vardict

vardict calls variants from next-generation sequencing (NGS) data directly from BAM files, detecting single nucleotide variants (SNVs), multi-nucleotide variants (MNVs), insertions and deletions (InDels), complex variants, and structural variants in DNA- and RNA-sequencing while supporting single and paired sample analyses for cancer genomics.

Key Features:

• Versatility: Processes DNA- and RNA-sequencing data and detects SNVs, MNVs, InDels, complex variants, and structural variants.
• BAM input and sample support: Performs single and paired sample variant calling directly from BAM files.
• Amplicon bias-aware variant calling: Incorporates amplicon bias-aware methods to mitigate PCR artifacts in PCR-based targeted sequencing experiments.
• Realignment for accuracy: Performs dynamic local realignments and rescues long indels by realigning BWA soft-clipped reads to improve allele frequency estimation.
• Scalability and performance: Scales linearly with sequencing depth to enable ultra-deep sequencing applications relevant to tumor evolution and circulating tumor DNA detection.
• Paired sample analysis: Detects somatic variant differences and loss of heterozygosity between paired samples.

Scientific Applications:

• Clinical cancer genomics: Reprocessing of The Cancer Genome Atlas (TCGA) Lung Adenocarcinoma dataset with VarDict identified known driver mutations in KRAS, EGFR, BRAF, PIK3CA, and MET in 16% more patients than previous calls.
• Tumor evolution and circulating tumor DNA: Enables ultra-deep sequencing analyses for studying tumor clonal dynamics and detecting circulating tumor DNA in blood samples.

Methodology:

Performs variant calling from BAM files with amplicon bias-aware calling, dynamic local realignments, and rescue of long indels via realignment of BWA soft-clipped reads, supporting single and paired sample comparisons.