Section: New Results

Modeling of alternative transcripts with long reads

In the context of transcriptomic analyses based on third generation sequencing data (ONT), we started to explore the following problem : given a transcriptomic experiment, a gene of interest, select reads related to the given gene and find exon junctions. As we have done in the CG-Alcode project, we aim to model the gene as an alphabet of exonic blocks, each transcript being a word over this alphabet. This work takes place in the context of ANR ASTER for which we deal with mouse transcriptomic data in brain and liver. Built models will allow to query human genes to discover putative transcripts.