HIV-1 is an important human pathogen showing exceptional variability within and between infected patients. Since HIV-1 packages two copies of its RNA genome, genetic recombination may result when the viral polymerase and nascent DNA hop between template genomes. Recombination impacts anti-HIV vaccine design, the evolution of drug resistance, and the increasing variability of circulating virus in the world today. Because recombination is the consequence of a nucleotide strand transfer, primary and secondary sequence features can enhance recombination. Biological studies confirm that local nucleotide content and stem loop structures indeed increase the frequency of recombination. We propose to analyze the strand transfer points of known HIV recombinants for evidence of ubiquitous sequence features. Identified features will be assayed for their recombination-promoting ability. The information gleaned from these joint efforts will be used to enhance recombinant detection techniques. While numerous in vitro studies have found local features that enhance viral recombination, this study would be the first comprehensive analysis of in vivo recombination. This informatics approach will identify recombination promoting sequence features by simultaneously examining many varied recombinants in many varied contexts. This approach will identify recombination-enhancing sequence features that are truly ubiquitous. Additionally, the proposal will provide the research community with a clean database of statistically confirmed HIV-1 recombinants and their crossover points. The sequence features near strand transfer points will be identified with motif search tools. Because the transfer points are not known with certainty and may vary naturally, the methods must allow for ambiguity in the signal. This fuzzy search, for secondary structure in particular, will extend existing search algorithms. This proposal takes an integrated in vitro and in silico approach. The biological and computational results will inform and confirm each other. Further, they will both inform better recombination detection methods. By incorporating knowledge about the recombination mechanism, the power of recombinant detection techniques can be improved. The Bayesian framework is particularly well-suited, and we propose to improve fan existing Bayesian-based method that simultaneously infers recombination and recombinant structure.
Showing the most recent 10 out of 16 publications
