Smoking-associated DNA methylation features link to HIV outcomes [HumanMethylation450 BeadChip]

Smoking is common in people who live with HIV infection and has significant adverse effects on HIV outcomes. The impacts of smoking on methylome has been well established in non-HIV populations. However, the smoking’s effects on host methylome in HIV-positive population has not been investigated and it is unknown if smoking-associated DNA methylation link to HIV outcomes. In this study, we applied machine learning methods selected smoking-associated DNA methylation features to predict HIV related frailty and mortality. We have collected blood samples from the Veteran Aging Cohort Study (VACS). Smoking status was based on self -report. HIV frailty was measured by VACS index. All DNA samples were extracted from whole blood in HIV-positive men. Epigenome-wide DNA methylation of each sample was profiled using the Illumina Infinium HumanMethylation450 BeadChip array. Data were normalized and controlled for quality using standard techniques. Please note that the 'readme.txt' contains a brief description for each sample characteristics. The following data is included in each sample records (as characteristics) and in the 'PCA_analysis_results.txt' file: Cont_Pr_850k_PC1-PC30 PCA analysis results (PC1 to PC30) on positive control intensities resi_850k_PC1-PC5 PCA analysis results (PC1 to PC5) on residual values of the first regression model