Re: [Geuvadis_rna_analysis] analysis group TC Nov 15 at 2pm

Here some update on the splicing companion outline: Tentative title: Splicing Singularities and Population Polymorphisms of Human Transcriptomes (1) transcriptome composition varies with gene expression - although gene discovery does not saturate, most genes of a certain tissue are expressed ubiquitously across individuals and populations - population-determinant genes (specific from 1 population) show proliferant factors age of cell line bias, pairwise distances and tree - differentially expressed genes accumulate for cell surface factors (2) transcriptome configuration changes by differential splicing (DS) - most genes show little transcript variability - most variable genes show enrichment for cell surface terms (?), but no significant overlap with DE genes (!) - alternative psi exons (splicing only) (3) genetic polymorphisms that influence DS - fraction of variants in splice site regions, compare to subgenic localization of sQTLs - classification of splice site variants in 5 distinct groups, characterization: activating variants differ in allele frequencies, deteriorating vs. neutral/enhancing indicative of psi exons (4) discovery of novel elements by RNAseq - intron variants, number of donors/acceptors...weak are predominantly rare in population - how many novel events, broken down by type, characterization Bottom line - the transcriptomes differ not markedly between individuals/populations - but quantitative (gene expression level) and qualitative (splicing level) differences occur not (only) at random On Thu, Nov 15, 2012 at 5:54 AM, Tuuli Lappalainen < Tuuli.Lappalainen(a)unige.ch&gt; wrote: