katdetectr

katdetectr performs unsupervised changepoint analysis to detect, characterize, and visualize kataegis loci from genomic variant data in cancer samples.

Key Features:

• Data format support: Imports industry-standard formats including MAF, VCF, and VRanges.
• Intermutation distance calculation: Determines the intermutation distance of genomic variants as input for changepoint detection.
• Changepoint detection: Utilizes the Pruned Exact Linear Time search algorithm for unsupervised changepoint analysis.
• Kataegis calling: Identifies putative kataegis foci according to user-defined parameters.
• Characterization and visualization: Provides functionalities to characterize and visualize detected kataegis loci.
• Standardized output: Returns results in a standardized format useful for subsequent analysis.
• Performance: Demonstrated robustness regarding tumor mutational burden with the fastest mean computation time and highest accuracy (0.99, 0.99) and normalized Matthews correlation coefficient (0.98, 0.92) compared to five evaluated kataegis detection packages.

Scientific Applications:

• Cancer genomics: Detects regional genomic hypermutation (kataegis) observed across a wide range of malignancies.
• Pan-cancer analysis: Applicable to whole-genome sequenced malignancies for identifying biologically and potentially clinically relevant kataegis loci.
• Synthetic and labeled dataset validation: Evaluated using synthetic data and an a priori labeled pan-cancer dataset of whole-genome sequenced malignancies.

Methodology:

katdetectr imports genomic variant data, determines intermutation distances, and performs unsupervised changepoint analysis using the Pruned Exact Linear Time search algorithm followed by kataegis calling with user-defined parameters.