Mitochondrial genome variation screening and automatic guide annotation platform

Abstract

The invention discloses a mitochondrial genome variation screening and automatic guide annotation platform, and a platform processor is configured to realize the following steps: data quality control: performing quality control on gene sequencing offline data to obtain a high-quality sequence; comparing data, namely comparing a high-quality sequence obtained after quality control to a human mitochondrial genome; re-comparing the data, and locally re-comparing the sequence of the region near the indel on the human mitochondrial genome; and mutation detection: carrying out gene mutation detection on the re-compared sequence. Mutation filtering: filtering gene mutation according to mutation frequency, sequencing depth and mutation reads number; mutation annotation: annotating the filtered gene mutation, and filtering the gene mutation again according to an annotation result after annotation of the gene mutation is completed. And quality control analysis: carrying out statistical analysis on data quality control according to the data volume, the comparison rate, the insert size and / or the coverage uniformity index.

Description

Technical Field

[0001] The present invention belongs to the field of bioinformatics, and in particular relates to a mitochondrial genome variation screening and automated guide annotation platform, method and program product. Background Art

[0002] Mitochondria are two-layer membrane-enclosed organelles that exist in most eukaryotic cells. They are structures that produce energy in cells and are the main place for cells to perform aerobic respiration. Mitochondrial DNA (mtDNA) is the genome located in the mitochondria of eukaryotic cells, which can convert chemical energy in cells into a usable form - adenosine triphosphate (ATP). The total length of mtDNA is 16569bp, and it is a double-stranded, closed-loop structure that encodes a total of 37 genes (13 protein-encoding genes, 22 tRNA genes, and 2 rRNA genes). mtDNA has the characteristics of maternal inheritance. The evolutionary tree constructed through the mitochondrial genome can reflect the evolutionary relationship of the maternal lineage and the evolutionary history of the population. mtDNA has genetic characteristics such as semi-autonomy and tissue heterogeneity. Due to the above characteristics and the structural characteristics of mtDNA, the mutation frequency of mtDNA is 10-20 times higher than that of nuclear DNA. Mitochondrial mutations can cause functional defects in the electron transport chain on the mitochondrial membrane, and further lead to ATP synthesis disorders, causing mitochondrial diseases. By detecting mitochondrial mutations and rating them based on guideline rules, the causes of diseases can be found more quickly at the genetic level, providing assistance to clinical researchers and further promoting the development of clinical medicine. Summary of the invention

[0003] One of the embodiments disclosed herein is a method for screening and annotating mitochondrial genome variations, comprising the steps of: data quality control, i.e., quality control of gene sequencing data, including removing adapter sequences and low-quality sequences to obtain high-quality sequences;

[0004] Data alignment, i.e. aligning the high-quality sequences obtained after quality control to the human mitochondrial genome (rCRS);

[0005] Data re-alignment, that is, local re-alignment of the sequences near the indel region on the human mitochondrial genome;

[0006] Mutation detection is to detect gene mutations in the re-aligned sequences.

[0007] Mutation filtering, i.e. filtering gene mutations based on mutation frequency, sequencing depth, and number of mutation reads;

[0008] Mutation annotation: annotate the gene mutations after filtering. After the gene mutations are annotated, perform another gene mutation filtering based on the annotation results.

[0009] Analytical quality control, that is, statistical analysis of data quality control based on data volume, comparison rate, insert size and / or coverage uniformity indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, in which:

[0011] Figure 1 A flow chart of a method for detecting mitochondrial DNA mutation according to one embodiment of the present invention.

[0012] Figure 2 A flow chart of a mitochondrial DNA mutation rating method according to one embodiment of the present invention.

[0013] Figure 3 A flow chart of a mitochondrial DNA mutation rating method according to one embodiment of the present invention.

[0014] Figure 4 A flow chart of a mitochondrial DNA mutation rating method according to one embodiment of the present invention. DETAILED DESCRIPTION

[0015] Existing methods for detecting mitochondrial DNA (mtDNA) mutations were examined. In a patent document, a mitochondrial mutation detection method and kit were disclosed. The method is a method for detecting 1494T and 1555G mutations. Five primers are used. Multiple PCR amplification is performed with the primers, and melting curve analysis is performed. Whether a corresponding mutation occurs is determined by the melting peak at a specific temperature. At the same time, the document also discloses a special kit, which includes: nucleic acid DNA extraction reagents, negative controls, amplification reaction reagents and instructions. The amplification reaction reagents include taq polymerase, PCR buffer, Mg2+, dNTP and primers. There are five primers in total.

[0016] In another patent document, a primer set and method for detecting mutations in 108 polymorphic sites of deafness-related genes are disclosed. A primer set is designed for the gene fragments where 108 mutation sites of 18 genes related to hereditary deafness are located, and an amplicon capture sequencing method is used. The primer set is first used for multiplex PCR amplification to construct a sequencing library, and then combined with the new generation high-throughput sequencing platform Illumina miniseq for detection. By analyzing all sequencing signals, the relative quantification of different bases at each site of the DNA sequence and the sequence determination of the DNA fragments are achieved.

[0017] In another patent document, a method for detecting mtDNA mutations in trace cells is proposed, comprising: extracting mtDNA from trace cells, wherein the number of the trace cells is 1 to 4; performing PCR amplification on different predetermined regions of the mtDNA to obtain amplification products of different predetermined regions, wherein the different predetermined regions of the mtDNA are superimposed to form the full-length mtDNA; mixing the amplification products of different predetermined regions to obtain a full-length mtDNA nucleic acid sequencing library; sequencing the nucleic acid sequencing library to obtain sequencing results; and determining the mtDNA mutations in the trace cells to be tested based on the sequencing results.

[0018] Analysis of existing methods for detecting mitochondrial DNA (mtDNA) mutations shows that existing methods inevitably have several limitations, including:

[0019] 1. It is impossible to amplify mtDNA completely. Only part of mtDNA is amplified or the full length of mtDNA is obtained by splicing;

[0020] 2. Both use amplicon capture sequencing, which is limited by amplification efficiency;

[0021] 3. The impact of nuclear genome sequence and mitochondrial DNA homologous region amplification on mutation detection was not considered;

[0022] 4. There is no rating for the detected mutations, so it is impossible to clearly show the impact of the mutations on the disease and accurately detect the cause of the disease.

