DISSECT

DISSECT identifies and characterizes structural variants in transcriptomes by aligning transcripts to reference genomes to detect duplications, inversions, rearrangements, and fusions.

Key Features:

• Algorithmic formulations: DISSECT implements two transcriptome-to-genome alignment formulations: a nucleotide-level alignment model that provides detailed mapping and a chaining-based formulation that links shared fragments between transcripts and the reference to improve computational efficiency.
• Structural variation discovery: Detects duplications, inversions, rearrangements, and gene fusions across whole transcriptomes, including analyses of long reads and accurately assembled contigs.
• Sensitivity and specificity: Demonstrated high sensitivity and specificity on simulated transcripts with known structural changes and on RNA-Seq data from the human prostate cancer cell line C4-2.
• Sequencing data support: Operates on transcripts derived from high-throughput sequencing, including RNA-Seq datasets.

Scientific Applications:

• Cancer research: Uncover novel structural alterations in cancer transcriptomes to inform studies of oncogenic processes.
• Transcriptome analysis: Identify structural variants that contribute to transcriptomic diversity and may impact gene function.

Methodology:

Align transcripts to a reference genome using nucleotide-level mapping and a fragment-chaining formulation that links shared sequence fragments to detect and characterize structural variants.