SVXplorer

SVXplorer identifies structural variants from short-read sequencing data by integrating discordant paired-end alignments, split-reads, and read-depth signals to detect and characterize complex SVs such as deletions, tandem and non-tandem duplications, inversions, translocations, and novel sequence insertions.

Key Features:

• Three-Tier Approach: Integrates discordant paired-end (PE) alignments, split-reads (SR), and read-depth (RD) information with sequential recombination of discordant cluster signatures to improve sensitivity and precision in SV detection.
• Graph-Based Clustering: Employs a graph-based clustering mechanism to integrate non-trivial discordant PE signatures and resolve complex structural variants involving three or more breakpoints.
• Comprehensive Variant Detection: Identifies deletions (DEL), tandem duplications (TD), inversions (INV), non-tandem duplications, translocations, novel sequence insertions (DN_INS), and undetermined breakend types tagged as BND.
• Input and Output Formats: Accepts BAM files as input and outputs results in BEDPE with variant-specific tags and VCF conforming to VCF 4.3.

Scientific Applications:

• Cancer genomics: Detection and characterization of somatic structural variants relevant to tumor genomes using short-read sequencing.
• Genetic disorders: Identification of germline structural variants underlying Mendelian and other genetic diseases.
• Evolutionary biology: Analysis of structural variation to study genomic architecture and evolutionary changes across populations or species.
• General genomic research: Accurate SV discovery and annotation in studies relying on paired-end short-read data.

Methodology:

Leverages discordant read-pairs, split-reads, and read-depth signals, applies sequential recombination of discordant cluster signatures, and uses graph-based clustering to integrate non-trivial signatures for identification of complex structural variants.