[0023] The reasons for the above limitations are mainly due to the limitations of instruments, operators and the complexity of mutation rating. The ordinary amplicon method has relatively low requirements for instruments, operators' operating level requirements are also relatively low, and the development time is short. The full length of the mitochondrial genome is only 16569bp. Long-range PCR can be used to directly amplify the full length of the mitochondrial genome, reducing or reducing the impact of splicing or amplification efficiency. The experimental cycle is long and the cost is high, and the requirements for the operating level of instruments and operators are also higher. In addition, there are many existing mutation rating standards, and the databases involved are numerous and scattered. The integration of rating standards and databases requires a lot of manpower costs.

[0024] In order to solve the limitations of existing mitochondrial DNA (mtDNA) mutation detection methods, the present disclosure optimizes a public mitochondrial Long-range PCR (long-range polymerase chain reaction) technology, optimizes its experimental conditions, and uses a second-generation sequencing method for sequencing. On this basis, a mitochondrial DNA mutation detection scheme as proposed in the embodiments of the present disclosure is proposed, and mutation detection and mutation rating are performed on the sequenced data. After the mutation detection and rating results of the present disclosure are verified by the gold standard method, more than 99% consistency is obtained.

[0025] According to one or more embodiments, Figure 1 As shown, a method for detecting mitochondrial DNA mutations comprises the steps of:

[0026] S101, data quality control, using the TrimGalore program to perform quality control on the offline data, i.e., quality control, including removing adapter sequences and low-quality sequences. The offline data here refers to the original sequence data generated by the sequencing platform or sequencer;

[0027] S102, data alignment, using the bwa program to align the quality-controlled high-quality sequences to the human mitochondrial genome (rCRS);

[0028] S103, data realignment, using the GATK program to perform local realignment to the region near the indel on the human mitochondrial genome. Through realignment, the alignment limitations of the bwa program were resolved and the error rate of mutation detection was reduced;

[0029] S104, mutation detection, using the VarScan program to perform mutation detection on the re-aligned sequences;

[0030] S105, mutation filtering, filtering mutations according to mutation frequency, sequencing depth, mutation reads and other data, removing mutations with mutation frequency < 0.02 and sequencing depth < 1000, retaining high-confidence mutations for subsequent analysis; S106, mutation annotation, using the Annovar program to annotate the filtered mutations, and performing mutation filtering based on the annotation results;

[0031] S107, quality control analysis, using programs such as seqtk, samtools, gatk, bedtools, picard, etc. to perform statistical analysis on quality control, including indicators such as data volume, alignment rate, average length of insert fragments (usually DNA or cDNA), and coverage uniformity.

[0032] in,

[0033] TrimGalore program, TrimGalore is a bioinformatics tool for high-throughput sequencing data preprocessing. It is mainly used to perform quality control and trimming of raw sequencing data (i.e., offline data) to improve the accuracy and efficiency of subsequent analysis.

[0034] bwa program, bwa (Burrows-Wheeler Alignment Tool) is a bioinformatics tool used to align short sequences to a reference genome, suitable for alignment analysis of high-throughput sequencing data.

[0035] GATK program, GATK (Genome Analysis Toolkit) is a toolkit for high-throughput sequencing (HTS) data analysis, used for tasks such as variant detection at the single sample and population level, DNA and RNA sequence alignment, batch processing and pipeline management.

[0036] VarScan program, VarScan is a software tool for high-throughput sequencing (NGS) data analysis, mainly used to detect genomic variations, including single nucleotide variations (SNVs) and insertions and deletions (Indels), suitable for tumor somatic mutation detection, and can process a variety of sequencing data such as target capture, exome and whole genome sequencing. Annovar program, Annovar is a bioinformatics tool for high-throughput sequencing data analysis, used to annotate genomic variations (such as SNPs, insertions and deletions, CNVs, etc.).

[0037] Seqtk program, Seqtk is a fast and lightweight bioinformatics tool set for processing high-throughput sequencing data, used to process sequence files in FASTA and FASTQ formats.

[0038] Samtools program, Samtools is a bioinformatics tool set used to process high-throughput sequencing data, especially for viewing, editing and converting SAM (Sequence Alignment / Map) and BAM (binary version of SAM) files.

[0039] Bedtools program, Bedtools is a bioinformatics tool set for genomic data analysis. It is used to process genomic file formats such as BED, GFF / GTF, VCF, etc. It is suitable for the analysis of genomic intervals, such as finding overlaps between intervals, calculating coverage, merging intervals, etc.

[0040] Picard program, Picard is a bioinformatics tool set for high-throughput sequencing data, used to process SAM / BAM files, and provides a series of tools for reading, writing, converting and analyzing high-throughput sequencing data.

[0041] Further, the embodiments of the present disclosure, such as Figure 2 As shown, a mitochondrial DNA mutation rating method comprises the steps of:

[0042] S201, rating database construction. Download the data required for rating from Mitomap, MitImpact, HmtVar, MitoIP, Clinvar and other databases, and convert their formats into a format that can be annotated by Annovar software;

[0043] S202, rating standard construction. According to the literature Specifications of the ACMG / AMP standards and guidelines for mitochondrial DNA variant interpretation, rating standards such as PVS1, PM4, silicoprediction, PM2, BA1, and BS1 were constructed;

[0044] S203, mutation rating: The filtered mutations are rated according to the database and rating criteria, and the pathogenicity of the mutation is evaluated based on the rating results.

[0045] In the disclosed embodiment, in step S102 of the mitochondrial DNA mutation detection method, the sequencing sequence is aligned to the mitochondrial reference genome based on the BWT data conversion algorithm. Here, BWT (Burrows-Wheeler Transform) is a data compression algorithm that converts a string into another string that can be reconstructed from the original string while maintaining a certain compression efficiency. The BWT algorithm is used to compress DNA sequence data, collects all possible cyclic permutations (i.e., all possible rotations) of a string, then sorts them in lexicographic order, and takes the last character of the sorted string as the output of the BWT. For longer strings, this process generates a BWT matrix. The steps of BWT conversion include:

[0046] Generate all cyclic permutations: For a given string, generate all possible cyclic permutations of it.

[0047] Lexicographical sorting: Sort these circular permutations into lexicographical order.

[0048] Extract the last column: Extract the characters of the last column from the sorted matrix. This string is the result of BWT conversion.

