MuTect

MuTect identifies somatic point mutations in next-generation sequencing (NGS) data from cancer genomes, emphasizing detection of low-allelic-fraction variants.

Key Features:

• Bayesian classifier: Employs a Bayesian classifier to detect somatic mutations, enabling calls from very low numbers of supporting reads.
• High-specificity filters: Applies carefully tuned filters to distinguish true somatic mutations from sequencing errors and other artifacts, preserving specificity.
• Benchmarking on real data: Evaluates performance using real sequencing data rather than simulations to assess sensitivity and specificity.
• Sensitivity at low allelic fractions: Demonstrates higher sensitivity for mutations with allelic fractions as low as 0.1 or below, addressing tumor heterogeneity and normal-cell contamination.

Scientific Applications:

• Tumor heterogeneity and subclone analysis: Detects low-frequency somatic variants to support characterization of cancer subclones and their evolutionary dynamics.
• Study of treatment resistance and disease progression: Enables identification of rare mutations relevant to treatment resistance and tumor evolution.
• Exome and genome sequencing analyses: Applied to standard exome and genome NGS data for comprehensive somatic mutation detection in cancer genomes.
• Low-allelic-fraction variant discovery: Facilitates discovery of rare variants that may be missed by methods less sensitive to low allelic fractions.

Methodology:

Implements a Bayesian classifier to call somatic mutations, applies tuned specificity filters, and benchmarks performance on real sequencing data across sequencing depth, base quality, and allelic fraction.