data files into a fifth data file such that the fifth data file has a first classification of non-repeating UMI tags.
[0007] In some embodiments, processing the first sequencing file into a second data file includes: merging multiple first sequencing files that originate from the same sequencing library and have the same UMI tag of the same first classification into a second data file; or directly providing the first sequencing file as the second data file based on the absence of multiple first sequencing files that originate from the same sequencing library and have the same UMI tag of the same first classification.
[0008] In some embodiments, the resolution of the second category of the UMI tag is A. Preferably, the resolution A is 2*2 or 16*16. More preferably, the resolution is 2*2.
[0009] In some embodiments, splitting the second data file into multiple third data files includes: splitting a single second data file into A third data files, wherein the first category of the UMI tags in the A third data files is the same, and the second category is different.
[0010] In some embodiments, the analysis component is a Sentieon component, a Megabolt component, and a LUSH component, preferably, the analysis component is a LUSH component.
[0011] In some embodiments, the method further includes removing suboptimal alignments from the fifth data file to obtain a sixth data file.
[0012] In some embodiments, the method further includes merging the sixth data file derived from the same sequencing library into a seventh data file.
[0013] In some embodiments, the method further includes merging the seventh data file in the same sample into an eighth data file.
[0014] In some embodiments, the first sequencing file, the second data file, and the third data file are at least one of a fastq file and a bam file, and the fourth data file, the fifth data file, the sixth data file, the seventh data file, and the eighth data file are at least one of a sam file, a bam file, and FASTA. The raw sequencing data includes any one of genome sequencing data, transcriptome sequencing data, proteomics data, metabolomics data, single-cell omics data, and spatiotemporal omics data.
[0015] A second aspect of the present invention provides a bioinformatics analysis apparatus, comprising: a data providing module for providing raw sequencing data of one or more samples, wherein the raw sequencing data includes a first sequencing file with one or more UMI tags derived from one or more sequencing libraries; a first processing module for processing the first sequencing file into a second data file such that the second data file has a first classification of non-repeating UMI tags; a second processing module for splitting the second data file into a plurality of third data files based on the second classification of the UMI tags; an analysis module for providing the plurality of third data files to a bioinformatics analysis component to obtain a plurality of fourth data files; and a third processing module for processing the fourth data files into a fifth data file such that the fifth data file has a first classification of non-repeating UMI tags.
[0016] A third aspect of the present invention provides a computer storage medium storing a computer program, the computer program including program instructions that, when executed by a processor, implement the method described in any embodiment of the first aspect of the present invention.
[0017] An embodiment of the fourth aspect of the present invention provides an electronic device, the electronic device including a memory and a processor connected together, the memory being used to store a computer program, and the processor being used to invoke the computer program to implement the method described in any embodiment of the first aspect of the present invention.
[0018] The advantages and technical effects brought about by the independent claims according to the embodiments of the present invention are as follows:
[0019] (1) This invention targets data with UMI tags and processes it through UMI-based classification, so that each group of data after processing carries the same UMI sequence information. This information can be directly provided to the biological analysis component for analysis, thereby achieving the effect of significantly shortening the tumor information analysis time.
[0020] (2) This invention only improves the steps (algorithm) of the acceleration method, which can accelerate the analysis process currently in use, and greatly reduces the cost compared with solutions that accelerate based on hardware resources.
[0021] (3) The present invention can effectively improve the analysis speed of tumor data with UMI tags. In samples with a sequencing depth of 3000X, it can reduce the analysis time from the original 8 hours to 2.2 hours and reduce CPU usage by 63.8%. Furthermore, while the analysis time is shortened, the consistency of tumor mutation detection results is above 95%, ensuring the consistency of mutation detection. Attached Figure Description
[0022] Figure 1 This is a schematic flowchart of the method for accelerating bioinformatics analysis according to an embodiment of the present invention.
[0023] Figure 2 The data structure of raw sequencing data in one embodiment of the present invention is shown.
[0024] Figure 3 This is a schematic diagram of the sequence composition of the UMI tag in an embodiment of the present invention.
