To develop new therapies aimed at HIV cure or reducing the sequelae of ART-treated infection, there is a critical need for more research on the persistence of HIV-infected cells that can contribute to immune activation on ART and allow virologic rebound after ART interruption. While the rebound competent reservoir is often assumed to be the same as the latent and/or the intact reservoir, sequences from the rebound virus usually do not match those from the latent reservoir but sometimes match those from cell-associated HIV RNA prior to ART interruption. These findings suggest the need to better understand the reservoirs that express HIV in vivo, which are poised to initiate rebound on interruption of ART and likely contribute to the immune activation, organ damage, and reduced life expectancy on ART. However, prior studies of the “transcriptionally active reservoir” have not been able to fully characterize the heterogeneity of these cells, which vary in terms of the types of HIV RNA transcribed (processive or complete, from defective or intact proviruses) and whether it is translated into HIV protein. We hypothesize that subsets of cells expressing different types of HIV RNA and/or protein will differ in terms of their frequency, survival/clearance rate, contribution to immune activation, cellular gene expression, and tissue distribution. To investigate these hypotheses, this project will apply a series of new and cutting-edge assays to longitudinal samples from different clinical phenotypes in order to: 1) measure how different proviruses (intact/defective), blocks to HIV transcription, defective or intact HIV transcripts, and HIV Gag protein change over time pre- and post-ART in the blood and how they differ between elite controllers and individuals who initiate ART during acute or chronic infection; 2) determine how levels of each type of HIV RNA and protein correlate with HIV-specific T and other immune responses as well as markers of immune activation/inflammati