Genetic basis of gene regulation: genome-wide models and applications to diagnosis

I will present recent results on the modelling of RNA metabolism rates from cis-regulatory elements. I will show that, for yeast, most of the between-gene variation in mRNA-stability can be predicted from sequence alone. Our model integrates known functional cis-regulatory elements, identify novel ones, and quantify their contribution at single-nucleotide resolution. We show quantitatively that codon usage is the major determinant of mRNA stability, and that this effect depends on canonical mRNA degradation pathways.

In a second part of the talk, I will present a pilot study that demonstrates the utility of RNA-sequencing for the genetic diagnosis of patients affected with rare disorders. Across 105 fibroblast cell lines derived from patients with a mitochondrial disorder, we found a median of 1 aberrantly expressed gene, 5 aberrant splicing events, and 6 mono-allelically expressed rare variants in and established disease-causing roles for each kind. Private exons often arose from sites that are weakly spliced in other individuals, providing an important clue for future prioritisation of deep intronic variants.