An amplicon sequencing data analysis method based on ion torrent platform

Abstract

The application provides an amplicon sequencing data analysis method based on an Ion Torrent platform, and belongs to the technical field of bioinformatics.The application provides an amplicon sequencing data analysis method based on an Ion Torrent platform.The application is based on the Torrent Suite (TS) and Ion Reporter (IR) system of the Ion Torrent platform, and through strict quality control, multi-algorithm integrated alignment, deep coverage analysis, accurate variation detection and automatic annotation, the risk of false positive and false negative results is significantly reduced, and the accuracy of data analysis is improved.Therefore, the amplicon sequencing data analysis method can significantly improve the accuracy of data analysis, the efficiency of clinical interpretation and the standardization of reports, and is especially suitable for genetic disease and tumor mutation detection scenarios.

Description

Technical Field

[0001] This invention belongs to the field of bioinformatics technology, and in particular relates to a method for analyzing amplicon sequencing data based on the Ion Torrent platform. Background Technology

[0002] Existing gene sequencing data analysis workflows often face the risk of false negatives and false positives in clinical applications. False negatives may lead to delays in the diagnosis and treatment of rapidly progressing diseases, while false positives may trigger unnecessary follow-up treatments. Furthermore, existing gene sequencing data analysis workflows have shortcomings in data quality control, algorithm optimization, and annotation resource integration, such as uneven sequencing depth, low alignment efficiency, and incomplete annotation information. Therefore, the clinical applicability of existing gene sequencing data analysis workflows needs to be improved. Summary of the Invention

[0003] The purpose of this invention is to provide an amplicon sequencing data analysis method based on the Ion Torrent platform. This amplicon sequencing data analysis method can improve the accuracy of data analysis and has high clinical applicability.

[0004] This invention provides a method for amplicon sequencing data analysis based on the Ion Torrent platform, comprising the following steps: 1) The amplicon sequencing data were obtained by sequencing the amplicon using the Ion Torrent platform; 2) Perform quality control on amplicon sequencing data, filter the sequencing data according to the chip operation quality parameters, and obtain sequencing data that meets the quality requirements; The chip operating quality parameters include multiple parameters from the following: total number of micropores on the chip, microbead loading rate, proportion of available microbeads, proportion of microbeads carrying the test template, proportion of multi-template microbeads, proportion of low-quality microbeads, and proportion of primer dimer microbeads. 3) The TMAP alignment tool of the Torrent Suite system is used to perform sequence alignment on the quality-controlled sequencing data. The Bowtie short, Bowtie long and SSAHA long algorithms are integrated to select the best alignment result to locate the human reference genome and generate a sequence alignment file in BAM format. 4) The coverage of the target region in the sequence alignment file is analyzed using the Coverage Analysis plugin of the Torrent Suite system to obtain amplicon coverage analysis data; the Coverage Analysis plugin performs statistical analysis on the coverage of the target region based on the BAM format alignment file generated by TMAP and the BED file of the target region. 5) Using Torrent Variant Caller in combination with diBayes and GATK algorithms, the human reference genome, AR panel target region BED file and hotspot variant VCF file are pre-set in the Torrent Suite system. Then, using Torrent Variant Caller to call diBayes and GATK algorithms, variant detection is performed on the BAM file generated by TMAP, identifying SNP and Indel variants respectively, and finally outputting variant site information in VCF format. 6) Upload the variant site information through the Ion Reporter system, automatically link it to the database, annotate the biological functions, and obtain the annotation results; 7) Filter the annotation results to screen for clinically relevant variants and remove germline mutations and low-quality variants to obtain clinically relevant variant data; the filter chain set for filtering the annotation results includes mutation coverage ≥250X, minimum allele frequency ≥5%, and exclusion of common SNPs and homopolymer interference.

[0005] Preferably, after obtaining clinically relevant variant data, the method further includes automatically integrating key annotation information based on the clinically relevant variant data to generate a standardized technical report containing details of the variant sites.

[0006] Preferably, the average sequencing depth of the target region is not less than 100×.

[0007] Preferably, the Indel includes small indel and longer indel.

[0008] Preferably, the variant site information includes genotype, frequency, type, and hotspot markers.

[0009] Preferably, the database includes ClinVar, dbSNP, and drug targets.

[0010] Preferably, the biological functions include the clinical significance of the variation, population frequency, amino acid changes, and pharmacogenomics information.

[0011] Preferably, the key annotation information includes gene regions, pathogenicity assessments, and drug targets.

