Population genomics of intrapatient HIV evolution

Abstract

The Human Immunodeficiency Virus 1 (HIV-1) is a rapidly evolving human retrovirus. HIV-1 nucleic acid sequences have been sampled from many pa- tients, with mostly one sequence per patient, to characterize HIV-1 genetics and epidemiology. Ultimately, however, HIV-1 replicates and evolves during single infections that last for several years. In my doctorate I performed whole-genome longitudinal deep sequencing on several HIV-1 patients and developed experimental, theoretical, and computational methods to (i) char- acterize HIV-1 evolution within single infections, (ii) organise and share the collected genomic data with the research community, and (iii) simulate evo- lution of rapidly adapting organisms like HIV-1 in silico. First, I quantified a number of central properties of intrapatient HIV-1 evolution such as genetic diversity, evolutionary rate, linkage disequilibrium, mutation rate, strength and prevalence of positive and purifying selection, and influence of RNA sec- ondary structures. Second, exploiting modern web technologies, I realized a web application that gives other researchers the chance to perform specific analyses on the same data set. Third, I coded a computer package, FFPop- Sim, to simulate the evolution of populations under selection; via a novel algorithm and a cross-language design, it has proven an ideal tool to bridge theoretical predictions and experimental results.