[0049] In addition, the mutation detection method includes the establishment of mutation filtering standards and quality control standards. The quality control standards are as follows:

[0050]

[0051] The analysis methods involved can be highly concurrent through the SGE system of the high-performance cluster. A high-performance cluster (HPC) is a group of loosely or tightly connected computers. These computers work together and can use multiple computers for parallel computing to achieve a very high computing speed. If any machine fails, the entire system can still operate normally. The SGE system is the main task scheduling system of the high-performance cluster. The specific implementation method is as follows:

[0052] qsub -Vv sample=$sample -cwd -N${sample}_analysis -l mf=100G -l cf=32 -q compute cluster -e elog -o olog run.sh

[0053] In a mitochondrial DNA mutation rating method disclosed in the present invention, the mutation rating process includes downloading multiple database data, and in the database construction step, removing the redundancy between multiple databases, retaining the latest database information, and reducing the calculation steps of data information in the analysis process. At the same time, in the mutation rating step, it also includes sorting out the rating standards that can be implemented through the program according to the literature, which greatly reduces the error rate, labor cost, and analysis time.

[0054] According to one or more embodiments, a mitochondrial DNA mutation detection process can be flexibly adjusted according to different types of data and requirements. The following is a specific description using a double-end sequencing sequence and a 4-core CPU for analysis. Specifically, the following steps may be included:

[0055] (1) Data quality control.

[0056] The TrimGalore program was used to perform quality control on the data from the next-generation sequencing machine to remove the adapter sequences and low-quality sequences. The specific analysis is as follows:

[0057] trim_galore--output_dir. / --fastqc--fastqc_args "-t 4"--paired--clip_R16--clip_R2 6--trim1--adapter ATCTCGTATGCCGTCTTCTGCTTG--adapter2ATCTCGTATGCCGTCTTCTGCTTG--path_to_cutadapt cutadapt input_R1.fastq.gz input_R2.fastq.gz

[0058] The meaning of each phrase is as follows:

[0059] trim_galore: TrimGalore program;

[0060] --output_dir. / : The output result path is the current path;

[0061] --fastqc --fastqc_args "-t 4": use 4 core CPUs to perform quality assessment on sequencing data;

[0062] --paired: The input second-generation sequencing data is in paired-end sequencing mode;

[0063] --clip_R1 6--clip_R2 6: cut off the first 6 bases of read1 and read2;

[0064] --trim1: delete 1 N from both sides of the sequence;

[0065] --adapter ATCTCGTATGCCGTCTTCTGCTTG--adapter2ATCTCGTATGCCGTCTTCTGCTTG: remove the adapter sequence to be removed from read1 and read2;

[0066] --path_to_cutadapt cutadapt: ​​path to the cutadapt program;

[0067] input_R1.fastq.gz input_R2.fastq.gz: read1 and read2 sequences of the second-generation sequencing sequence.

[0068] This step is the preprocessing stage of high-throughput sequencing data analysis. It is crucial to perform quality control on the raw sequencing data to ensure the accuracy and reliability of subsequent analysis. Specifically, it includes:

[0069] S1011, use the TrimGalore program to perform quality control and trimming on the paired-end sequencing data `input_R1.fastq.gz` and `input_R2.fastq.gz`;

[0070] S1012, cut off the first 6 bases of each read by specifying `--clip_R1` and `--clip_R2` options; S1013, remove low-quality N bases from both sides of the sequence using the `--trim1` option;

[0071] S1014, specify and cut adapter sequences via the `--adapter` and `--adapter2` options to remove possible sequencing primers or adapters;

[0072] S1015, use the `--fastqc` option to generate quality assessment reports before and after pruning, and specify `--fastqc_args` to use 4 threads to speed up processing;

[0073] S1016, specify the path of the Cutadapt program so that TrimGalore can call Cutadapt to perform sequence trimming.

[0074] S1017, output the trimmed sequence data to the current directory and generate a corresponding quality assessment report.

[0075] This step can effectively improve the quality of sequencing data and provide high-quality input data for subsequent steps such as alignment, variation detection, and gene expression analysis.

[0076] (2)Data comparison.

[0077] The bwa program was used to align the high-quality sequences after quality control to the human mitochondrial genome (rCRS, https: / / www.ncbi.nlm.nih.gov / nuccore / NC_012920.1). The specific analysis method is as follows: bwa mem-Mt 4-K 10000000rCRS.fa input_R1.clean.fastq.gz input_R2.clean.fastq.gz>align.sam&&

[0078] samtools sort-@4align.sam>align.bam&&

[0079] samtools index align.sam

[0080] The meaning of each phrase is as follows:

[0081] bwa: bwa program;

[0082] mem: Set the comparison mode to mem;

[0083] -t 4: Use 4 CPU cores for comparison;

[0084] -K 10000000: input 10000000 bases per batch during the alignment process;

[0085] rCRS.fa: input reference genome (bwa index format);

[0086] input_R1.clean.fastq.gz input_R2.clean.fastq.gz: high-quality sequences for alignment analysis;

[0087] samtools sort: sort the sam files and convert them into bam;

[0088] -@4: sorting uses 4-core CPU;

[0089] samtools index: Create an index for bam files.

[0090] This step compares the sequence obtained by sequencing with the reference genome, and specifically includes:

[0091] S1021, use BWA's MEM mode to align the paired-end sequencing reads `input_R1.clean.fastq.gz` and `input_R2.clean.fastq.gz` to the human mitochondrial genome reference sequence `rCRS.fa`, and output the alignment results in SAM format;

[0092] S1022, use the sorting function of Samtools to sort the alignment results in SAM format in coordinate order and output them as a file `align.bam` in BAM format;

[0093] S1023, use the indexing function of Samtools to create an index for the sorted BAM file `align.bam` to facilitate quick access to the file in subsequent analysis.

[0094] This step ensures accurate alignment of the sequencing data with the reference genome and provides an organized and easily accessible BAM file for subsequent steps such as variant detection and gene expression analysis.

[0095] (3) Data re-comparison.

[0096] The GATK program was used to perform local realignment of the region near the indel on the human mitochondrial genome, and the specific analysis is as follows.

