Systematic Comparison of SNV Callers Based on Single-cell Sequencing Data
Open AccessSingle-cell sequencing is a novel method that sheds light on many currently challenging biological and medical problems. For example, the variants called from single-cell genomics and transcriptomics data can be utilized to reconstruct lineage tracing; the differentially expressed genes between normal and tumor tissues can be identified with single-cell transcriptomics data. The regular single-cell sequencing comprises single-cell WGS sequencing, single-cell WES sequencing, single-cell transcriptomics sequencing and single-cell proteomics. Single-cell ATAC sequencing data are special category of single-cell genomics data. Single-cell ATAC sequencing extracts all of open chromatin regions in a cell and generates the profile of the open chromatin regions in a cell. Genomic profiles of open chromatin regions (OCRs) are a precious asset to the biological and clinical research. The genomic profiles of OCRs can reveal all regulatory genomic regions in a cell, so it can help to understand the gene regulation at single-cell level. Especially, the SNVs identified in the OCRs in cells can be used in downstream research in many ways. For example, the SNVs detected in the OCRs can help find the mechanism of the abnormal gene expression in a disease condition; the variants in OCRs can also be used to reconstruct the cellular lineage. Despite of the research significance of the SNVs in scATAC data, almost no variant callers have been developed for single-cell sequencing data. Of note, even less previous research can be found about calling variants from single-cell ATAC data. Therefore, a pioneering benchmarking research is conducted to compare the variant calling performance of several SNV callers (BCFtools, GATK Mutect2 and strelka2) based onsingle-cell ATAC data. The three metrics employed to benchmark the performance of variant callers are precision, recall and F-score. Meanwhile, time efficiency of three SNV callers are also compared. After comparison, due to strelka2’s highest F-score metric, strelka2 is determined to be the most recommended SNV caller for most labs when they are dealing with single-cell sequencing data. On the other hand, BCFtools has unique advantages of time-efficiency and highest recall rate, which renders BCFtools the first choice in SNV discovery research that prioritizes time efficiency and recall.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.