HLA profiles predict known and novel HIV-1 escape mutations at a population level

Poster number: 21

Roomp K (1), Ahlenstiel G (2), Beerenwinkel N (1), Rockstroh J (2), Däumer M (3), Spengler U (2), Lengauer T (1)

  1. MPII, Saarbruecken, Germany
  2. Department of Internal Medicine, University of Bonn, Germany, 3 Institute of Virology, University of Cologne, Germany

We have examined HIV-1 sequences in a HLA-diverse patient population to identify virus adaptation to host immune responses (escape mutations). HIV-1 sequence data was collected from a population of 230 HIV-positive individuals. HLA-A, HLA-B and HLA-DRB1 typing was performed in all cases. A relational database was designed specifically for the analysis of host immune response effects in HIV-infected patients. All amino acids between positions 94 and 214 in the reverse transcriptase were examined in the most recent sequences from all patients. We analyzed each amino acid position using Fisher’s exact test to identify all HLA allele covariates with P-values less than or equal to 0.05. Subsequently, logistic regression models were fitted for all remaining covariates. Only those covariates with P-values less than or equal to 0.01 were retained. We are able to confirm the statistical significance of the majority of associations, which were reported as significant after correction for the total number of residues examined across the entire region, by Moore et al., Science, 2002. Additionally, we have found a number of novel escape mutations.