[0097] gatk-T RealignerTargetCreator-R rCRS.fa-I align.bam-L target.bed-osample.realigner.intervals&&

[0098] gatk-T IndelRealigner-R rCRS.fa-I align.bam-targetIntervalssample.realigner.intervals-L target.bed-o realign.bam&&

[0099] samtools index realign.bam

[0100] The meaning of each phrase is as follows:

[0101] gatk: gatk program;

[0102] -T RealignerTargetCreator, -T IndelRealigner: specify the analysis function;

[0103] -R rCRS.fa: specifies the reference genome;

[0104] -I align.bam: input bam file;

[0105] -L target.bed: input bed interval file;

[0106] -o sample.realigner.intervals, -o realign.bam: output result file;

[0107] samtools index: Create an index for bam files.

[0108] This step is aimed at re-aligning the mitochondrial genome and is an important step in analyzing mitochondrial DNA variation. Specifically, it includes:

[0109] S1031, use the `RealignerTargetCreator` module to identify regions near indels that may need realignment, and output these regions to the `sample.realigner.intervals` file;

[0110] S1032, use the `IndelRealigner` module to perform local realignment on these specific regions to improve the accuracy of indel detection, and output the results to the `realign.bam` file;

[0111] S1033, use the `samtools index` command to create an index for the realigned BAM file to facilitate quick access to the file in subsequent analysis.

[0112] This step can effectively deal with the alignment problem of indel regions on the mitochondrial genome and improve the accuracy and reliability of variation detection.

[0113] (4) Mutation detection.

[0114] The VarScan program was used to detect mutations in the realigned bam. The specific analysis is as follows: samtoolsmpileup -Bf rCRS.fa -l target.bed -d 200000realign.bam>sample.pileup&&

[0115] varscan mpileup2snp sample.pileup--min-var-freq 0.001--output-vcf--strand-filter 0--p-value 1|vcf_sort>sample.snp.vcf&&

[0116] varscan mpileup2indel sample.pileup--min-var-freq 0.001--output-vcf--strand-filter 0--p-value 1|vcf_sort>sample.indel.vcf&&

[0117] bzip sample.snp.vcf&&tabix-p vcf sample.snp.vcf.gz&&

[0118] bzip sample.indel.vcf&&tabix -p vcf sample.indel.vcf.gz&&

[0119] bcftools concat sample.snp.vcf.gz sample.indel.vcf.gz-aO v-osample.vcf

[0120] The meaning of each phrase is as follows:

[0121] samtools mpileup: Use samtools to calculate the bases of each region on the mitochondrial genome of the bam file;

[0122] -B: ignore base quality;

[0123] -f rCRS.fa: specifies the fasta file of the reference genome;

[0124] -d 200000: maximum sequencing depth for each site;

[0125] varscan: varscan program;

[0126] mpileup2snp, mpileup2indel: analyze SNPs and indels respectively;

[0127] --min-var-freq 0.001: minimum mutation frequency for analyzing mutations;

[0128] --output-vcf: output file is in vcf format;

[0129] --strand-filter 0: The chain-specific cutoff is 0;

[0130] --p-value 1: p-value cutoff is 1;

[0131] bgzip: Use bgzip program to compress vcf files;

[0132] tabix-p vcf: Use the tabix program to index the vcf file;

[0133] bcftools concat: merge bcf files;

[0134] -a: Allows the coordinates of the first site in the second merged file to be before the coordinates of the last site in the first file;

[0135] -O v: output file format is vcf;

[0136] -o sample.vcf: The output file is sample.vcf.

[0137] This step uses the VarScan program to perform mutation detection on the realigned BAM files to identify single nucleotide polymorphisms (SNPs) and insertions / deletions (indels).

[0138] This step first generates a pileup file using samtools, then detects SNPs and indels using VarScan, and outputs the results in VCF format. After that, the VCF file is compressed and indexed using bgzip and tabix, and finally the VCF files of SNPs and indels are merged into a unified VCF file using bcftools. This step helps to systematically identify and annotate genomic variations, providing high-quality input data for subsequent genetic analysis.

[0139] (5) Mutation filtering.

[0140] The mutation frequency, sequencing depth, and chain specificity of each mutation site were counted, and mutation sites with too low mutation frequency, too shallow sequencing depth, and obvious chain specificity were filtered out, as follows:

[0141] dict["af"]>=0.05and dict["DP"]>=1000and dict["strand_bias"]>=0.3and dict["strand_bias"]<=0.7

[0142] This example is implemented in Python, and its specific meaning is as follows:

[0143] dict["af"]>=0.05: filter mutations with mutation frequency>=0.05;

[0144] dict["DP"]>=1000: filter mutations with sequencing depth>=1000;

[0145] dict["strand_bias"]>=0.3and dict["strand_bias"]<=0.7: filter mutations with strand bias between 0.3 and 0.7;

[0146] The above standards can be changed during the process.

[0147] In this step, in the genomic variation analysis, mutation sites are counted and screened. Among them, the mutation frequency refers to the relative proportion of a specific mutation in a sample. In statistics, it is usually expressed as a decimal between 0 and 1, where 1 means that the site is completely composed of mutant alleles in the sample, and 0 means there are no mutant alleles. In the above description, `dict["af"]>=0.05` means that mutation sites with a mutation frequency greater than or equal to 5% are screened. This threshold can be adjusted according to the purpose of the study to ensure that the screened mutations are representative in the population.

[0148] Sequencing depth refers to the number of times the sequence coverage is obtained in a specific genomic region. A higher sequencing depth can improve the accuracy and reliability of mutation detection. The condition `dict["DP"]>=1000` means that a mutation site will only be considered if the sequencing depth of the site is at least 1000 times. This helps to exclude false positive mutations caused by insufficient sequencing depth.

[0149] Strand specificity refers to the preference of mutations on the positive or negative strand. In some cases, mutations may be over-represented on one strand, which may be due to bias in the sequencing process or experimental error. The condition `dict["strand_bias"]>=0.3and dict["strand_bias"]<=0.7` ensures that the distribution of the selected mutation sites on the strand is relatively balanced, excluding those mutations that are over-represented on a single strand, thereby reducing the possibility of false positives.

[0150] In the above Python language, the above conditions can be implemented through a dictionary (`dict`), where the key represents different statistical indicators, such as `"af"`, `"DP"` and `"strand_bias"`, and the value is a specific numerical value. By combining these conditions with logical operators (`and`), a composite screening criterion can be constructed to achieve accurate screening of mutation sites.

