We are using the single genome sequencing (SGS) technology that we developed in the previous period to analyze and understand the accumulation of genetic variation in gag/pol and env in a number of different patient groups, including chronically infected patients, both nave and on therapy, as well as in primary and early HIV infection (in collaboration with Drs. Joseph Margolick, Johns Hopkins University; Eric Daar, Harbor-UCLA Medical Center; and Shyam Kottilil, NIAID). The goals of this study are to understand the nature of the forces (mutation, selection, drift, recombination) that mold the genetic diversity of virus populations in vivo. This analysis is being further extended by developing two new sequencing applications: 1) We are applying the SGS technique to analyze the proviral DNA content of cells from patients in parallel with analysis of plasma-derived HIV, as a means of understanding the distribution of the virus population in the body, as well as the potential for recombination. 2) In collaboration with Dr. John Mellors (University of Pittsburgh), we are developing a new assay (multiple genome sequencing, or MGS) as a potentially efficient way to detect resistance mutations in virus populations at low frequency. In addition, we are developing new tools to analyze sequence information, including a novel method to classify drug resistance mutations in the phylogenetic context, which is currently in development and has been applied to understanding the spread of drug resistance in experimental SHIV infections of macaques (in collaboration with Dr. Zandrea Ambrose, University of Pittsburgh). Bioinformatic approaches are also being developed to investigate correlations between the mutations identified by SGS and standard commercially available gentotypes and phenotypes (in collaboration with Dr. Margriet VanHoutte, Virco-Tibotec). In addition, we have applied phylogenetic approaches in collaboration with Joann Mican and H. Clifford Lane (NIAID) and Henry Masur (CCMD) to investigate the presence of non-B subtypes in HIV-1 infected persons in Washington, D.C. We are obtaining a more comprehensive picture of HIV genetic variation on replication in vivo in the presence or absence of drug resistance. We are using the SGS approach to look at resistance more directly, asking, for example, (in collaboration with Dr. Mellors) whether the presence of low-frequency resistance mutations in patients switching from a failed therapy is predictive of subsequent failure. We are extending the portion of the HIV genome we are sequencing to detect novel mutations associated with RT inhibitor resistance in the C-terminal region of RT (in collaboration with Dr. Vinay Pathak, National Cancer Institute). In order to investigate the presence of mutations in non-subtype B viruses, we are collaborating with several groups around the world to obtain useful samples from patients prior to and following drug therapy (Dr. Marcelo Soares, Brazil; Dr. Sunil Arora, India). We are also investigating genetic diversity of the gag p6 gene, a viral gene product that binds to host protein TSG101 in cholesterol-laden rafts and is essential for proper virus budding. Following initial observations we made on the inconsistent effects of pravastatin on HIV viral RNA levels in vivo (cross-sectional study in collaboration with Dr. Peter Sklar, Drexel University), we are studying whether gag p6 genetic diversity and specific TSG101 genotypes are associated with viral RNA levels or response to statin therapy in a new randomized multi-center clinical protocol (NIH Protocol 06-I-0197, in collaboration with Drs. Eric Freed, Mary Carrington, Anuradha Ganesan, Nancy Crum-Clianfone, Henry Masur, and Peter Sklar). The nature of HIV-1 populations in patients undergoing antiretroviral therapy remains uncertain, and we are conducting an extensive genetic analysis of HIV-1 before and after initiation of antiretroviral therapy (NIH Protocol 97-I-0082). These results will yield new information regarding the nature and timing of genetic bottlenecks occurring during antiretroviral therapy. Analysis of HIV-1 sequences at relatively low viremia has been limited by technical issues in amplifying the relatively few HIV-1 sequences present in plasma during therapy. In the last year, the HIV Drug Resistance Program (DRP) Virology Core has successfully adapted the SGS procedure to obtain acceptable numbers of sequences from patients suppressed on antiretroviral therapy. In collaboration with Drs. Michael Polis (NIAID) and Deborah Persaud (Johns Hopkins University, NIH Bench to Bedside Award, 2006), we are analyzing genetic variation in patients enolled in NIH Protocol 97-I-0082 who have been suppressed on antiretroviral therapy for a prolonged time ( more than 7 years). The SGS approach, as developed in the DRP, is rapidly becoming the standard approach to investigate HIV populations, with a number of groups and large networks employing the technique, notably the Center for HIV/AIDS Vaccine Immunology (CHAVI). In addition, the SGS technique has been utilized to investigate population genetics of other pathogens, and NIH investigators (Joseph Kovacs, NIH Clinical Center) have collaborated with our Host-Virus Interaction Branch in the first demonstration of genetic evidence for recombination in Pneumocystis jeroveci, an opportunistic fungus that is responsible for substantial morbidity and mortality in HIV-infected individuals. The DRP is also developing new technologies to investigate HIV-1 genetic variation. We are investigating massively parallel pyrosequencing techniques to study HIV populations genetics. Although such ultra-deep technology has been used to study HIV-1, the utility of the approach remains uncertain, because it is not clear whether the approach can accommodate a highly genetically diverse virus population and yield accurate phylogenetic data and allele frequencies. The DRP has an extensive database of single genome sequences from a large cohort of well-characterized patients. These single genome sequences will provide the gold standard to compare results of pyrosequencing and determine the potential utility of massively parallel sequencing in genetic analysis of HIV-1 populations. [Corresponds to Project 2 in the April 2007 site visit report of the Host-Virus Interaction Branch, HIV Drug Resistance Program]