Genetic distance, positive selection and predicted T cell epitopes.

Subjects are shown in descending order of those with the highest mean genetic distance, with positive selection occurring at the population level (all sequences), shown in the bottom row of the table.
1 Mean substitutions/site for all time points per subject. Numbers are shown for the 1–2 year time point for each individual, since this is the one time point represented in all individuals, with the exception of NP 4, for whom the 4.9 year time point was the only one available.
2 Sites listed in bold type were picked up by more than one method (FUBAR plus SLAC or FEL); those in standard type were indicated by FUBAR alone.
3 Sites indicated by MEME.
4 The majority species for each time point was entered into an online T cell epitope prediction tool tailored to the HLA type of the individual. Only epitopes predicted by more than one method are shown, and in order of where they occur from N- to C-terminus of Vpu. Numbering indicates the amino acid start and end positions. Where more than one similar epitope was predicted, for example two epitopes overlapping the same region but of 9 and 10 amino acids, the one with the highest predicted binding affinity is shown. Peptides in bold type have a high predicted affinity (0–50 nM); those in regular type have medium predicted affinity (51–500 nM).
NP = none predicted.