RNA-seq transcriptome profiling uncovers how structural variants influence alternative splicing. E. Ait Yahya Graison1, A. Necsulea2, A. Reymond1 1) CIG, University of Lausanne, Lausanne, Switzerland; 2) Swiss Institute for Experimental Cancer Research (ISREC) Ecole Polytechnique Fédérale de Lausanne (EPFL) CH-1015 Lausanne Switzerland.

   Structural variants (SVs) are a major contributor to genetic diversity but a comprehensive understanding of their functional impact remains to be established. We previously showed that they impact tissue transcriptome by modifying the level and timing of expression of genes that localize within and on their flanks. Here we used mouse inbred strains as a model to extensively gauge by RNA-seq the influence of structural variants on the transcriptome complexity and regulation. Towards this aim, we generated extensive RNA-seq data from liver and brain of C3H/HeJ, 129S2, DBA/2J and C57BL/6J strains using the Illumina HiSeq2000 and Genome Analyzer IIx. These high-throughput sequencing expression data were intersected with the catalog of structural variants encompassing insertions, inversions, deletions, copy number gains and complex SVs produced within the frame of the Mouse Genomes project to assess simultaneously the impact of genome structural changes on both gene expression and alternative splicing. While large SVs (mostly copy number gains or deletions) have a direct impact on transcript expression levels as embedded gene generally show a direct correlation between number of copies and expression levels, smaller SVs significantly influence splicing diversity in several manners. First, the presence of either a deletion or an insertion in an exon significantly favors the emergence of alternative splice donor sites. Second and conversely, a deletion or insertion within an intron significantly increases the number of alternative acceptor sites in downstream exon. This effect is independent of maintenance or disruption of the open reading-frame. Third, the splicing machinery more rarely skips out the SV-containing exons. Fourth, on the contrary, exons that lie just upstream or downstream of an SV-containing intron are more likely to be skipped out by splicing. This latter effect is not dependent on the size of the SV suggesting that it is a true property of the rearrangement rather than the consequence of an increased or decreased size of the flanking intron. To conclude , we show that SVs do impact tissue transcriptome on a global scale also by altering its complexity and diversity through alternative splicing.

You may contact the first author (during and after the meeting) at