[0025] Figure 4 This is a schematic diagram of a list of third data files in one embodiment of the present invention.
[0026] Figure 5 This is a schematic diagram of the analysis process of the LUSH component in one embodiment of the present invention.
[0027] Figure 6 This is a schematic diagram of a list of fourth data files in one embodiment of the present invention. Detailed Implementation
[0028] The present invention will now be described in further detail with reference to specific embodiments. The embodiments given are merely illustrative of the invention and are not intended to limit its scope. The embodiments provided below can serve as a guide for further improvements by those skilled in the art and do not constitute a limitation on the invention in any way.
[0029] This application is based on the inventor's following understanding:
[0030] Bioinformatics analysis is now widely used to gain a deeper understanding of the molecular mechanisms underlying tumor biology and its development; to assist clinicians in more accurate tumor diagnosis and classification by analyzing the molecular characteristics of tumor cells; to provide patients with personalized treatment plans, including targeted therapy and immunotherapy; and to conduct prognostic assessments and drug resistance studies. Tumor information analysis relies on the massive amounts of data generated by high-throughput technologies, resulting in large data volumes and long analysis times. For patients with intermediate or advanced-stage tumors, providing tumor information analysis results as quickly as possible to formulate treatment plans is particularly important.
[0031] Detecting low-frequency mutations in tumors is crucial for clinical diagnosis, treatment, and prognostic assessment. Introducing unique molecular tags (UMIs) to correct systematic errors in sequencing and improve the detection sensitivity of low-frequency mutations is a commonly used method. This reduces the probability of false positives and increases mutation detection sensitivity. The principle of UMI technology is to add a unique short tag sequence to each original DNA molecule before PCR amplification. After library construction and sequencing, the original DNA can be traced back based on the tag sequence and alignment position. By comparing the consensus sequence of multiple sequences from the same DNA molecule, it is possible to distinguish whether the mutation is real or an introduced random error. UMI tags come in different resolutions, with 2x2 and 16x16 being commonly used. Different resolutions of UMI tags can significantly improve the sensitivity of tumor detection, providing important information for early tumor diagnosis, treatment monitoring, and prognostic assessment.
[0032] However, there is currently a lack of workflow acceleration algorithms for tumor data with UMI tags.
[0033] Therefore, an embodiment of the first aspect of the present invention proposes an accelerated method for bioinformatics analysis. This method involves splitting tumor data tagged with UMI tags into UMI-based data, ensuring that each split data set contains the same UMI sequence information. The LUSH component is then used to analyze the sequencing data containing the same UMI sequence information, completing four functions: data quality control, alignment, sorting, and duplicate labeling. This collectively accelerates the analysis process of tumor information tagged with UMI tags. Figure 1 As shown, when the input file is configured, such as raw sequencing data and its corresponding sample information, it determines whether the same sample contains data with the same library and the same UMI. If data with the same library and the same UMI exists, this type of data is merged, and the merged UMI type is then split according to resolution to obtain data corresponding to different resolution UMIs. If data with the same library and the same UMI does not exist, the data is directly split according to resolution within the UMI type to obtain fq data corresponding to different resolution UMIs. The split data is then accelerated, including data quality control, alignment, sorting, and duplicate labeling. The BAM files of the same library but different UMIs obtained in the previous step are merged, and the alignment BAM files of different libraries are merged to obtain the final alignment BAM for the sample. The BAM file can be used for other downstream analyses, such as quality control and variant detection.
[0034] The bioinformatics analysis acceleration method of this invention includes: providing raw sequencing data of one or more samples, wherein the raw sequencing data includes a first sequencing file with one or more UMI tags derived from one or more sequencing libraries; processing the first sequencing file into a second data file such that the second data file has a first classification of non-repeating UMI tags; splitting the second data file into a plurality of third data files based on the second classification of the UMI tags; providing the plurality of third data files to a bioinformatics analysis component to obtain a plurality of fourth data files; and processing the fourth data files into a fifth data file such that the fifth data file has a first classification of non-repeating UMI tags.
