ABSTRACT:
HIV-1 genotyping assays are faster and less complex than phenotyping but interpreting complex mutational patterns is often difficult. In order to improve this interpretation, we have developed a system that couples genotypic data in real time to a phenotype-genotype database. This database currently comprises over 7000 samples which all have drug susceptibility phenotypic data, determined by ABI DNA sequencing. To relate a 'test' genotype to phenotypic resistance information in the database, a series of genotypic patterns were related to specific drugs.
These patterns were attached to all genotypic samples in the database thus enabling rapid searches to be performed. The phenotypes of samples in the database that match a particular genotype can then be retrieved and displayed as the proportion resistant or sensitive to each drug. The following are examples of searches with specific genotypic patterns in RT: 184V alone gave 1281 phenotype-genotype matches, 90% of which were >10-fold lamivudine resistant; 41L, 210W and 215Y, 435 matches, 88% >10-fold zidovudine resistant; 41L, 184V, 210W and 215Y, 938 matches, 71% >10-fold zidovudine resistant. Samples with specific protease inhibitor mutational patterns gave the following search results: 10V, 46I, 54V, 71T, 77I, 82A and 90M, 73% nelfinavir resistant, 53% indinavir resistant, 77% ritonavir resistant, 45% saquinavir resistant. This system makes it possible to obtain a rapid indication of the likely phenotype of a genotyped sample by matching substantial archived phenotypic data to a mutation pattern. The use of phenotype-genotype database searching should enhance the value of genotyping information for routine resistance testing.
Furthermore, complex genotypes with uncertain interpretations will be highlighted, where phenotyping would then be beneficial to clarify the actual resistance patterns.
