Our efforts resulted in a publication on a meta-analysis of studies satisfying the following criteria: (1) known donors infected recipients (e.g., mothers infected breast-feeding babies or viral stock infected macaques); and (2) the recipient sequences are from acute phase of infection, as defined from Fiebig staging. Such studies provide data relevant to a question posed by Drs. Lewis and Devico: are the sequences of the infecting and non-infecting immunodeficiency virus particles in a inoculum statistically distinguishable? Our meta-analysis indicated that the founder viruses of the SIV infection differ significantly from the typical virions in the stock inoculum. Moreover, the statistical differences can be assigned to specific signature residues in the HIV attachment protein (Env). Our meta-analysis showed that some Env residues were not signatures of vaccine breakthrough, as was believed, but signatures normally required for naive infection. An independent confirmed our results. Currently, we are exploring the signature residues in the broader evolutionary setting of lentiviruses, to help discover the residues' function. Using clinical HIV sequence data, Dr Manzourolajdad is also exploring the evolution of the Nef and Gag genes, and in particular, the selection pressures on their RNA structures. Most studies examine protein selection, but our results indicate that at least one Nef region is under strong selection pressure for alternative RNA folds. Nef may perform some virological functions through an RNA switching mechanism. Accordingly, we are developing general computational tools for identifying alternative folds in riboswitches.
