Section: New Results

KisSplice: de-novo calling alternative splicing events from RNA-seq data

We addressed the problem of identifying polymorphisms in RNA-seq data when no reference genome is available, and avoiding an assembly. Based on the fundamental idea that each polymorphism will correspond to a recognisable pattern in a De Bruijn graph constructed from the RNA-seq reads, we propose a general model for all polymorphisms in such graphs. We then introduce an exact algorithm to extract alternative splicing events and show that it enables to identify more correct events than current transcriptome assemblers. Additionally, when we applied our method on a 50M reads dataset from human, we were able to identify 3884 events, out of which 57% are not present in the annotations, which confirms recent estimates showing that the complexity of alternative splicing has been largely underestimated so far. This work has been submitted to publication. This work was done in collaboration with P. Peterlongo.