• This query is designed to show the prevalence of protease and RT mutations according to subtype and PI, NRTI, and NNRTI exposure.
• Data are available for subtypes A, B, C, D, F, G, and the two most common circulating recombinant forms: CRF01 (AE) and CRF02 (AG)
• Protease sequences are classified as PI-naive or PI-treated based upon whether or not the person from whom the sequence was obtained had ever received a PI (regardless of RTI history).
• RT sequences are classified as RTI-naive, NRTI-treated, and NNRTI-treated (regardless of PI history). Sequences are classified as NRTI-treated only if they were obtained from persons who had received one or more NRTIs but no NNRTIs. Sequences are classified as NNRTI-treated regardless of the NRTI treatment history.
• The total number of sequences for a subtype and treatment class is shown in the column header beneath the subtype letter. By default, persons having more than one isolate with the same mutation are counted only once ("Position Mutation Count" = "one mutation per person").
• The output tables contain separate rows for each protease or RT position.
• All mutations that occur above the user-selected threshold appear in the table followed by a superscript indicating the prevalence (percent) of that mutation. For consistency, mutations are defined as differences from the consensus B reference sequence (amino acids shown in the second column)
• Each mutation is hyperlinked to a web page with information about the isolates containing the mutation and meeting the column's subtype and treatment criteria.

Users select the following parameters:

1. ARV Drug class
2. Position Mutation Count: The default option ("One mutation per person") allows different mutations at a position from the same person to count toward the mutation total. However, persons having more than one isolates with the same mutation are counted once. An even stricter option ("One isolate per person") counts only the last 'on-therapy' isolate.
3. Mutation threshold: Selecting ">= 1% AND >= 2 persons" will exclude mutations occurring in <1% of sequences of a subtype and treatment category or in <2 persons; Selecting >0.5% will exclude fewer sequences; Selecting "All" will exclude no sequences.
4. Selecting Include for Mixture means that mutations present as part of an electrophoretic mixture will be counted as mutant.

Based on a review of the published literature from the Stanford HIV Drug Resistance Database, and collaborative efforts of the Non-B Workgroup.