[0035] In some embodiments, raw sequencing data of one or more samples are provided. Due to the large volume of tumor data, to meet data quality control requirements, there may be situations where a single sample has multiple sequencing libraries with different UMIs, or where the same sequencing library has different UMIs. In some embodiments, such as... Figure 2 As shown, the Normal sample with sample number 22B28475430 includes 8 raw sequencing files. These raw sequencing files are from different sequencing libraries V350120886 and V350120916, and both libraries contain 4 lane data. Furthermore, the first category of the UMI tag in all raw sequencing files is the same, which is 1246.
[0036] In some embodiments, processing the first sequencing file into a second data file includes: merging multiple first sequencing files that originate from the same sequencing library and have the same UMI tag of the same first classification into a second data file; or directly providing the first sequencing file as the second data file based on the absence of multiple first sequencing files that originate from the same sequencing library and have the same UMI tag of the same first classification.
[0037] In some embodiments, the step of processing the first sequencing file into a second data file so that the second data file has a first classification with non-repeating UMI tags can be achieved by the following operation, fastq* lane This represents fastq* data from the same library, the same UMI, but different lanes. stb This represents the jastq* data resulting from merging data with the same UMI tags from the same library:
[0038] [BEGIN
[0039] cat
[0040] cat
[0041] END]
[0042] In some embodiments, four lane data points from the Normal sample (sample number 22B28475430) originating from sequencing library V350120886, all with a first category of UMI tag 1246, are merged into a second data file, fastq1. stb Furthermore, the four lane data points of the Normal sample (sample number 22B28475430) originating from sequencing library V350120916, with UMI tags all having a first category of 1246, were merged into a second data file, fastq2. stb .
[0043] In some embodiments, the resolution of the second category of the UMI tag is A. Preferably, the resolution A is 2*2 or 16*16, and more preferably, the resolution is 2*2. UMI tags have different resolutions, with 2*2 and 16*16 being the most commonly used. UMI tags with different resolutions can significantly improve the sensitivity of tumor detection, providing important information for early tumor diagnosis, treatment monitoring, and prognostic assessment.
[0044] In some embodiments, the second data file is split into multiple third data files based on the second classification of the UMI tags. It is understood that the first classification of the UMI tags in the processed second data is the same. At this time, the base sequences corresponding to the first classification of the UMI tags are loaded or read to split the data according to the second classification of the UMI tags. Figure 3 As shown, when the base sequence corresponding to the first category 1001 of the UMI tag is CTCTCT,TGTGTG, its resolution is 2*2, meaning the base sequence of the UMI is divided into two parts. CTCTCT can be represented as A and TGTGTG can be represented as B, thus A and B can be arranged and combined into AA, AB, BA, and BB. By arranging and combining the base sequence corresponding to the first category 1001 of the UMI tag into the form AA, AB, BA, and BB, the second category of the UMI tag is assigned.
[0045] Furthermore, based on the second classification of the UMI tags, the second data file can be split into multiple third data files. As discussed above, a 2*2 resolution UMI tag, after splitting, results in four pairs of fastq files. In some embodiments, these can be represented as *__0__0_1.fq and *__0__0_2.fq (UMI tag AA); *__0__1_1.fq and *__0__1_2.fq (UMI tag AB); *__1__0_1.fq and *__1__0_2.fq (UMI tag BA); and *__1__1_1.fq and *__1__1_2.fq (UMI tag BB), each pair of fastq data containing the same UMI sequence, such as... Figure 4 As shown. Similarly, a 16*16 UMI tag, after being split, results in a total of 256 pairs of fq data.
[0046] In some embodiments, splitting the second data file into multiple third data files includes: splitting a single second data file into A third data files, wherein the first category of the UMI tags in the A third data files is the same, and the second category is different.
[0047] In some embodiments, multiple third data files are provided to a bioinformatics analysis component to obtain multiple fourth data files, and the fourth data files are processed into a fifth data file such that the fifth data file has a first classification with non-repeating UMI tags. Figure 5 As shown, the split third data file is presented to a bioinformatics analysis component for data quality control, alignment, sorting, and repeat labeling. In some embodiments, the analysis component is a Sentieon component, a Megabolt component, and a LUSH component; preferably, the analysis component is a LUSH component. It should be noted that the repeat labeling mentioned here refers to identifying the large number of repeats generated during PCR to obtain a more accurate mutation abundance.