[0012] Preferably, the filtering of amplicon sequencing data is based on Thermo Fisher quality control standards: bead loading rate ≥60%, usable bead ratio ≥90%, bead ratio with test template ≥90%, multi-template bead ratio ≤50%, primer dimer bead ratio ≤10%, and low-quality bead ratio ≤20%.

[0013] Preferably, the human reference genome includes GRCh38.

[0014] This invention provides a method for analyzing amplicon sequencing (AmpliSeq) data based on the Ion Torrent platform, comprising the following steps: sequencing amplicon sequences using the Ion Torrent platform to obtain amplicon sequencing data; performing quality control on the amplicon sequencing data through multi-parameter filtering to ensure data reliability; performing sequence alignment on the quality-controlled data using the TMAP alignment tool of the Torrent Suite system, integrating Bowtie short, Bowtie long, and SSAHA long algorithms, comprehensively evaluating the alignment results of different algorithms, and automatically selecting the optimal alignment result. Integrating multiple algorithms not only enables efficient alignment but also improves localization accuracy; evaluating the amplicon coverage of the aligned target region using the Coverage Analysis plugin of the Torrent Suite system to obtain amplicon coverage analysis data for evaluating the target region capture efficiency; accurately detecting SNPs and Indel variants using Torrent Variant Caller combined with diBayes and GATK algorithms; and using Ion... The Reporter (IR) system uploads the variant site information, automatically links it to a database, and performs biological function annotation on the variants to obtain annotation results. The annotation results are then filtered to select clinically relevant variants and remove germline mutations and low-quality variants, resulting in clinically relevant variant data. In summary, this invention, based on the TorrentSuite (TS) and Ion Reporter (IR) systems of the Ion Torrent platform, significantly reduces the risk of false positives and false negatives and improves the accuracy of data analysis through rigorous quality control, multi-algorithm integration and comparison, in-depth coverage analysis, accurate variant detection, and automated annotation. Therefore, the amplicon sequencing data analysis method of this invention can significantly improve the accuracy of data analysis, clinical interpretation efficiency, and reporting standardization, and is particularly suitable for genetic disease and tumor mutation detection scenarios. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of the entire process from amplicon sequencing to data analysis results generated using the Ion Torrent platform in Example 1; Figures 2-3 This is a graph showing the total number of addressable micropores containing ISPs and the percentage of active ISPs in Example 1; Figure 4 This is a diagram illustrating the standard alignment format BAM file in Example 1; Figure 5 The results of the standard comparison in Example 1 are displayed graphically. Figure 6 This is a schematic diagram of amplicon coverage analysis in Example 1; Figure 7 This is a schematic diagram illustrating the uniformity of amplicon sequencing in Example 1; Figure 8 The flowchart of the Torrent Variant Caller algorithm in Example 1 is shown below. Figure 9 This is a diagram illustrating the Torrent Variant Caller parameter settings in Example 1; Figure 10 This is a diagram showing the TVC results in Example 1; Figure 11 This is a visualization of the variant sites in Example 1; Figure 12 This is a diagram illustrating the variant sites in Example 1; Figure 13 This is a diagram illustrating the functional annotation information in Example 1; Figure 14 This is a diagram illustrating the group frequency information in Example 1; Figure 15 This is a diagram showing the information of Ontologies in Example 1; Figure 16 This is a diagram illustrating pharmacogenomics information from Example 1; Figures 17-18 This refers to the filter built into the Ion Reporter in Example 1; Figure 19 This is a template for the analysis parameter report generated by the Ion Reporter system in Example 1; Figures 20-23 This is a schematic diagram of the detection results generated by the Ion Reporter system in Example 1; Figure 24 This is a diagram showing the quality control results of the sequencing data of the positive sample in Example 3; Figure 25 A diagram illustrating the filter condition settings in Example 3; Figure 26 This is a display of the mutation results after the filtering step in Example 3; Figure 27 This is a demonstration of the mutation results without the filtering step in Example 3. Detailed Implementation

