MutationAligner

MutationAligner identifies somatic mutation hotspots in protein domains by aggregating recurrent mutations at homologous positions across paralogous genes using multiple sequence alignments of human protein domains and cancer sequencing data.

Key Features:

• Dataset composition: Leverages somatic mutation data from over 5,000 cancer patient samples across 22 distinct tumor types (mid-2015).
• Multiple sequence alignments: Uses multiple sequence alignments of human protein domains to map homologous positions across paralogs.
• Domain-centric aggregation: Aggregates mutations occurring at homologous positions within sets of paralogous genes to extend recurrence analysis beyond single genes.
• Recurrence analysis extension: Applies an extended recurrence analysis to detect hotspots by combining evidence from homologous domain positions.
• Statistical power enhancement: Increases power to detect cancer-relevant mutations by pooling signals at conserved domain positions.
• Hotspot identification: Identifies significant mutation hotspots within protein domains across multiple genes and tumor types.
• Functional implication linkage: Connects domain-centric mutation hotspots to putative functional consequences for encoded biological functions.
• Applications to cancer genomics: Provides domain-level mutation maps useful for downstream experimental design and translational interpretation.

Scientific Applications:

• Functional genomics experiment design: Prioritizes domain positions and paralog groups for targeted functional validation studies.
• Precision medicine interpretation: Highlights recurrent domain mutations that may inform clinical decision-making and biomarker discovery.
• Cancer molecular characterization: Aids analysis of the molecular underpinnings of cancer by revealing domain-level mutation patterns across tumor types.

Methodology:

Uses multiple sequence alignments of human protein domains and extends recurrence analysis by aggregating somatic mutations at homologous positions within sets of paralogous genes using somatic mutation data from >5,000 cancer patients across 22 tumor types (mid-2015).