[0151] Therefore, in the disclosed embodiment, mutation frequency>=0.05, sequencing depth>=1000 and / or mutation with 0.3<=chain preference<=0.7 are set, because:

[0152] (1) Mutation frequency >= 5%, which can filter out low-frequency mutations. Too low a mutation frequency can lead to false positives;

[0153] (2) Sequencing depth >= 1000, which is because the depth is too low, indicating that the coverage of the region is poor, which may lead to inaccurate detection;

[0154] (3) Chain preference >= 0.3 and chain preference <= 0.7. Chain preference can eliminate the influence caused by PCR or sequencing process.

[0155] Therefore, this combination of screening criteria for mutation filtering plays an important role in improving the accuracy of detection results.

[0156] In addition, these screening criteria, including mutation frequency (Allele Frequency, AF), sequencing depth (Depth of Coverage, DP), and strand specificity (Strand Bias), can be adjusted according to the specific needs of the study and the characteristics of the data set. For example, if the study focuses on low-frequency mutations, the threshold of mutation frequency may be lowered; if the sequencing depth of the data set is generally low, the sequencing depth standard may need to be adjusted. In this way, the analysis process can be flexibly customized to adapt to different research objectives and data conditions.

[0157] (6) Mutation annotation.

[0158] The Annovar program was used to annotate the filtered mutations. The specific analysis is as follows: perl annovar.plsample.filter.vcf annovar_db -out sample -nastring.-vcfinput --thread4 --dot2underline

[0159] The meaning of each phrase is as follows:

[0160] perl: Use perl language for analysis;

[0161] annovar.pl: calls the annovar program;

[0162] sample.filter.vcf: input file, filtered vcf file;

[0163] annovar_db: annovar annotation database path;

[0164] -out sample: output file prefix;

[0165] -nastring.: There is no information marked as. in the annotation database;

[0166] -vcfinput: input file format is vcf;

[0167] --thread 4: Use 4 CPU cores for analysis;

[0168] --dot2underline: Replace dots (.) in file names with underscores (_).

[0169] This step annotates the genetic variation data to understand the biological significance and potential impact of the variation. This step calls the Annovar program to perform efficient and automated annotation processing on the filtered VCF files, while ensuring the naming conventions of the output files and the efficiency of thread usage during the processing.

[0170] (7) Analytical quality control.

[0171] The quality control indicators of each step in the statistical process, including data volume, data quality, comparison rate, coverage uniformity, etc., are summarized and marked as unqualified samples. The specific usage examples are as follows:

[0172] if QC_index <QC_critieria:

[0173] qc_result.append("QC_index_lower_QC_critieria")

[0174] The above example is implemented in Python, and the specific meaning is as follows:

[0175] If a quality control index (QC_index) is lower than the quality control standard (QC_critieria), the sample will be marked QC_index_lower_QC_critieria.

[0176] This step involves quality control (QC) in the statistical analysis process, which is a step to ensure data accuracy and reliability. Quality control indicators are used to evaluate the integrity and usability of data sets. Common quality control indicators include:

[0177] Data Quantity: refers to the total amount of data collected, reflecting the richness of the sample.

[0178] Data Quality: involves the accuracy of data and the authenticity of signals, which may include signal-to-noise ratio, repeatability, etc.

[0179] Mapping Rate: In genomics, it refers to the ratio of the sequence obtained by sequencing to the reference genome that is successfully mapped.

[0180] Coverage Uniformity: refers to whether the sequencing data is evenly distributed across the entire genome, reflecting the consistency of sequencing depth.

[0181] In this step, the quality control index of each sample is evaluated and compared with the preset quality control standard (QC_criteria). If the quality control index of the sample does not meet the standard, the sample is considered to be unqualified in quality. Among them:

[0182] `QC_index`: represents the quantitative value of a quality control indicator of the current sample.

[0183] `QC_criteria`: The preset quality control standard threshold, which is used as the basis for judging whether the sample is qualified.

[0184] `qc_result`: A list used to store quality control results.

[0185] `append`: is a list method used to add elements to the end of the list.

[0186] When the value of `QC_index` is lower than `QC_criteria`, it means that the sample does not meet the quality control standard. At this time, a specific tag `"QC_index_lower_QC_criteria"` is added to the `qc_result` list to record the quality control failure of the sample. This method helps to further analyze or exclude the unqualified samples.

[0187] Therefore, in this step of analytical quality control, when the quality control standard is:

[0188] Data volume >= 0.5Gb, alignment index >= 0.9 and 5000X coverage >= 0.99, due to:

[0189] (1) Data volume: >= 0.5 Gb, ensuring sufficient data volume for subsequent analysis;

[0190] (2) The comparison rate index is greater than or equal to 0.9. This index can be used to evaluate the contamination of the data. If the index is too low, there will be too little effective analysis data and there may be contamination from other sources.

[0191] (3) 5000X coverage>=0.99. Meeting this analysis condition can satisfy both the average depth and coverage conditions. If this indicator is too low, it will lead to false negatives in subsequent mutation detection.

[0192] Therefore, this step also plays an important role in ensuring the accuracy of the test results.

[0193] According to one or more embodiments, a method for rating mitochondrial DNA mutations comprises the following steps:

[0194] (1) Rating database construction. Download the data required for rating from various official databases and convert the format. The required official database information is as follows:

[0195] ClinVar https: / / www.ncbi.nlm.nih.gov / clinvar /

[0196] COSMIC https: / / cancer.sanger.ac.uk / cosmic

[0197] dbNSFP http: / / database.liulab.science / dbNSFP#database

[0198] HmtVar https: / / www.hmtvar.uniba.it

[0199] MitImpact https: / / mitimpact.css-mendel.it

[0200] MITOMAP https: / / www.mitomap.org / MITOMAP

[0201] mtDB http: / / www.mtdb.igp.uu.se

[0202] After the data is downloaded, extract the mitochondrial genome data in the database (the chromosome number is MT or chrMT), and then convert the extracted data into the following format:

[0203] Table 1

[0204] (2) Construction of rating standards.