[0017] This invention provides a method for amplicon sequencing data analysis based on the Ion Torrent platform, comprising the following steps: 1) The amplicon sequencing data were obtained by sequencing the amplicon using the Ion Torrent platform; 2) The amplicon sequencing data is filtered according to the chip operating quality parameters to obtain sequencing data that meets the quality requirements. The chip operating quality parameters include multiple parameters such as the total number of microwells on the chip, bead loading rate, proportion of available beads, proportion of beads carrying the test template, proportion of multi-template beads, and proportion of low-quality beads. 3) The TMAP alignment tool of the TS system is used to perform sequence alignment on the quality-controlled sequencing data. The Bowtieshort, Bowtie long, and SSAHA long algorithms are integrated to select the optimal alignment result and locate it to the human reference genome, generating a sequence alignment file in BAM format. 4) The amplicon coverage of the target region in the sequence alignment file is evaluated using the Coverage Analysis plugin of the TS system to obtain amplicon coverage analysis data; the Coverage Analysis plugin performs statistical analysis on the coverage of the target region based on the BAM format alignment file generated by TMAP and the BED file of the target region. 5) Using Torrent Variant Caller (TVC) in combination with diBayes and GATK algorithms, the human reference genome, AR panel target region BED file and hotspot variant VCF file are pre-set in the TorrentSuite system. Then, the Torrent Variant Caller calls the diBayes and GATK algorithms to perform variant detection on the BAM file generated by TMAP, identify SNP and Indel variants respectively, and finally output variant site information in VCF format. 6) Upload the variant site information through the Ion Reporter system, automatically link it to the database, annotate the biological functions, and obtain the annotation results; 7) Filter the annotation results to screen for clinically relevant variants and remove germline mutations and low-quality variants to obtain clinically relevant variant data; the filter chain set for filtering the annotation results includes mutation coverage ≥250X, minimum allele frequency ≥5%, and exclusion of common SNPs and homopolymer interference.

[0018] This invention first uses the Ion Torrent platform to sequence amplicon sequences, obtaining amplicon sequencing data.

[0019] In one implementation, the average sequencing depth of the target region is not less than 100×; in another implementation, the average sequencing depth of the target region is 100×~10000×, and can further be 200×~5000×. For example, the average sequencing depth for germline variation detection in genetic diseases typically requires ≥100×, but can also be 200~500×; the average sequencing depth for somatic mutation detection in tumors is typically 500×~5000×; and the average sequencing depth for liquid biopsy (ctDNA) can reach over 10000×. In this invention, a higher sequencing depth helps improve the detection capability of low-frequency variations and the reliability of detection results.

[0020] After obtaining amplicon sequencing data, this invention also includes filtering the amplicon sequencing data according to chip operating quality parameters to obtain sequencing data that meets quality requirements. The quality control parameters of this invention include multiple parameters selected from the following: total number of microwells on the chip, bead loading rate, usable bead ratio, bead ratio carrying test template, multi-template bead ratio, low-quality bead ratio, and primer dimer bead ratio. As one implementation, the bead loading rate is higher than 60%; the usable bead ratio is ≥90%; the bead ratio carrying test template is ≥90%; the multi-template bead ratio is not higher than 50%; the low-quality bead ratio is less than 20%; and the primer dimer bead ratio is ≤10%.

[0021] As one implementation method, amplicon sequencing data filtering is based on Thermo Fisher's quality control standards; this invention filters amplicon sequencing data based on the quality control indicators recommended in the official technical documentation of the Ion Torrent platform. These official technical documents include, but are not limited to, the Ion Torrent S5 System User Guide and the Torrent Suite Software User Guide; ISP preparation and quality control are set according to the parameter requirements in the Ion Sphere Quality Control Documentation, Thermo Fisher Technical Note, and Thermo Fisher Application Note. The quality control parameters described in this invention refer to the official quality evaluation indicators of the Ion Torrent platform and are set in conjunction with the amplicon sequencing application scenario. By filtering sequencing data through quality control, the impact of low-quality data on subsequent sequence alignment, coverage analysis, variant detection, and functional annotation can be effectively reduced, thereby improving the reliability and reproducibility of the analysis results. This invention ensures data quality and reduces the impact of low-quality data on subsequent analyses by setting strict quality control parameters.

[0022] After obtaining sequencing data that meets quality requirements, this invention uses the TMAP alignment tool in the TS system to perform sequence alignment on the quality-controlled sequencing data. It integrates Bowtie short, Bowtie long, and SSAHA long algorithms, selecting the optimal alignment result to locate the human reference genome and generating a BAM format sequence alignment file. This invention uses the TMAP alignment tool in the TS system to perform sequence alignment on qualified sequencing data. This tool integrates Bowtie short, Bowtie long, and local alignment algorithms, selecting the optimal result through multiple alignment strategies, thereby achieving precise localization of the human reference genome. Compared to conventional alignment tools, it is more compatible with the Ion Torrent sequencing platform, exhibiting higher consistency and accuracy.

[0023] In this invention, qualified data is saved in the universal file format BAM, and subsequent data analysis by TS will be based on the BAM file. Meanwhile, the TS system provides a file exporter plugin that can automatically or manually convert BAM files to other universal formats, such as FASTQ.

