All-FIT

All-FIT estimates tumor purity from allele-frequency data derived from high-depth targeted sequencing to enable discrimination of somatic mutations from germline variants in clinical sequencing.

Key Features:

• Iterative Weighted Least Squares Methodology: Employs an iterative weighted least squares algorithm that leverages allele frequencies of detected variants to impute and refine tumor purity.
• High-Depth Sequencing Data Utilization: Tailored for high-depth targeted sequencing data to exploit extensive variant allele frequency information for precise purity estimates.
• Enhanced Accuracy and Performance: Validation on simulated and clinical datasets demonstrated improved accuracy and performance compared with existing computational methods.
• Adaptability to Genomic Heterogeneity: Handles genomic heterogeneity and sub-clonal mutations when estimating tumor purity.
• Interpretation Based on Tumor Biology: Produces purity estimates intended to be interpreted in the context of expected tumor biology.

Scientific Applications:

• Clinical Sequencing: Supports clinical sequencing workflows that require distinguishing somatic mutations from unfiltered germline variants for diagnosis and treatment planning.
• Research on Tumor Heterogeneity: Provides reliable purity estimates for studies of tumor heterogeneity and subclonal architecture.
• Development of Computational Methods: Serves as a benchmark and reference for developing and evaluating new tumor purity estimation approaches.

Methodology:

Applies an iterative weighted least squares algorithm to allele-frequency data from high-depth targeted sequencing to infer and iteratively refine tumor purity estimates while accounting for genomic heterogeneity and sub-clonal mutations.