KisSplice

KisSplice detects and quantifies polymorphisms and alternative splicing events from RNA-seq data without requiring a reference genome or full transcript assembly.

Key Features:

• De Bruijn Graph-Based Polymorphism Detection: Constructs a De Bruijn graph from RNA-seq reads to identify graph patterns corresponding to polymorphic events without global transcript reconstruction.
• KISSPLICE Algorithm for Alternative Splicing: Implements an exact KISSPLICE algorithm to extract alternative splicing events with high sensitivity, enabling discovery of novel events absent from existing annotations.

Scientific Applications:

• Reference-Free Transcriptome Variation Analysis: Enables detection of alternative splicing and polymorphisms in species lacking reference genomes and in highly polymorphic transcriptomes.

Methodology:

KisSplice builds a De Bruijn graph from RNA-seq reads, models polymorphisms as specific graph patterns, and applies the KISSPLICE algorithm to perform local assembly and precise extraction of alternative splicing events.