[0024] In one implementation, the human reference genome includes GRCh38 (the USCS version of hg38). Compared to the previous version of the reference sequence GRCh37 (hg19), GRCh38 fills more than 100 gaps and closes 12 gaps in the GRCh38 reference assembly using ultralong reads and nanopore sequencing, which allows for the design of primers for a small number of SNPs associated with allergic rhinitis (AR).

[0025] As one implementation method, this invention uses TMAP (Torrent Mapper) to locate each sequenced read on the human reference genome, obtaining the read's position on the genome. In this invention, TMAP is specifically optimized for the characteristics of Ion Torrent data, and its core includes three popular alignment algorithms: Bowtie short, Bowtie long, and SSAHA long. TMAP is optimized for the data characteristics of the Ion Torrent sequencing platform, integrating multiple alignment algorithms such as Bowtie short, Bowtie long, and SSAHA long. By comprehensively evaluating the alignment results of different algorithms, it automatically selects the optimal alignment result and outputs a standard BAM format file. The BAM file can be viewed and analyzed using visualization software such as IGV. By adopting the TMAP multi-algorithm integrated alignment strategy, the localization accuracy of the sequenced sequence in the reference genome can be improved, providing reliable data support for subsequent variant detection and result analysis.

[0026] After obtaining the generated BAM format sequence alignment file, this invention analyzes the coverage of the target region using the Coverage Analysis plugin of the TS system to obtain amplicon coverage analysis data. In this invention, the Coverage Analysis plugin can directly read the BAM format sequence alignment file and target region file generated by TMAP, perform statistical analysis on the target region coverage, and output indicators such as the on-target rate, mean sequencing depth, uniformity of base coverage, and coverage of each amplicon, thereby evaluating the capture effect and sequencing quality of the target region. As one implementation method, the coverage analysis parameters of this invention include the on-target rate, mean sequencing depth, and uniformity of base coverage, wherein the on-target rate is not less than 90%, the mean sequencing depth is not less than 100X, and the uniformity of base coverage is not less than 90%.

[0027] After obtaining amplicon coverage analysis data, this invention utilizes Torrent Variant Caller (TVC) combined with diBayes and GATK algorithms, inputting the human reference genome, AR panel target region file, and hotspot file, to detect SNPs and Indels respectively, and then outputs variant site information by merging them.

[0028] In one implementation, the Indel includes small indel and longer indel.

[0029] In one implementation, the variant site information includes genotype, frequency, type, and hotspot markers.

[0030] This invention employs (TVC) combined with diBayes and GATK algorithms to distinguish between SNPs and Indels, and combines AR panel specific parameters to improve the sensitivity of variant detection.

[0031] After obtaining the variant site information, this invention uploads the variant site information through the Ion Reporter system, automatically associates it with the database, annotates the biological functions, and obtains the annotation results.

[0032] In one implementation, the database includes ClinVar, dbSNP, and drug targets.

[0033] In one implementation, the biological functions include the clinical significance of the variation, population frequency, amino acid changes, and pharmacogenomics information.

[0034] This invention uses an IR system to integrate databases such as ClinVar, dbSNP, and pharmacogenomics, providing rich biological function annotations.

[0035] After obtaining the annotation results, the present invention filters the annotation results, screens clinically relevant variants, and removes germline mutations and low-quality variants to obtain clinically relevant variant data; the filter chain set for filtering the annotation results includes mutation coverage ≥250X, minimum allele frequency ≥5%, and exclusion of common SNPs and homopolymer interference.

[0036] This invention enables rapid screening of clinically relevant variants using custom filters (filter chains).

[0037] After obtaining clinically relevant variant data, the present invention also includes automatically integrating key annotation information based on the clinically relevant variant data to generate a standardized technical report containing details of variant sites.

[0038] In one implementation, the key annotation information includes gene regions, pathogenicity assessments, and drug targets.

[0039] As one implementation method, the standardized technical report is intended for clinical interpretation.

[0040] This invention automatically generates clinical reports containing key biological information based on a standardized process, improving analysis efficiency and consistency.

[0041] The amplicon sequencing data analysis method based on the Ion Torrent platform described in this invention has the following advantages: 1) High-reliability data quality control: By setting strict quality control parameters (such as Live ISP ratio ≥90%, sequencing depth 500X), data quality is ensured and the impact of low-quality data on subsequent analysis is reduced.

