|Home | About | Journals | Submit | Contact Us | Français|
The HIV Reverse Transcriptase and Protease Sequence Database is an on-line relational database that catalogs evolutionary and drug-related sequence variation in the human immunodeficiency virus (HIV) reverse transcriptase (RT) and protease enzymes, the molecular targets of anti-HIV therapy (http://hivdb.stanford.edu). The database contains a compilation of nearly all published HIV RT and protease sequences, including submissions from International Collaboration databases and sequences published in journal articles. Sequences are linked to data about the source of the sequence sample and the antiretroviral drug treatment history of the individual from whom the isolate was obtained. During the past year 3500 sequences have been added and the data model has been expanded to include drug susceptibility data on sequenced isolates. Database content has also been integrated with didactic text and the output of two sequence analysis programs.
Human immunodeficiency virus (HIV) reverse transcriptase (RT) and protease enzymes are the molecular targets of the 15 currently licensed antiretroviral drugs. Sequence changes in the genes coding for these enzymes are directly responsible for phenotypic resistance to RT and protease inhibitors. Individuals infected with drug-susceptible HIV isolates experience reductions in morbidity and mortality with appropriate antiretroviral drug therapy. In contrast, individuals infected with drug-resistant isolates do not usually respond to drug therapy (1). Assays for sequencing HIV RT and protease are commercially available and are widely used in clinical settings. However, the optimal means of interpreting RT and protease sequences is not known, and the potential utility of such sequence results in clinical settings is an area of intense clinical investigation (2).
HIV RT is a heterodimer composed of a p66 subunit, which contains the DNA-binding groove and the active site, and a p51 subunit, which appears to function as a scaffold for the enzymatically active p66 subunit. HIV RT is responsible for RNA-dependent DNA polymerization, RNase H activity and DNA-dependent DNA polymerization. Although it is 560 amino acids in length, almost all known drug resistance mutations are found in the 5′-polymerase coding region, which is the region sequenced by most clinical laboratories. HIV protease is responsible for the post-translational processing of the viral gag and gag-pol polyproteins to yield the structural proteins and enzymes of the virus. The enzyme is an aspartic protease composed of two non-covalently associated, structurally identical monomers 99 amino acids in length. The protease has a binding cleft that specifically recognizes and cleaves at least 10 different sequences on viral precursor polyproteins.
Genetic analysis of HIV isolates has revealed 10 different group M (main) subtypes, differing from one another by 10–30%, and several highly divergent group O and group N outlier sequences (3). Sequences of HIV isolates collected from around the world provide evidence for naturally occurring genetic polymorphisms; sequences of isolates from persons receiving anti-HIV drug therapy indicate genetic changes selected under antiretroviral drug pressure that may be associated with drug resistance. Sequences within the HIV RT and Protease Sequence Database indicate that ~40% of the 99 amino acids in the protease are polymorphic in untreated individuals, and up to 67% are polymorphic in patients receiving antiretroviral therapy. In the RT, percentages of polymorphisms are ~25 and 40%, respectively.
The HIV RT and Protease Sequence Database is a relational database begun in 1998 as a catalog of sequences linked to data about the source of the sequence sample and the antiretroviral drug treatment history of the individual from whom the isolate was obtained. The database schema has been described in detail in earlier publications (4,5). During the past year, the data model has been expanded to include phenotypic drug susceptibility data and other data pertaining to the interpretation of RT and protease sequences obtained in clinical settings. In addition, two sequence analysis programs and didactic text have been integrated via hyperlinks to specific database queries. In the sections that follow, we will review key database features with an emphasis on those that were introduced within the past year.
Table Table11 contains a summary of the key web pages in the database divided into three sections: documentation, queries and sequence analysis programs. As of September 15, 2000, the database contained 7909 sequences from 2215 individuals, including 4038 RT and 3871 protease sequences. These sequences include 5820 sequences published in GenBank and 2089 sequences from published journal articles. The database also contains >3500 drug susceptibility results.
The ‘Background’ is a brief bulleted description of the database rationale targeted towards a lay audience. The ‘Primer’ is a more detailed description of the database rationale, particularly with respect to the problem of HIV drug resistance; the ‘Primer’ also describes how the HIV RT and Protease Sequence Database differs from other databases containing HIV sequences, including the International Collaboration databases and Los Alamos HIV Sequence Database (6). The ‘Data Model for Understanding HIV Drug Resistance’ describes an ontology of HIV drug resistance based on four types of correlations between genetic sequences and other types of data, including drug treatment histories of patients from whom HIV isolates are sequenced, in vitro drug susceptibility data from laboratory HIV isolates, in vitro drug susceptibility data from clinical HIV isolates and clinical outcome data from patients receiving anti-HIV therapy.
‘Summary Statistics’ is a dynamically generated page based on SQL queries that provides the number of patients, isolates and sequences in the database meeting a variety of criteria. The ‘Drug Resistance Notes’ page contains graphical didactic summaries of the most common drug resistance mutations. The graphical summaries are image maps linked to data within the database relating the mutation to each of the clinically available anti-HIV compounds (Fig. (Fig.11).
Users can retrieve theoretically limitless numbers of different sequence sets matching selection criteria based on specific references, drug treatments, RT and protease mutations, and drug susceptibility patterns (Table (Table1,1, Database query forms). Each query returns a new table and each record in the new table contains 8–12 columns of associated data selected by the user. The data returned may include: (i) hyperlinks to MEDLINE abstracts and GenBank records; (ii) complete nucleotide sequences and translations; (iii) classification of the sequences by individual and time point; (iv) data on the HIV species, group and subtype represented by each sequence; (iv) data on drug treatment including lists of drugs and complete summaries of drug regimens received by the patient from whom the isolates were obtained; and (v) technical data on the methods of virus isolation and sequencing. Following retrieval of a sequence set, users can view or download complete sequence alignments in a variety of formats, including a composite sequence alignment summarizing the frequency and type of mutation at each position in the sequence (Fig. (Fig.22).
HIV-SEQ (HIV RT and Protease Search Engine for Queries, formerly HRP-ASAP) accepts user-submitted RT and protease sequences, compares them to a reference sequence and uses the differences (mutations) as query parameters for interrogating the sequence database (7). The program allows users to discover associations between a submitted sequence and previously published sequences containing the same mutations. The web site also contains a beta test version of an anti-HIV drug resistance interpretation program, which accepts user-submitted RT and protease sequences and returns inferred levels of resistance to 15 clinically available anti-HIV drugs. Drug resistance is inferred using a comprehensive set of rules hyper-linked to the output of specific database queries.
The data model will be broadened to link sequence data with additional types of in vitro and clinical phenotypic data. The model will continue to utilize relational features to represent data such as sequences and in vitro drug susceptibility results. However, object-oriented features will be incorporated to model higher-level data such as disease stage, complex treatment regimens and response to anti-HIV therapy. These changes in the data model will expand the usefulness of the database, increasing the user base from beyond a core group of sequence analysts to include clinical investigators studying other aspects of HIV drug resistance. Preliminary research on a third molecular target of HIV therapy, the gp41 envelope protein (8), has also begun and will be included in the database when appropriate.
Please refer to this article, when citing the HIV RT and Protease Sequence Database.