[0048] Then, the four BAM files with duplicate markers are merged, which can be accomplished through the following steps: `bam` represents the BAM file after accelerated analysis of the data after the i-th split. stb The merged BAM file is represented using the Samtcols software. The command is to merge.
[0049] BEGIN
[0050]
[0051] END.
[0052] In some embodiments, the method further includes removing suboptimal alignments from the fifth data file to obtain a sixth data file. The purpose of removing suboptimal alignments is primarily to improve the accuracy and reliability of sequencing data analysis. Suboptimal alignments may contain erroneous alignment information, and removing them can reduce false positives.
[0053] In some embodiments, a sequencing library based on a sample contains multiple UMI tags. The BAM files generated from all UMI tags within the same library for that sample are merged. That is, the sixth data file from the same sequencing library is merged into a seventh data file, which can be accomplished through the following steps: This represents the bam file after merging the i-th umt, with This indicates that the BAM files resulting from merging different UMI files in this library were merged using the samscols software's merge command, as shown below:
[0054] BEGIN
[0055]
[0056] END.
[0057] In some embodiments, based on the fact that a sample contains multiple sequencing libraries, the BAMs of all sequencing libraries in the sample are merged. That is, the method further includes: merging the seventh data file from the same sample into an eighth data file, which can be accomplished through the following steps:
[0058] by Represents the i-th document library The merged BAM file, with This refers to the BAM file of the sample after merging different libraries, which is merged using the merge command in the semtools software.
[0059] BEGIN
[0060]
[0061] END.
[0062] In some embodiments, the first sequencing file, the second data file, and the third data file are at least one of a fastq file and a bam file, and the fourth data file, the fifth data file, the sixth data file, the seventh data file, and the eighth data file are at least one of a sam file, a bam file, and FASTA. The raw sequencing data includes any one of genome sequencing data, transcriptome sequencing data, proteomics data, metabolomics data, single-cell omics data, and spatiotemporal omics data.
[0063] It is understandable that the above steps already include the acceleration process related to the invention, and subsequent steps can be customized according to their respective needs, such as in-depth quality control, mutation detection, annotation, etc.
[0064] A second aspect of the present invention provides a bioinformatics analysis apparatus, comprising: a data providing module for providing raw sequencing data of one or more samples, wherein the raw sequencing data includes a first sequencing file with one or more UMI tags derived from one or more sequencing libraries; a first processing module for processing the first sequencing file into a second data file such that the second data file has a first classification of non-repeating UMI tags; a second processing module for splitting the second data file into a plurality of third data files based on the second classification of the UMI tags; an analysis module for providing the plurality of third data files to a bioinformatics analysis component to obtain a plurality of fourth data files; and a third processing module for processing the fourth data files into a fifth data file such that the fifth data file has a first classification of non-repeating UMI tags.
[0065] A third aspect of the present invention provides a computer storage medium storing a computer program, the computer program including program instructions that, when executed by a processor, implement the method described in any embodiment of the first aspect of the present invention.
[0066] An embodiment of the fourth aspect of the present invention provides an electronic device, the electronic device including a memory and a processor connected together, the memory being used to store a computer program, and the processor being used to invoke the computer program to implement the method described in any embodiment of the first aspect of the present invention.
[0067] Example
[0068] In this embodiment of the invention, a sample with an average depth of 3000X after deduplication was selected. By applying the accelerated bioinformatics analysis method described in this invention, the analysis time and variation results of conventional detection and analysis procedures were compared.
[0069] The comparison results are shown in Table 1. The total analysis time using the accelerated bioinformatics analysis method described in this invention was 2.2 hours, while the same sample took approximately 8 hours using the conventional analysis procedure, with a 63.8% reduction in CPU usage. The method described in this invention significantly shortens the analysis time and has excellent acceleration effects. Furthermore, the consistency between the sample and the final variant detection results was 100%, ensuring the consistency of variant detection.