[0042] 2) Optimized alignment efficiency: Integrating Bowtie short, Bowtie long and SSAHA long algorithms, the optimal alignment strategy is selected for reads of different lengths to improve positioning accuracy.

[0043] 3) Precise variant detection: TVC (Torrent Variant Caller) combined with diBayes and GATK algorithms is used to distinguish between SNPs and Indels. Combined with AR panel specific parameters, the sensitivity of variant detection is improved.

[0044] 4) Comprehensive annotation and intelligent filtering: The IR system integrates databases such as ClinVar, dbSNP, and pharmacogenomics to provide rich biological function annotations and quickly screens clinically relevant variants through custom filters (such as mutation coverage ≥250X, minimum allele frequency ≥5%).

[0045] Automated report generation: Based on standardized processes, clinical reports containing key biological information are automatically generated, improving analysis efficiency and consistency.

[0046] To further illustrate the present invention, the following detailed description of an amplicon sequencing data analysis method based on the IonTorrent platform, in conjunction with the accompanying drawings and embodiments, is provided. However, these descriptions should not be construed as limiting the scope of protection of the present invention.

[0047] Example 1 A method for analyzing amplicon sequencing data based on the Ion Torrent platform, the flowchart of which is shown below. Figure 1 As shown, the steps are as follows: 1. Sequencing data quality control The Ion Torrent platform was used to sequence amplicon samples, and the sequencing quality directly affects the results of subsequent data analysis. Following ThermoFisher and internationally accepted NGS data quality control requirements (including Illumina and industry-standard sequencing quality control specifications), this embodiment performed a multi-parameter filtering process after the data was processed to rigorously screen the raw signals.

[0048] Key quality control parameters are as follows: Total address wells: This indicates the total number of microwells on the chip. This parameter is a fixed value and depends on the chip type. Wells with ISP: This indicates the number and proportion of microbeads successfully loaded in the microwells. This parameter is above 60%. Live ISP: This indicates the number and proportion of usable microbeads. This parameter is above 90%. Live Library: This indicates the number and proportion of microbeads carrying the test template. This parameter is above 90%. Polyclonal: This indicates the number and proportion of multi-template microbeads (multiple test templates on one microbead). This parameter is below 50%. Primer Dimer: This indicates the number and proportion of primer dimer microbeads. This parameter is below 10%. Low Quality: This indicates the number and proportion of microbeads with low sequencing quality. This parameter is below 20%.

[0049] After the above filtering, qualified data are saved in BAM format (Binary Sequence Alignment File) for subsequent alignment and variant analysis. In addition, the Torrent Suite (TS) system's built-in File Exporter plugin can convert BAM files to common formats such as FASTQ to meet different analytical needs.

[0050] An ISP Summary report was exported using Torrent Suite software to perform quality control statistics on the distribution of ion-associated microspheres (ISPs) on the chip. Statistical indicators included the total addressable wells with ISPs and the percentage of live ISPs, to assess the effective reaction unit ratio of this sequencing run. The results are illustrated in the diagram below. Figures 2-3 As shown.

[0051] 2. Perform sequence alignment based on the quality control data from step 1. After the sequencing data passes quality control in step 1, the TS system will automatically perform alignment analysis based on the reference genome. The reference sequence for the AR panel is GRCh38 (i.e., the USCS version of hg38). Using its built-in alignment software TMAP (Torrent Mapper), each sequenced read is mapped to the genome, revealing its location. A schematic diagram is shown below. Figure 4 As shown. Figure 4 The contents of the BAM file are displayed. The first column, Read Name, indicates the name of the reads; the second column, FLAG, indicates the forward or reverse alignment of the reads on the genome, with 0 indicating forward alignment and 16 indicating reverse alignment; the third column, Reference, indicates the reference chromosome to which the reads were aligned; and the fourth column, Position, indicates the starting alignment position of the reads on the reference chromosome. Figure 5 The results of the alignment are displayed, including the location on the genome and the region of the gene in which it is located.

[0052] 3. Perform amplicon coverage analysis based on the alignment results from step 2. Ampliseq technology uses multiplex PCR to amplify a target region for sequencing. Therefore, amplification efficiency, i.e., the coverage of the target region, is a crucial technical indicator. The Coverage Analysis Plug-in of the TS system evaluates and analyzes aspects such as capture rate and uniformity of amplicon sequencing results. The main parameters are as follows: Percent reads on target: This parameter indicates the percentage of reads obtained from amplification and sequencing of the target region. It reflects the specificity of amplicon sequencing. A reading above 90% is desirable.