[0205] The rating standards were constructed according to the literature Specifications of the ACMG / AMP standards and guidelines for mitochondrial DNA variant interpretation (https: / / europepmc.org / article / MED / 32906214). The specific ratings and standards include: PVS1, PM4, silico prediction, PM2, BA1, and BS1. Here, the rating standards of PVS1 / PM4, silico prediction, PM2 / BA1 / BS1 are derived from a set of guidelines jointly developed by the American College of Medical Genetics and Genomics (ACMG) and the Association for Molecular Pathology (AMP) for assessing the pathogenicity of genomic variants. These guidelines are designed to provide a standardized framework for clinical genetics and molecular diagnostic laboratories to help interpret whether genetic variants may cause disease. Among them,

[0206] The PVS1 (Pathogenic Very Strong) standard means that if a variant causes loss of function in a gene with a known disease mechanism (such as nonsense variants, frameshift mutations, or authoritative splice site variants, etc.), it is considered to be very strong evidence of pathogenicity.

[0207] The PM4 (Pathogenic Moderate) criterion involves changes in protein length, such as protein truncation or elongation due to non-frameshift mutations or loss of stop codons, which may affect the function of the protein.

[0208] In silico prediction refers to the use of computer algorithms and databases to predict the effect of variants on protein function. These tools usually consider factors such as sequence conservation, protein structure prediction, and the effect of variants on protein stability.

[0209] The PM2 (Pathogenic Moderate) standard involves the frequency of the variant in the normal control population. If a mutation does not exist in the healthy population or has a very low probability of occurrence (usually <0.00002), but a patient carries the variant, the mutation is considered to be moderately pathogenic.

[0210] The BA1 (Benign Strong) criterion refers to a very high allele frequency of a variant in a specific population (e.g., greater than 5%), which generally indicates that the variant is benign because it exists in a large number of normal individuals and does not cause an obvious disease phenotype.

[0211] The BS1 (Benign Strong) criterion involves the frequency of the variant in a normal control population; if the variant frequency is between 0.5% and 1%, this is considered strong evidence supporting the benign nature of the variant.

[0212] These standards are used in genetic diagnosis and research to help determine the pathogenicity of gene variants. Although they are based on a large amount of scientific research and clinical data, with the emergence of new data, these standard guidelines are no longer consistent with new research conclusions and need to be updated and revised. At the same time, the embodiments of the present disclosure propose to combine multiple standards to comprehensively evaluate the pathogenicity of a variant. The evaluation criteria proposed in the embodiments of the present disclosure are as follows.

[0213] (a) PVS1 / PM4 rating.

[0214] Rating Rating Criteria PVS1_Undetermined The mutation is located in the first two digits of the start codon and is a missense mutation PVS1_Moderate Nonsense mutation or frameshift mutation, and the remaining amino acids after the mutation account for more than 90% of the total amino acids PVS1_Strong Nonsense mutation or frameshift mutation, and the remaining amino acids after the mutation account for less than 90% of the total amino acids PM4 Non-frameshift mutations, insertions or deletions of multiples of 3 bases NA Mutations in other situations

[0215] Table 2

[0216] Here,

[0217] Start codon: refers to the codon at the start site of protein synthesis;

[0218] Nonsense mutation: refers to a single base mutation caused by the codon encoding an amino acid in the sequence mutating into a stop codon. Frameshift mutation: refers to the change of the reading frame caused by the deletion or insertion of a base at a certain site in the DNA molecule, resulting in a series of code changes downstream, causing the gene that originally encoded a certain peptide chain to encode another completely different peptide chain sequence;

[0219] Non-frameshift mutation: refers to an insertion or deletion of a base number that is a multiple of 3.

[0220] According to the above rating criteria, mitochondrial DNA mutations are rated. Determine whether the mitochondrial DNA mutation is located in the coding gene region. If so, it is judged as PVS1_Undetermined, PVS1_Moderate, PVS1_Strong, PM4, and NA according to the mutation type and location. Figure 3 As shown. Among them, the rating standard for PVS1_Undetermined is that the mutation position is located in the first two digits of the start codon and is a missense mutation. PVS1 means Pathogenic Very Strong. Due to the mutation of the start codon, protein expression may be abnormal. The construction of this rating standard plays an important role in clearly showing the impact of mutations on diseases and helping researchers to more accurately identify the cause of the disease.

[0221] (b) silico prediction.

[0222]

[0223]

[0224] Table 3

[0225] In the standard, mitochondrial DNA mutations are graded, comprising the following steps:

[0226] S401, mutation type identification, determining whether the mutation is located in mRNA or tRNA;

[0227] S402, score calculation:

[0228] -For mutations on mRNA, calculate the APOGEE score.

[0229] -For mutations on tRNA, the MitoTIP score and HmtVAR disease score are calculated.

[0230] S403, rating decision: according to the mutation type and the calculated score, the rating criteria in Table 3 are applied for rating. S404, the rating result of the mutation is output, which is PP3, BP4, NA or UNKNOWN.

[0231] (c)PM2 / BA1 / BS1.

[0232] Rating Rating Criteria PM2_Supporting GenBank Variant Frequency is empty or < 0.00002 BA1 GenBank Variant Frequency>=0.01 BS1 0.005<=GenBank Variant Frequency<0.01 NA 0.00002<=GenBank Variant Frequency<0.005

[0233] Here GenBank Variant Frequency: the mutation frequency of the population in the GenBank database;

[0234] like Figure 4 As shown, a specific process for mitochondrial DNA mutation rating is as follows.

[0235] S501, check "GenBank_Variant_Frequency" and compare the observed variant frequency with the frequency recorded in the GenBank database.

[0236] S502, determine whether there is missing data, if "no", continue the process, if "yes", give PM2_Supporting rating;

[0237] S503, HOTSPOT check. After confirming that there is no vacancy, it is asked whether it is "HOTSPOT". HOTSPOT refers to the mutation hotspot area. If it is "yes", it means that the mutation is located in a known hotspot area. If it is "no", it is not a hotspot area.

[0238] S504, determine whether there is PM2 support, if it is "no", skip the PM2-related rating steps, if it is "yes", further rating is required;

[0239] S505, different rating standards are given according to different ranges of Genomic Variant Frequency (GVF):

[0240] - If GVF is less than 0.00002, then PM2_Supporting,

[0241] -NA if 0.00002 is equal to GVF and GVF is less than 0.005,

[0242] - If 0.005 is less than or equal to GVF and GVF is less than 0.01, then BS1,

[0243] - If GVF is greater than or equal to 0.01, then BA1.

