Bioinformatický seminár

Tue 26 Feb. 2013, 17:20
I-9

Title: Topfer et al. (2013) Probabilistic inference of viral quasispecies subject to recombination
Speaker: Mic Nánási

Abstract RNA viruses exist in their hosts as populations of different but related
strains. The virus population, often called quasispecies, is shaped by a
combination of genetic change and natural selection. Genetic change is due to
both point mutations and recombination events. We present a jumping hidden Markov
model that describes the generation of viral quasispecies and a method to infer
its parameters from next-generation sequencing data. The model introduces
position-specific probability tables over the sequence alphabet to explain the
diversity that can be found in the population at each site. Recombination events 
are indicated by a change of state, allowing a single observed read to originate 
from multiple sequences. We present a specific implementation of the expectation 
maximization (EM) algorithm to find maximum a posteriori estimates of the model
parameters and a method to estimate the distribution of viral strains in the
quasispecies. The model is validated on simulated data, showing the advantage of 
explicitly taking the recombination process into account, and applied to reads
obtained from a clinical HIV sample.