[0053] Average base coverage depth: This parameter reflects the average sequencing depth of the sample. A suitable sequencing depth depends on the purity of the sample. For applications detecting germline mutations in blood samples, this value should be higher than 100X; for cancer tissue samples mixed with normal cells, applications detecting somatic mutations, this value should be greater than 250X. Peripheral blood, umbilical cord blood, (dried blood smears), saliva, and oral swabs are used for germline mutation detection. DNA extracted from whole blood samples has high purity, while the purity and concentration of DNA from saliva, oral swabs, and other samples vary. Therefore, the sequencing depth is set at 500X.

[0054] Uniformity: The uniformity of amplicon sequencing. This parameter reflects the consistency of multiplex PCR efficiency. The parameter value is typically above 90%.

[0055] A schematic diagram of amplicon coverage analysis is shown below. Figure 6 As shown in the diagram, the uniformity analysis results of amplicon sequencing are illustrated below. Figure 7 As shown.

[0056] 4. Detect variant sites based on the coverage analysis data from step 3. Targeting the characteristics of Ion Torrent data, the Ion Torrent technology platform integrates the variant detection software TVC (Torrent Variant Caller), whose algorithm flow is as follows: Figure 8 As shown: TVC has been integrated into the AR analysis platform, which can run automatically. Inputs include the reference genome (hg38), the AR panel target region location file (bed), the parameter file, and the AR panel hotspot file. Figure 9 Different algorithms, diBayes, GATK Genotyper, and GATK read walker, are used to detect SNPs, small indels, and longer indels, respectively. The results of SNPs and indels are then combined as the output of the variant call, as illustrated in the diagram. Figure 10 In addition, parameter options were set based on the specific characteristics of the AR panel. The target area location file and the AR panel hotspot file are protected by this patent.

[0057] The TVC will indicate the location of the mutation (e.g.) Figure 10 In the data analysis, chr4:1807894, the genotype of the variant (G->A, Hom or Het), the frequency, reliability, type (SNP or Indel), whether it is a hotspot, and the gene in which it is located can be displayed. When performing data analysis manually, IGV can be used to display information such as the mutation and the location of the variant on the gene (intron or exon), as shown in the illustration. Figure 11 As shown.

[0058] 5. Perform biological function annotation based on the variation data from step 4. The data analysis workflow based on Ion Reporter allows data to be transferred from the TS system to the IR system via the Ion Reporter Uploader plugin. Within the IR system, selecting the appropriate analysis workflow automatically completes the necessary data processing steps, yielding rich biological functional annotation information.

[0059] Rich annotation information diagrams, such as Figures 12-16 As shown (only partial data is provided for demonstration purposes), the annotation information includes the variant sites ( Figure 12 Functional annotation information ( Figure 13 ), group frequency information ( Figure 14 ExAC is divided into multiple subgroups (LAF, EAAF, OAF, EFAF, SAAF, ENFAF, AAF, and GAF), and Ontologies information ( Figure 15 Important information such as pharmacogenomics Figure 16 ),in Figures 14-16 To facilitate the presentation of the content, the original Excel spreadsheet has been integrated and displayed.

[0060] 6. Apply a smart filter based on the annotation results from step 5. IR provides powerful analytical filters that allow for arbitrary combinations and filtering of different biological functional information, enabling analysts to quickly and flexibly filter data and interpret data diagrams, such as... Figures 17-18 As shown.

[0061] The analysis parameters are as follows: Software: ThermoFisher Ion Reporter 5.10. Workflow version: 2.0. Increase Indel detection capability and SNP quality score by modifying the "Mutation Lookup" settings as follows: Mutation Coverage >= 250X Indel FuncSize offset = 0Kmer Len = 11 indel min variant score = 10 Indel min allele frequency = 0.05 Min Indel Size = 1, Min Var Freq = 0.05 SNP = TRUE local read realignment SNP min_variant_score = 20 SNP min allele frequency = 0.05 Additionally, the following filter chain selects variables for calculating TMB: Mutation Coverage >= 250X Mutation Type = SNV or Indel 0.0 <= Minor Allele Frequency <= 1.0E-6 (This filter uses 1,000 genome dbSNPs) 0.0 <= 5000 Exomes Global MAF (20161108) <= 1.0E-6 0.0 <= ExAC GAF (1) <= 1.0E-6 UCSC common SNPs = not in 0 <= homopolymer length <= 40.05 < allele frequency <= 1.0 database filter removes possible germline mutations. The UCSC filter removes mutation hotspots to avoid bias introduced by the presence of many frequently mutated oncogenes in this group. The "homopolymer" filter removes poor-quality variants. Finally, a minimum allele frequency of 5% guarantees 100% detection of variants with an allele frequency of at least 10%.

