home · mobile · calendar · colloquia · 2007-2008 · 

Colloquium - Stamatakis

Models, Algorithms, and High Performance Computing for Evolutionary Biology
Alexandros Stamatakis
Ludwig-Maximilians-University Munich

The continuously accelerating accumulation of molecular data poses new challenges for bioinformatics. Therefore, the application of High-Performance Computing (HPC) techniques to solve biological problems as well as the development of novel, faster algorithms is becoming increasingly important and crucial to the success of bioinformatics.

In this talk I will provide a general introduction to phylogenetic inference under the Maximum likelihood (ML) criterion and outline my contributions to the field as well as the impact of my work so far. Initially, I will address work on search algorithms that has lead to run-time improvements of more than 2 orders of magnitude. I will then outline work on new models and model approximations that further accelerate the computation of the phylogenetic Maximum Likelihood function. Another part of my work focuses on the application of HPC techniques to phylogenetic inference. I will outline how the phylogenetic ML function can be mapped to a broad variety of parallel hardware architectures such as, e.g., Graphics Processing Units, the IBM Cell, shared-memory as well as multi-core systems, and the IBM BlueGene/L supercomputer.

Finally, I will list several current collaborative projects with biologists to solve specific evolutionary problems.

Hosted by Rob Knight.
EKLC 338 is the CIRES Auditorium and is on the third floor of the Ekeley Sciences Building.

Department of Computer Science
University of Colorado Boulder
Boulder, CO 80309-0430 USA
May 5, 2012 (14:13)