Publications

Next-generation sequencing (NGS) technologies have significantly expanded clinical genetic testing for inherited conditions and diseases like cancer, with accurate variant calling being a crucial step that underpins downstream analysis and interpretation. The software tools and approaches for detecting sequence variants in clinical samples have evolved significantly, much like NGS technologies, in the last decade. Variant calling is a crucial step in analysing NGS data, as it enables the identification of genomic variations such as single nucleotide polymorphisms (SNPs), insertions, and deletions. The Genome Analysis Toolkit (GATK) is a widely used software suite that provides a range of tools for variant calling from NGS data. GATK uses a probabilistic model to identify variants while accounting for sequencing errors, mapping errors, and other sources of noise. The GATK workflow for variant calling typically involves several preprocessing steps, including mapping the reads to a reference genome, marking duplicate reads, and recalibrating the base quality scores. These preprocessing steps help improve the accuracy of variant calling. The core variant calling algorithm in GATK is called HaplotypeCaller, which uses local de novo assembly to identify haplotypes (i.e., sets of closely linked variants) in the reads. UnifiedGenotyper, a previous variant-calling tool in GATK, has been replaced by HaplotypeCaller due to limitations, including challenges in detecting indels, handling complex variants, and performance issues related to computational resources and limited scalability. HaplotypeCaller then uses these haplotypes to improve the accuracy of variant calling, particularly for regions of the genome with complex variants, such as indels. After variant calling, the resulting variant call format (VCF) file can be further processed and filtered using GATK tools, such as VariantFiltration, to remove low-quality variants and reduce false positives. GATK also provides tools for variant annotation, such as VariantAnnotator and SnpEff, which can provide additional information about the functional effects of variants, including their effects on genes and proteins. Overall, GATK provides a powerful and flexible toolkit for variant calling from NGS data. However, effective use of GATK requires expertise in bioinformatics and careful attention to quality control and filtering to minimise false positives and ensure the accuracy of variant calls. Nonetheless, GATK's popularity and wide adoption in the scientific community attest to its usefulness and effectiveness for variant calling from NGS data.