[0244] (3) Mutation rating.

[0245] The filtered mutations are rated according to the rating database and rating criteria, and the pathogenicity of the mutations is evaluated based on the rating results. The specific usage examples are as follows:

[0246] if site_gene in hotspot_dict.keys():

[0247] related_disease=[hotspot_dict[site_gene][3]]

[0248] verify_info=[hotspot_dict[site_gene][6]]

[0249] Pathogenic_condition=[hotspot_dict[site_gene][5]]

[0250] The above example is implemented using Python language, and the specific meaning is as follows.

[0251] Determine whether the mutation site is in the list of rated mutation sites:

[0252] Associating mutation sites with relevant diseases;

[0253] Associating mutation sites with rating information;

[0254] Associate mutation sites with disease treatment information.

[0255] Therefore, in order to solve the problems in the existing mitochondrial DNA (mtDNA) mutation detection, the present disclosure proposes a mitochondrial genome variation screening and automated guide annotation system, which has the following features:

[0256] 1. The full length of mitochondrial DNA can be amplified by long range polymerase chain reaction (Long Range PCR) technology;

[0257] 2. It can remove the influence of homologous regions of nuclear genome and mitochondrial genome;

[0258] 3. It can detect mutation information in all regions of mitochondrial DNA;

[0259] 4. Based on databases such as MitoMap, HmtVar, ClinVar, and MitoIP, disease-related hotspots on mitochondrial DNA can be automatically rated. The program code can not only automatically rate disease hotspots and improve work efficiency, but also reduce the error rate of manual rating.

[0260] The program is based on Python code and is mainly divided into three parts as a gene variation analysis: silico prediction, PM2 / BA1 / BS1 judgment and PVS1 / PM4 rating.

[0261] In silico prediction is the use of computer simulation to predict biological events or molecular behaviors, such as protein structure, gene expression patterns, or the effects of gene mutations. PM2 / BA1 / BS1 judgment is the classification or rating of gene mutations. PVS1 / PM4 rating is used to assess the pathogenicity of gene mutations.

[0262] Wherein, the gene variation analysis specifically comprises the steps of:

[0263] (a) In silico prediction is used to predict the impact of gene mutations. First, determine the type of gene (GENE_TYPE), and then make a judgment based on different scoring systems (such as APOGEE_score or HmtVAR disease score, MitoTIP score). Incompatible cases will be marked as UNKNOWN to facilitate subsequent code search and repair. The specific logic is as follows:

[0264] - If the gene type is mRNA and the value of APOGEE_score is not undefined ("."), then according to the value of APOGEE_score:

[0265] If APOGEE_score is greater than or equal to 0.5, the prediction result is "PP3".

[0266] If APOGEE_score is less than 0.5, the prediction result is "BP4".

[0267] Otherwise, the prediction result is marked as "UNKNOWN".

[0268] - If the gene type is tRNA, and both the HmtVAR disease score and MitoTIP score are defined, based on the combination of these two scores:

[0269] If the HmtVAR disease score is "H" and the MitoTIP score is in "Q1" or "Q2", the prediction result is "PP3".

[0270] If the HmtVAR disease score is "H" and the MitoTIP score is in "Q3" or "Q4", the prediction result is "NA".

[0271] If the HmtVAR disease score is "L" and the MitoTIP score is in "Q1" or "Q2", the prediction result is "NA".

[0272] If the HmtVAR disease score is "L" and the MitoTIP score is in "Q3" or "Q4", the prediction result is "BP4".

[0273] In other cases, the prediction result is marked as "UNKNOWN".

[0274] - If the gene type is neither mRNA nor tRNA, or the score is undefined, the prediction result is directly marked as "UNKNOWN", which means "unknown".

[0275] (b) PM2 / BA1 / BS1 judgment (gene variant frequency assessment), the credibility of the variant is judged according to GenBank Variant Frequency (gene variant frequency), see Figure 4 The logic is as follows:

[0276] - If GenBank Variant Frequency is not defined ("."), the result is "PM2_Supporting". - If GenBank Variant Frequency is defined, convert it to decimal form and based on its value:

[0277] If the frequency is greater than or equal to 0.01, and hotspot_info (hotspot information) is not a hotspot or is undefined, the judgment result is "BA1".

[0278] If the frequency is between 0.01 and 0.005, and hotspot_info is not a hotspot or is undefined, the result is "BS1".

[0279] If the frequency is less than 0.00002, the judgment result is "PM2_Supporting".

[0280] In other cases, the judgment result is "NA", which means "not available".

[0281] (c) PVS1 / PM4 rating (mutation severity assessment), which assesses the severity of mutations based on their type and location. Figure 3 , does not conform to Figure 3 In such cases, they will be marked as NA, STOPGAIN_bug, or FRAMESHIFT_bug to facilitate subsequent search and repair. The logic is as follows:

[0282] - If the gene name (gene_name) is in the known coding gene dictionary (codinggene_dict), the start site and protein length are calculated and then judged according to the format of hgvsp (Human Genome Variation Society):

[0283] If the mutation occurred in the start codon and was not a synonymous mutation, and the amino acid after the start codon was not "I" or "M", it was rated as "PVS1_Undetermined".

[0284] If the mutation results in the addition of a stop codon (stopgain) and is not at the start site, the rating may be "NA", "PVS1_Moderate", or "PVS1_Strong", depending on the position of the mutation relative to the length of the protein.

[0285] If the mutation is a frameshift and is not at the start site, the rating may be "NA", "PVS1_Moderate", or "PVS1_Strong", depending on the position of the mutation relative to the length of the protein.

[0286] If the mutation is a non-frameshift insertion or deletion and is not in the start site, it is rated "PM4".

[0287] Otherwise, the rating is "NA".

[0288] - If the gene name is not in the dictionary of known coded genes, the rating is simply "NA".

[0289] In summary, the present disclosure overcomes the limitations of existing mitochondrial DNA (mtDNA) mutation detection methods and obtains the following beneficial effects:

[0290] 1. Based on the complete amplification of mtDNA, it is no longer limited to the analysis of partial sequences, which will greatly improve the comprehensiveness and accuracy of the test and solve the limitation that the full length of mitochondrial DNA cannot be obtained or the full length of mitochondrial DNA can only be obtained by splicing.

