Challenges and Opportunities in Computational Biology: Interpreting the human genome

With the completion of the human genome, the challenge of interpreting the 3 billion letters that encode a human being has put computational biology at the center stage of biomedical research. Comparative genomics of multiple mammalian species has emerged as one of the most powerful and systematic ways to identify the functional elements encoded in the human genome, by virtue of their conservation across millions of years of evolution. These elements are subtle and hidden in millions of non-functional nucleotides, requiring new techniques to decipher them and to understand their roles across diverse experimental and functional datasets. In this talk, I will cover how comparative genomes of multiple closely related species can be used to systematically interpret the human genome, and in particular for the de-novo discovery of genes, regulatory motifs, microRNAs, gene targets, and enhancer elements. By studying the conservation properties of known functional elements, we define evolutionary signatures, specific to each type of functional element and dictated by the precise selective constraints that it evolves under. We have successfully applied such approaches to study multiple fungal, fly, and mammalian genomes, refining the annotation of existing elements and revealing hundreds of new functional elements. In addition, comparative genomics has enabled us to study the evolutionary dynamics of genomes, gene families, and regulatory motifs, and their role in the emergence of new gene functions.