Gene Network Identification and Integration (GNetii) Approach to Understanding the Biology Underlying HIV and Drug Abuse.

PROJECT SUMMARY/ABSTRACT The goal of the proposed study is to advance our understanding of the complex networks of biology underlying variation in HIV viral load (VL) and latent reservoir (LR) among HIV+ individuals, and how cocaine abuse (CA) affects identified biological networks. With the success of combination antiretroviral therapy (cART) and public health strategies to reduce HIV incidence, much of the HIV burden in developed countries is now as a chronic disease, including among drug users. Managing HIV progression (HP) and searching for an HIV cure are of paramount importance. Higher pretreatment VL is associated with HP and is associated with a larger LR. An HIV cure is dependent on eliminating the LR. Cocaine is one of the most frequently abused illicit drugs among HIV+ individuals and is known to increase VL, worsen HP, slow decline of viral production after cART, and, we hypothesize, affect the quantity of LR. Thus, there is a complex web of relationships among VL, LR, and CA, which are partially driven by and mediated through genetic susceptibility and gene regulation. As concluded by Le Cleric et al. (2019) in their recent review: “Only integrative approaches that combine all big data results and consider their complex interactions will allow us to capture the global picture of HIV molecular pathogenesis. This novel challenge will require large collaborative efforts and represents a huge open field for innovative bioinformatics approaches.” We propose Gene Network Identification and Integration (GNetii) as a multi-method, multi-omic framework for discovering and understanding the biology underlying HIV outcomes and the effect of CA. We will apply Explainable Artificial Intelligence, network mapping, and Lines-of-Evidence integration to existing genome-, methylome-, and transcriptome-wide data across a number of cohorts in the following aims:  Aim 1: Identify gene networks underlying variation in HIV VL and LR applying GNetii.  Aim 2: Identify differences in HIV associated gene networks by CA. This robustly designed study is significant and innovative: targeting key HIV outcomes affected by CA, applying big data techniques to identify gene networks across multiple omics (enhancing discovery and biological interpretation), and leveraging unique LR data. Our multiple Principal Investigator team includes expertise in HIV, drug abuse, and computational biology. Thus, we are likely to produce important new insights into key elements of HIV as a chronic disease: providing a basis for targeting unique features of CA that impact VL and LR, which make HIV management and a cure more challenging in this population.