GASV

GASV identifies and compares structural variants in genomic sequences by representing measurement uncertainty as geometric polygons and computing their intersections to localize variant boundaries.

Key Features:

• Geometric uncertainty representation: Represents structural-variant measurement uncertainty as polygons in a plane to capture imprecise breakpoints.
• Intersection-based variant matching: Computes intersections between polygons to identify measurements that support the same structural variant.
• Computational geometry algorithms: Uses computational geometry algorithms to efficiently determine all polygon intersections.
• Integration of aCGH and paired-end sequencing: Integrates array comparative genomic hybridization (aCGH) and paired-end sequencing/mapping measurements for unified analysis.
• Boundary localization improvement: Improves precision of structural-variant boundary localization compared with individual measurement methods.
• Classification and comparison of SVs: Enables identification, classification, and comparison of duplications, insertions, deletions, and inversions across measurements.
• Cancer versus germline distinction: Facilitates distinguishing genetic structural variants from putative somatic rearrangements in cancer genomes.
• Multi-sample and cross-technique comparison: Provides a general framework for comparing structural variants across multiple samples and measurement techniques.

Scientific Applications:

• Structural variant discovery and classification: Identification and classification of duplications, insertions, deletions, and inversions in genomic data.
• Integration of heterogeneous measurements: Combining aCGH and paired-end sequencing/mapping data to reconcile SV calls from multiple technologies.
• Boundary refinement in sequencing studies: Refining breakpoint localization in paired-end sequencing and aCGH analyses.
• Cancer genomics and somatic rearrangements: Distinguishing putative somatic structural variants from germline variants in cancer genome analyses.
• Cross-study SV comparison and meta-analysis: Comparing and consolidating SV measurements across studies, samples, and techniques, including analyses of human and cancer genomes.

Methodology:

Represents SV measurement uncertainty as polygons and computes polygon intersections using computational geometry algorithms to identify measurements supporting the same variant.