INFERRING HIV VARIABILITY AND FITNESS FROM HIGH THROUGHPUT SEQUENCING DATA

The human immunodeficiency virus (HIV), similar to many other viruses, is characterized by an exceptionally high mutation rate. Coupled with a huge population size, this mutation rate leads to high levels of genetic heterogeneity in populations of HIV particles replicating inside a patient. This heterogeneity enables the virus to quickly adapt to new challenges, in and in particular to newly administered drugs, potentially leading to drug resistant variants. The probability of emergence and retainment of a drug resistant mutant is dependent, among other factors, on the replicative fitness of the drug resistance mutation. such as different hosts and immune system pressure. Mutations can be broadly classified into deleterious, neutral, or adaptive.

Next Generation Sequencing (NGS) techniques have revolutionized our ability to probe heterogenous populations of HIV replicating inside patients. Here, we use a populations-genetics approach to characterize the amount and nature of the genetic diversity in NGS data from drug-naive patients. In particular, we focused on patients harboring drug resistant mutations that were presumably transmitted from the previous carrier. Our goal was to test whether these mutations impacted the genetic diversity of the viral population, and to further test whether a relationship exists between the fitness of a drug resistant mutation and the probability it is transmitted and retained in the new host. We investigated how errors derived from sequencing and library preparation impact our ability to infer the viral genetic variability, and discuss novel approaches aimed at overcoming these problems. Ultimately, our approach will allow dissecting the relationship between drug resistant mutations, their fitness, and their impact on viral replication and viral evolution.