RDscan

RDscan identifies deletions and putative regions of difference (RDs) in Mycobacterium tuberculosis complex (MTBC) genomes from short-read whole-genome sequencing to support species differentiation and evolutionary analysis.

Key Features:

• Snakemake pipeline: Implements the analysis as a Snakemake workflow for reproducible computational execution.
• Deletion detection from short-read WGS: Detects deletions in short-read whole-genome sequencing data across MTBC genomes.
• Comprehensive dataset analysis: Applied to a reference dataset of 721 samples representing the full diversity of MTBC for extensive variation discovery.
• Correlation with curated RDs: Compares detected deletions against a curated list of high-specificity RDs to assign taxonomic specificity at species, lineage, or sublineage levels.
• Novel RD discovery: Identifies 17 regions not previously described in existing RD catalogs.
• Large-scale validation: Demonstrated high specificity when tested on approximately 7,000 samples.
• Phylogenetic locus identification: Identifies 79 loci associated with phylogenetic units that can serve as lineage or species markers.
• Adaptation and intersection analysis: Analyzes intersections among loci to highlight nonrandom patterns and potential roles in bacterial adaptation.

Scientific Applications:

• Reductive genome evolution studies: Enables investigation of deletion patterns underlying reductive genome evolution in Mycobacterium species.
• Species and lineage differentiation: Supports assignment of species, lineage, or sublineage specificity within the MTBC based on RD presence/absence.
• Phylogenetic marker discovery: Provides loci useful for phylogenetic analyses and for developing diagnostic markers within MTBC.
• Bacterial adaptation research: Facilitates study of genomic loci intersections that may reflect adaptive processes to external conditions.

Methodology:

Implemented as a Snakemake pipeline that detects deletions from short-read whole-genome sequencing and correlates detected deletions with a curated list of high-specificity RDs; analyses were performed across a 721-sample diversity dataset and validated on ~7,000 samples.