[0291] 2. Break through the efficiency limitations of existing amplicon capture sequencing methods, thereby speeding up detection and improving experimental throughput.

[0292] 3. Reduce the possible interference of nuclear genome sequences and mitochondrial DNA homologous regions on mutation detection, and improve the reliability of test results.

[0293] 4. Rating the detected mutations and clearly showing the impact of the mutations on the disease will help doctors and researchers identify the cause more accurately and provide stronger support for disease diagnosis and treatment.

[0294] 5. With a deeper understanding of mtDNA mutations, more personalized treatment plans can be provided to patients, improving treatment outcomes.

[0295] Therefore, the present disclosure solves the defects of existing mitochondrial DNA mutation detection, promotes research in the field of mitochondrial genetics, and helps to reveal more knowledge about the association between mitochondrial function and disease. While reducing the cost of detection, it improves the scientificity and practicality of detection, allowing more patients to more effectively prevent and intervene in diseases through early and accurate detection, reducing the occurrence and development of diseases.

[0296] It should be understood that in the embodiments of the present invention, the term "and / or" is only a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0297] 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 computer-readable storage medium. Based on this understanding, the technical solution of the present invention 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 several instructions for 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 method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0298] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for screening and annotating mitochondrial genome variations, characterized in that: The method is to screen for mitochondrial genome variation after performing next-generation sequencing on mitochondrial DNA fragments obtained by long-distance polymerase chain reaction amplification. The following steps are involved: A1, data quality control, that is, quality control of the obtained gene second-generation sequencing data, including removing adapter sequences and low-quality sequences to obtain high-quality sequences; A2, data alignment, i.e. aligning the high-quality sequences obtained after quality control to the human mitochondrial genome rCRS; A3, data re-alignment, that is, local re-alignment of the sequences near the indel region on the human mitochondrial genome; A4, mutation detection, i.e., performing gene mutation detection on the re-aligned sequences; A5, mutation filtering, that is, filtering gene mutations according to mutation frequency, sequencing depth, and mutation read number; A6, mutation annotation, annotates the gene mutations after filtering; A7, analytical quality control, that is, statistical analysis of data quality control based on data volume, alignment rate, insert size and / or coverage uniformity indicators.

2. The method according to claim 1, characterized in that The step A1 specifically includes: A11, quality control and trimming of the obtained sequencing data; A12, removes several low-quality bases at the beginning and end of each read; A13, remove low-quality N bases from both sides of the sequencing data sequence; A14, specifying and excising adapter sequences to remove sequencing primers or adapters; A15, trims the sequencing sequence, outputs the trimmed sequence data, and generates a corresponding quality assessment report.

3. The method according to claim 1, characterized in that The step A2 specifically includes: A21, aligning the sequencing sequence obtained in step A1 to the human mitochondrial genome reference sequence rCRS.fa, and outputting the alignment result in SAM format; A22, sort the alignment results in SAM format according to the coordinate order and output the file in BAM format; A23, create index for the sorted BAM file.

4. The method according to claim 3, characterized in that The step A3 specifically includes: The BAM file aligned in step A2 is locally realigned according to the sequence of the region near the indel, and the result is output as a BAM file; Create an index for the re-aligned BAM files.

5. The method according to claim 4, characterized in that The step A4 specifically includes: Perform mutation detection on the re-aligned BAM file, detect SNPs and indels, and output the results as VCF files. Compress and index the VCF files, and merge the SNP and indel VCF files into a unified VCF file.

6. The method according to claim 1, characterized in that The step A5 specifically includes: The mutation frequency, sequencing depth, and chain specificity of each mutation site were counted, and a composite screening standard was constructed to filter out mutation sites with too low mutation frequency, too shallow sequencing depth, and obvious chain specificity.

7. The method according to claim 1, characterized in that The method also includes rating the obtained mitochondrial genome variations, the rating method comprising the steps of: Construct a rating database, download the data required for rating from the human mitochondrial DNA database, and convert its format into an annotable format; Construct rating standards, including PVS1 / PM4, silico prediction, PM2 / BA1 / BS1 rating standards; And, the filtered gene mutations are rated according to the rating database and rating criteria, and the pathogenicity of the mutations is evaluated according to the rating results.

8. The method according to claim 7, characterized in that The PVS1 / PM4 rating standards are: The rating criteria for rating PVS1_Undetermined were that the mutation position was located in the first two digits of the start codon and was a missense mutation; The rating criteria for rating PVS1_Moderate are nonsense mutation or frameshift mutation, and the residual amino acids after the mutation account for more than 90% of the total amino acids; The rating criteria for rating PVS1_Strong are nonsense mutation or frameshift mutation, and the residual amino acids after the mutation account for 90% or less of the total amino acids; The rating criteria for rating PM4 are non-frameshift mutations with insertion or deletion of bases in multiples of 3; The silicoprediction rating criteria are: The rating criteria for rating PP3 include:

1. The mutation is located in mRNA and the APOGEE score is greater than or equal to 0.5; 2. The mutation is located in tRNA, the MitoTIP score is in the top 50% and the HmtVAR disease score is greater than or equal to 0.35; the rating criteria for rating BP4 include:

1. The mutation is located in mRNA and the APOGEE score is < 0.5; 2. The mutation is located in tRNA, the MitoTIP score is in the bottom 50% and the HmtVAR disease score is <0.35; the rating criteria for rating NA include:

1. The mutation is located in tRNA, MitoTIP score is in the bottom 50% and HmtVAR disease score>=0.35; 2. The mutation is located in tRNA, MitoTIP score is in the top 50% and HmtVAR disease score<0.35; PM2 / BA1 / BS1 The rating criteria are: The rating criteria for rating PM2_Supporting is that the GenBank Variant Frequency is empty or <0.00002; The rating criteria for rating BA1 is GenBank Variant Frequency>=0.01; The rating standard for rating BS1 is 0.005<=GenBank Variant Frequency<0.01; The rating standard for rating NA is 0.00002<=GenBank Variant Frequency<0.

005.

9. A mitochondrial genome variation screening and annotation platform, characterized in that: The platform includes a server having a memory; as well as A processor coupled to the memory, the processor being configured to execute instructions stored in the memory to implement the method according to any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that The computer program is executed by a processor to implement the method according to any one of claims 1 to 8.

Images