Subjunc

Subjunc maps RNA-seq reads and detects exon-exon junctions using a seed-and-vote alignment strategy for high-precision transcriptome mapping.

Key Features:

• Seed-and-Vote Paradigm: Extracts multiple short subreads from each sequencing read and uses their collective votes to determine optimal genomic locations, enabling flexible mapping and indel detection.
• Efficiency and Speed: Performs an initial genomic position pre-selection based on subread votes before detailed mismatch and indel analysis to reduce search space and accelerate alignment.
• Sensitivity and Accuracy: Requires consensus among multiple subreads rather than exact mapping or proximity of individual subreads, increasing sensitivity and placement accuracy.
• Exon Junction Detection: Identifies reads whose subreads map to different exons within the same gene to detect exon-exon junctions.
• Short-read Support: Handles reads shorter than 160 base pairs with overlapping subreads to support detailed transcriptomic profiling.

Scientific Applications:

• RNA-seq junction detection: Detection of exon-exon junctions in RNA-seq data for gene expression and splicing analyses.
• Transcriptome profiling: Comprehensive transcriptome analysis and detailed transcriptomic profiling, including sequencing projects using short reads (<160 base pairs).

Methodology:

Extract multiple short subreads from each sequencing read; allow subreads to vote on candidate genomic locations; select the location with highest consensus before performing granular alignment to refine mismatches and indels; detect exon junctions by identifying reads with subreads mapping across different exons.