Advances in Statistical Modeling of High Dimensional Data:
Variable selection, and Challenges in Image Analysis

Osvaldo Zagordi, ETH Zürich

Deep sequencing of a mixed sample

Genetic heterogeneity within organisms of the same species plays an essential role in nature. A typical example of this heterogeneity is the HIV population present in a single patient, where the diversity can be so high to represent one of the main problem in the development of an effective vaccine. Sequencing technologies of the new generation, thanks to their higher throughput and lower cost per base, allow us to address this problem in a way that was unimaginable only a few years ago. By sequencing a genetically heterogenuous sample, in fact, we can estimate not only the amount of variation, but also the identity and frequency of the haplotypes responsible for it. This is done by means of a procedure that corrects the erros produced in the sequencing process, thus separating the technical errors from the real biological variation. After giving a brief description of the techniques used, implemented in the freely available software ShoRAH, I will present results on both simulated and real data.

[Back to the Schedule]