Pathways and Mechanisms of Drug Resistance Evolution in HIV

About This Grant

PROJECTSUMMARY/ABSTRACT There are ~40 million people worldwide living with the Human Immunodeficiency Virus Type 1 (PLWH), commonly referred to as HIV. In the absence of a functional cure, antiretroviral therapy (ART) represents the primary treatment option. However, ART fails over time due to the development of drug resistance, even against latest generation drugs. Thus, there is an urgent need to understand the pathways and mechanisms through which resistance emerges, which can then be harnessed for developing next-gen therapeutics to specifically target drug-resistant variants. The proposed research seeks to elucidate the mutational pathways and molecular mechanisms through which resistance evolves against the latest ART drugs. Firstly, a Potts sequence- covariation fitness landscape model built on HIV sequences from the Stanford HIV database will be used to determine the epistatic interactions modulating drug resistance and to identify the specific sequential order in which mutations arise forming pathways to complex resistant variants. This aim builds directly from the candidate’s doctoral research developing the Potts fitness landscape models for HIV. Secondly, a combination of structural biology using cryogenic electron microscopy (cryo-EM) and multiscale molecular simulations will be employed to determine the atomistic mechanisms of resistance against latest generation ART drugs. This aim builds on the candidate’s ongoing training in cryo-EM and will serve as a vehicle to obtain complementary research skills in protein biochemistry, and hybrid quantum mechanics molecular mechanics simulations. Finally, in the independent phase, the project aims to combine the Potts fitness landscape models, structural biology and molecular simulations within a Reinforcement Learning (RL) framework, a state-of-the art artificial intelligence (AI) method, to forecast the evolution of drug resistance in HIV. The aim will prognostically predict the dominant emerging pathway trajectories and mechanisms in the PLWH population to prospectively inform the design of optimum therapeutic strategies. The principles learned (by the RL model) and the methodology developed can be applied to forecast emerging resistance in other RNA viruses including those with scarce data. This aim builds on the candidate’s ongoing training in AI methods through the Eric & Wendy Schmidt AI-in- Science Fellowship and combines the multidisciplinary research methods and training obtained through his doctoral and postdoctoral tenure. Overall, the project seeks to produce critical insights into the mechanisms of HIV evolution under drug selection pressure that can help design next generation therapeutic strategies targeting resistant strains and lead to the development of surveillance tools to monitor and forecast emerging drug resistance. The career development plan will take place at The Salk Institute and provide an important opportunity for training and development that will prepare the candidate to start his independent research career.