dbAARD & AGP: A computational pipeline for the prediction of genes associated with age related disorders

The atrocious behavioral and physiological shift with aging accelerate occurrence of deleterious disorders. Contemporary research is focused at uncovering the role of genetic associations in age-related disorders (ARDs). While the completion of the Human Genome Project and the HapMap project has generated huge amount of data on genetic variations; Genome-Wide Association Studies (GWAS) have identified genetic variations, essentially SNPs associated with several disorders including ARDs. However, a repository that houses all such ARD associations is lacking. The present work is aimed at filling this void. A database, dbAARD (database of Aging and Age Related Disorders) has been developed which hosts information on more than 3000 genetic variations significantly (p-value <0.05) associated with 51 ARDs. Furthermore, a machine learning based gene prediction tool AGP (Age Related Disorders Gene Prediction) has been constructed by employing rotation forest algorithm, to prioritize genes associated with ARDs. The tool achieved an overall accuracy in terms of precision 75%, recall 76%, F-measure 76% and AUC 0.85. Both the web resources have been made available online at http://genomeinformatics.dce.edu/dbAARD/ and http://genomeinformatics.dce.edu/AGP/ respectively for easy retrieval and usage by the scientific community. We believe that this work may facilitate the analysis of plethora of variants associated with ARDs and provide cues for deciphering the biology of aging.