[0062] 7. Generate standardized clinical reports After being interpreted and reviewed by analysts, the IR system can automatically generate standardized reports, as illustrated in the diagram below. Figure 19 The displayed template shows a report containing information on the selected site and its associated biological function. An example of the detection results generated by the Ion Reporter system is shown below. Figures 20-23 As shown.

[0063] The Ion Reporter has a rich collection of biological knowledge bases, as shown in Table 1.

[0064] Table 1. Ion Reporter's built-in rich biological knowledge base

[0065] Example 2 The accuracy verification of the amplicon sequencing data analysis method follows these steps: To verify the accuracy of the amplicon sequencing data analysis method of this invention, 54 independent clinical samples (peripheral blood and saliva samples), a positive control (NA13805, HapMap reference DNA from Coriell Institute for Medical Research), a negative control (NA12878, HapMap reference DNA from Coriell Institute for Medical Research), and a template-free control (NTC) were analyzed. All samples underwent library preparation, template preparation, Ion Torrent platform sequencing, and bioinformatics analysis according to the method described in Example 1 of this invention.

[0066] (1) The accuracy of clinical sample variation detection is shown in Table 2.

[0067] Table 2. Results of variant detection in clinical samples

[0068] A total of 11,298 variant sites were detected in 54 samples. After screening by OncoKB (OKB) and manual verification, 61 SNV / INDEL variants with clinical significance or requiring key verification were selected for orthogonal confirmation, and each sample contained at least one variant confirmed by orthogonal confirmation.

[0069] The variant validation results are shown in Table 3. All 61 variants were confirmed by orthogonal validation, with 54 true positives for SNV and 7 true positives for INDEL. No false positives or false negatives were found. The positive concordance rate (PPA) and positive predictive value (PPV) for both SNV and INDEL reached 100%. Furthermore, the allele frequency (AF) of the detected SNV variants ranged from 0.4128 to 0.5500, with an average of 0.4902; the AF range for INDEL variants ranged from 0.4100 to 0.5400, with an average of 0.5006. The average sequencing depth was 741×, indicating that the method of this invention can stably and accurately detect variants at high sequencing depths.

[0070] Table 3. Results of variant validation

[0071] (2) Validation of positive control samples The standard positive control NA13805 was used for repeated testing according to the method in Example 1, and the results are as follows: Mapped Reads: 570,496~952,056, average 783,750; On Target: 98.80%~99.19%, average 98.95%; Mean Depth: 544×~905×, average 751×; Uniformity: 95.60%~97.42%, average 96.74%.

[0072] The results show that the method of the present invention can stably detect target variants in positive control samples, and has good repeatability and sequencing quality.

[0073] (3) Validation of negative control samples The standard negative control NA12878 was tested according to the method in Example 1, and the results are as follows: Mapped Reads: 747,290~1,500,957, average 1,007,024; On Target: 99.00%~99.15%, average 99.08%; Mean Depth: 713×~1473×, average 984×; Uniformity: 91.39%~97.83%, average 95.64%.

[0074] The results show that the method of the present invention has stable data analysis performance and can be used for variation analysis of high-quality sequencing data.

[0075] (4) Template-free control (NTC) verification Using a DNA template-free control, the detection was performed according to the method in Example 1, and the results showed: The average number of Mapped Reads is only 2358; The average sequencing depth is only 0.54×; No valid sequencing data for the target region were generated.

[0076] This demonstrates that the method of the present invention can effectively distinguish between real samples and negative controls, and no false positive results were observed due to obvious background interference or contamination, exhibiting good analytical specificity.

[0077] Example 3 The positive sample PC-DNA-M_v493 (Seraseq) ®Myeloid Mutation DNA Mix (catalog number: 0710-0408, sourced from SeraCare) highlights the importance of the "filtering" step in amplicon sequencing data analysis method for sample validation. The steps are as follows: 1) The positive sample PC-DNA-M_v493 underwent library preparation, template preparation, Ion Torrent platform sequencing, and bioinformatics analysis according to the method described in Example 1 of this invention. The sequencing quality control results are as follows: Figure 24 As shown.

[0078] After applying the preset filter conditions (Filtered, filter settings are as follows) Figure 25 As shown in the figure, the system filters out low-quality variants, false positive sites, and other variants that do not meet the screening criteria, ultimately retaining 32 variants. These 32 variants are the true positive variants obtained through quality control and filtering strategies, and can be used as the basis for subsequent analysis and report generation. The results are as follows. Figure 26 As shown.

