DeepVariant is a deep learning germline variant caller that turns aligned sequencing reads into a VCF of SNVs and small indels. This page covers what it does, how to run it, how to read its output, and how to measure its accuracy on your own data.
A step-by-step WGS pipeline for a single human genome: alignment with bwa-mem2, variant calling with DeepVariant, structural variants with Manta, annotation with VEP, and the QC numbers that tell you whether any of it worked.
A working pipeline for analyzing the SNPs in your own whole-genome sequence: alignment, variant calling with DeepVariant, QC thresholds that catch real failures, annotation, and the queries worth running first.
A step-by-step pipeline from raw FASTQ to a filtered, benchmarked VCF for a single human genome, with the tool choices, parameters, and quality checks we would use.
A practical comparison of DeepVariant, GATK HaplotypeCaller, bcftools mpileup, Strelka2, and Clair3 for calling SNPs from your own whole-genome data, with commands, benchmarks, and the failure modes that matter.
A working end-to-end guide to going from FASTQ to a filtered, benchmarked VCF on a personal 30x human genome, with the specific tools, flags, reference files, and failure modes we would use.