岩崎研公開ラボセミナー

Making the most of noisy, low-quality genomes

Dr. Christophe Dessimoz(University of Lausanne, University College London, and Swiss Institute of Bioinformatics)

2017年04月19日(水)    15:00-16:00  理学部3号館 310号室   

Despite technological progress, most eukaryotic genomes remain stuck in
draft quality. In theory, the more we know about the sequence universe,
the easier characterising new genomes should be. Frustratingly however,
integrating information across genomes of varying quality levels remains
conceptually and practically challenging. In the talk, I will first present an
evolutionary framework to address this problem, then I will show how this
can be used to extract good sequences from different annotation sets, or
to identify genes that might be associated with functional innovation or
adaptation in the evolutionary history of these genomes. To address the
issues of contamination, lateral gene transfer, or hidden paralogy, I will
also present a process-agnostic method to cluster subsets of genes that
have a common evolutionary history.