[0079] 2) The positive sample PC-DNA-M_v493 underwent library preparation, template preparation, Ion Torrent platform sequencing, and bioinformatics analysis according to the method described in Example 1 of this invention. The only difference from step 1) is that the filtering analysis in step 5 is not performed. The results are as follows: Figure 27 As shown, without setting a filter (No Filter), a total of 2,646 variants were detected in this positive control sample. Figure 27 The top right corner shows "Filtered In Variants". The results at this point include true variance as well as false positives and low-quality variances caused by various error factors, and therefore cannot be directly used as the final analysis result.

[0080] Although the above embodiments have provided a detailed description of the present invention, they are only some embodiments of the present invention, and not all embodiments. People can obtain other embodiments based on these embodiments without creative effort, and these embodiments all fall within the protection scope of the present invention.

Claims

1. A method for analyzing amplicon sequencing data based on the Ion Torrent platform, characterized in that, Includes the following steps: 1) The amplicon sequencing data were obtained by sequencing the amplicon using the Ion Torrent platform; 2) Perform quality control on amplicon sequencing data, filter the sequencing data according to the chip operation quality parameters, and obtain sequencing data that meets the quality requirements; The chip operating quality parameters include multiple parameters from the following: total number of micropores on the chip, microbead loading rate, proportion of available microbeads, proportion of microbeads carrying the test template, proportion of multi-template microbeads, proportion of low-quality microbeads, and proportion of primer dimer microbeads. 3) The TMAP alignment tool of the Torrent Suite system is used to perform sequence alignment on the quality-controlled sequencing data. The Bowtie short, Bowtie long and SSAHA long algorithms are integrated to select the best alignment result to locate the human reference genome and generate a sequence alignment file in BAM format. 4) The coverage of the target region in the sequence alignment file is analyzed using the Coverage Analysis plugin of the Torrent Suite system to obtain amplicon coverage analysis data; the Coverage Analysis plugin performs statistical analysis on the coverage of the target region based on the BAM format alignment file generated by TMAP and the BED file of the target region. 5) Using Torrent Variant Caller in combination with diBayes and GATK algorithms, the human reference genome, AR panel target region BED file and hotspot variant VCF file are pre-set in the Torrent Suite system. Then, using Torrent Variant Caller to call diBayes and GATK algorithms, variant detection is performed on the BAM file generated by TMAP, identifying SNP and Indel variants respectively, and finally outputting variant site information in VCF format. 6) Upload the variant site information through the Ion Reporter system, automatically link it to the database, annotate the biological functions, and obtain the annotation results; 7) Filter the annotation results, screen for clinically relevant variants, and remove germline mutations and low-quality variants to obtain clinically relevant variant data; The filter chain for filtering the annotation results includes mutation coverage ≥250X, minimum allele frequency ≥5%, and exclusion of common SNPs and homopolymer interference.

2. The amplicon sequencing data analysis method according to claim 1, characterized in that, After obtaining clinically relevant variant data, the system also includes automatically integrating key annotation information based on the clinically relevant variant data to generate a standardized technical report containing details of the variant sites.

3. The amplicon sequencing data analysis method according to claim 1, characterized in that, The average sequencing depth of the target region is not less than 100×.

4. The amplicon sequencing data analysis method according to claim 1, characterized in that, The Indel includes smallindel and longer indel.

5. The amplicon sequencing data analysis method according to claim 1, characterized in that, The variant site information includes genotype, frequency, type, and hotspot markers.

6. The amplicon sequencing data analysis method according to claim 1, characterized in that, The database includes ClinVar, dbSNP, and drug targets.

7. The amplicon sequencing data analysis method according to claim 1, characterized in that, The biological functions include the clinical significance of the variation, population frequency, amino acid changes, and pharmacogenomics information.

8. The amplicon sequencing data analysis method according to claim 1, characterized in that, The key annotation information includes gene regions, pathogenicity assessments, and drug targets.

9. The amplicon sequencing data analysis method according to claim 1, characterized in that, The amplicon sequencing data was filtered based on Thermo Fisher quality control standards: bead loading rate ≥60%, usable bead ratio ≥90%, bead ratio with test template ≥90%, multi-template bead ratio ≤50%, primer dimer bead ratio ≤10%, and low-quality bead ratio ≤20%.

10. The amplicon sequencing data analysis method according to claim 1, characterized in that, The human reference genome includes GRCh38.

Images