[0070]
[0071]
[0072]
[0073]
[0074]
[0075]
[0076] Note: "common" indicates that the original conventional analysis process is consistent with the test results of this invention.
[0077] Based on the comparison of the above batch test data, the method of the present invention not only improves the analysis speed, but also does not affect the mutation detection, with a consistency of 100%.
[0078] In this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0079] In this invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0080] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.
Claims
1. A method for accelerating bioinformatics analysis, characterized in that, include: Provide raw sequencing data for one or more samples, wherein the raw sequencing data includes a first sequencing file with one or more UMI tags derived from one or more sequencing libraries; The first sequencing file is processed into a second data file so that the second data file has a first classification with non-repeating UMI tags; Based on the second classification of the UMI tag, the second data file is split into multiple third data files; Multiple third data files are provided to the bioinformatics analysis component to obtain multiple fourth data files; and The fourth data file is processed into a fifth data file so that the fifth data file has a first category with non-repeating UMI tags.
2. The method for accelerating bioinformatics analysis according to claim 1, characterized in that, The step of processing the first sequencing file into a second data file includes: Based on multiple first sequencing files that originate from the same sequencing library and bear the same first-class UMI tag, these multiple first sequencing files that originate from the same sequencing library and bear the same first-class UMI tag are merged into the second data file; or Based on the absence of multiple first sequencing files originating from the same sequencing library and bearing the same first category UMI tag, the first sequencing file is directly provided as the second data file.
3. The method for accelerating bioinformatics analysis according to claim 1, characterized in that, The resolution of the second category of the UMI tag is A. Preferably, the resolution A is 2*2 or 16*16. More preferably, the resolution A is 2*2. The step of splitting the second data file into multiple third data files includes: The single second data file is split into A third data files, wherein the first category of the UMI tags in the A third data files is the same, and the second category is different.
4. The method for accelerating bioinformatics analysis according to claim 1, characterized in that, The analysis components are the Sentieon component, the Megabolt component X, and the LUSH component, preferably the LUSH component.
5. The method for accelerating bioinformatics analysis according to claim 1, characterized in that, Also includes: The suboptimal comparisons in the fifth data file are removed to obtain the sixth data file.
6. The method for accelerating bioinformatics analysis according to claim 5, characterized in that, Also includes: The sixth data file derived from the same sequencing library is merged into the seventh data file.
7. The method for accelerating bioinformatics analysis according to claim 6, characterized in that, Also includes: The seventh data file in the same sample is merged into the eighth data file.
8. The method for accelerating bioinformatics analysis according to claim 7, characterized in that, The first sequencing file, the second data file, and the third data file are at least one of a FastQ file and a BAM file. The fourth, fifth, sixth, seventh, and eighth data files are at least one of sam, bam, and FASTA files. The raw sequencing data includes any one of the following: genome sequencing data, transcriptome sequencing data, proteomics data, metabolomics data, single-cell omics data, and spatiotemporal omics data.
9. A bioinformatics analysis device, characterized in that, include: A data providing module is used to provide raw sequencing data for one or more samples, wherein the raw sequencing data includes a first sequencing file with one or more UMI tags derived from one or more sequencing libraries; A first processing module processes the first sequencing file into a second data file, so that the second data file has a first classification with non-repeating UMI tags; The second processing module, based on the second classification of the UMI tag, splits the second data file into multiple third data files; The analysis module provides multiple third data files to the bioinformatics analysis component to obtain multiple fourth data files; and The third processing module processes the fourth data file into a fifth data file so that the fifth data file has a first classification with non-repeating UMI tags.
10. A computer storage medium, characterized in that, The computer storage medium stores a computer program, which includes program instructions that, when executed by a processor, implement the method according to any one of claims 1 to 8.
11. An electronic device, characterized in that, The electronic device includes a memory and a processor connected together. The memory is used to store a computer program, and the processor is used to call the computer program to implement the method according to any one of claims 1 to 8.