Schedule:
Wednesday April 11
10 a.m., CSE 4217
Pevzner/Bafna journal club

4 p.m., Stein Clinical Research building 344
Doug Richman's lab "Work in Progress" meeting

(Same title/abstract for both talks, but different emphasis.)

Speaker:
Nicholas Eriksson
Dept. of Statistics,
Stanford University

Title:
Viral population estimation using pyrosequencing

Abstract:
Recently developed pyrophosphate based sequencing technologies
(pyrosequencing) can be used for ultra-deep sequencing of entire viral
populations.  Although pyrosequencing is faster and cheaper than Sanger
sequencing, it produces short (~100 bp), error prone reads that are
challenging to deal with.

I will present an answer to the following problem:  given ultra-deep
pyrosequencing reads from a HIV population, reconstruct the population
structure (i.e., the haplotypes occurring in the population and their
frequencies).  This inference is based on a statistical approach to error
correction, followed by a combinatorial algorithm for constructing a set
of haplotypes which explain the data.  Using this set of explaining
haplotypes, we apply a statistical model to infer the frequencies of the
haplotypes in the population via an EM algorithm.

I'll present the results of simulations and perhaps some results on the
structure of HIV populations within patients harboring drug resistant
virus strains.

This is joint work with Lior Pachter, Soo-Yon Rhee, Yumi Mitsuya, Robert
W. Shafer, and Niko Beerenwinkel.


Speaker biography:
Nicholas Eriksson is an NSF postoctoral fellow in statistics at Stanford
University.  He has degrees in pure mathematics from MIT (SB, 2001) and UC
Berkeley (PhD, 2006) with a thesis titled "Algebraic combinatorics for
computational biology".  He is interested in problems in evolutionary and
computational biology that can be solved using tools from discrete math,
statistics, and computer science.