Splicing is a crucial aspect of gene regulation and this step in the processing of pre-mRNA to mature mRNA must occur with extreme precision to ensure that the correct message is translated by the ribosome. The coding potential and functional versatility of genes is dramatically increased by alternative splicing pathways that generate different protein isoforms, very often in a developmental stage- or tissue-specific manner. About 95% of multiexon human genes undergo alternative splicing but the functional significance of the vast majority of splice variants is entirely unknown. Moreover, little is known about the evolution of the mechanisms of alternative splicing in vertebrate species.The main goals of this proposal are therefore to: (a) generate the first large-scale RNA-Seq analysis of splice variants across major vertebrate species; (b) develop a “splicing code” for vertebrates that predicts tissue-specific splicing patterns; (b) identify and characterize the regulation and function of both conserved and tissue/species-specific splice variants; (d) determine the minimal number of code elements sufficient to promote constitutive and different types of regulated alternative splicing.We will develop an RNA-Seq computational pipeline for the large-scale profiling of alternative splicing in a set of physiologically-equivalent tissues across representative species of vertebrates. From the resulting data we will derive probabilistic models that will allow us, not only to identify biologically informative conserved and condition-specific splicing patterns, but also to make predictions on how individual transcripts are spliced in specific tissues. Such predictions will then be tested in the lab and the experimental results will be used to refine and improve the models. These models will be used to de novo design synthetic splicing reporters with minimal sets of code elements predicted to confer different classes of tissue-dependent alternative splicing.