sequana_coverage

sequana_coverage analyzes genomic coverage from high-throughput sequencing to identify regions of significant over- or underrepresentation and to characterize coverage-based variations such as repetitive regions, deleted genes, and copy number variations (CNVs).

Key Features:

• Detection of Genomic Regions of Interest (ROIs): Identifies ROIs by analyzing base-level coverage deviations from expected values, including both overrepresented and underrepresented regions.
• Statistical Robustness: Uses z-score statistics, normalization of genome coverage, and a Gaussian mixture model to estimate coverage distribution and assess significance.
• Data Detrending and Clustering: Applies an efficient running median algorithm for detrending and a double-threshold mechanism to cluster ROIs.
• Comprehensive Reporting: Generates HTML reports with standard plots and metrics and summarizes coverage alongside genomic variations such as single-nucleotide variants and CNVs.
• Output Formats: Exports CSV files listing regions with low or high coverage relative to the average.

Scientific Applications:

• Genomic variation detection: Aids identification of CNVs, deleted genes, and coverage anomalies associated with single-nucleotide variants.
• Repeat and structural anomaly analysis: Highlights repetitive regions and structural anomalies inferred from coverage deviations.
• Origin-of-replication and bias detection: Reveals trends such as the origin of replication or unknown sequencing biases.
• Genome-wide coverage profiling: Supports genome-wide studies of coverage patterns from high-throughput sequencing data.

Methodology:

Data detrending with a running median; normalization and coverage distribution estimation with a Gaussian mixture model; assignment of z-scores to base positions for significance assessment; and ROI clustering using a double-threshold mechanism.