RDXplorer

RDXplorer detects copy number variants (CNVs) in whole human genome sequence data using read depth (RD) coverage and the Event-Wise Testing (EWT) algorithm to identify deletions, duplications, and polymorphisms across multiple genomes.

Key Features:

• Read Depth Coverage Analysis: Analyzes read depth (RD) from whole-genome sequencing to identify CNVs, including large insertions and events in complex genomic regions.
• Event-Wise Testing (EWT): Applies the EWT algorithm on intervals of data points to test the significance of CNV events rather than evaluating likelihoods at every genome position.
• Multiple-Comparison Adjustment: Adjusts significance tests for multiple comparisons to control the overall false-positive rate.
• Alternative to Paired-End Mapping: Operates independently of paired-end read mapping methods, providing detection for CNV classes that paired-end approaches may miss.
• Polymorphism Identification: Compares CNV calls across multiple genomes to identify polymorphisms and recurrent variants.
• Performance Evaluation: Demonstrated high sensitivity in comparisons with ten CNV detection methods across sequencing depths from 5X to 50X using DGV gold standard variants and real sequencing data.

Scientific Applications:

• Genetic association studies: Detects CNVs for studies aimed at uncovering genetic risk factors for diseases.
• Structural variant characterization: Identifies deletions and duplications that contribute to complex phenotypes and disease mechanisms.
• High-sensitivity sequencing studies: Suited for applications requiring maximized true-positive detection across sequencing depths (5X–50X).

Methodology:

Uses read depth (RD) from whole-genome sequencing and applies the Event-Wise Testing (EWT) algorithm on intervals to test CNV significance with multiple-comparison adjustment; compares RD across multiple genomes for polymorphism identification and was evaluated against DGV gold standard variants and real sequencing data across sequencing depths of 5X–50X in comparisons with ten CNV detection methods.