The NCI Laboratory of Genomic Diversity has identified 15 AIDS restriction genes (ARGs). Such genes affect the probability that HIV-exposed individuals become infected or the rate that HIV-infected individuals progress to AIDS or death. In order to facilitate the discovery of new ARGs, a program named ARG Analysis was developed for the analysis of associations of genotypes from diallelic markers with HIV/AIDS-related outcomes. The statistical tests performed include categorical analyses of infection status and disease progression as well as survival analyses using the Cox proportional hazards model. The user may either specify the major and minor alleles for a marker or have the program do this automatically using its autoassign feature. An exhaustive analysis of a single marker takes about 10 minutes with a 2.8 GHz Microsoft Windows-based computer. In order to allow analyses to be performed faster, the ARG Analysis program was adapted to run on the Obelix Linux-based cluster at the NCI-Frederick Advanced Biomedical Computing Center (ABCC). When run on a single node, a complete analysis for a single marker takes about 25 seconds. The program may be run either by submitting a job to the Portable Batch System (PBS) from the Linux command line or through a web interface. The output consists of a set of tables containing detailed results as well as a digest table containing P-values and odds or hazard ratios. These tables may be examined by a table viewer or processed by other programs that produce human friendly reports. One such program, ARG Tracks, produces a six-page report with tables and graphs for each marker. Another program, ARG Ranks, uses several ranking methods to identify markers that are most likely to be true ARGs.

Agency
National Institute of Health (NIH)
Institute
Division of Basic Sciences - NCI (NCI)
Type
Intramural Research (Z01)
Project #
1Z01BC010317-06
Application #
7049876
Study Section
(LGD)
Project Start
Project End
Budget Start
Budget End
Support Year
6
Fiscal Year
2004
Total Cost
Indirect Cost
Name
Basic Sciences
Department
Type
DUNS #
City
State
Country
United States
Zip Code