DNA Methylation of gDNA from blood and bronchoalveolar lavage of tuburbulosis patients with and without type 2 diabetes via Illumina Infinium MethylationEPIC array

Tuberculosis (TB) remains a leading cause of illness and death globally. Type 2 diabetes (T2D) patients are more likely to become infected with Mycobacterium tuberculosis (M.tb). When they are infected, they are more likely to develop active TB disease and to present with more severe TB and those who receive TB treatment are more likely to experience adverse treatment outcomes. Genomic DNA (gDNA) was isolated from unstimulated whole blood and bronchoalveolar lavage (BAL) cells. Analysis of methylation arrays was carried out following the Bioconductor ‘methylationArrayAnalysis’ workflow, using the IlluminaHumanMethylationEPICanno.ilm10b4.hg19 annotation set for Illumina MethylationEPIC arrays. Probes were excluded where detection p-value was <0.01, where they were located on the Y chromosome, or were annotated as containing SNPs or known to be cross-reacting. Probes on the X chromosome were retained as only females were included in this analysis. Background correction was carried out via the normal-exponential out-of-band (Noob) method and probe intensities were quantile normalized. To identify differentially methylated positions (DMPs), the data was filtered using the criteria shown in extended data figure 8. To permit comparison between DMPs a secondary cutoff of uncorrected p<0.01 and beta values >0.8 and <0.2 was retained. Gene set enrichment testing was carried out via the “gsameth” function in the missMethyl package using hallmark and GO gene sets from the MSigDB Molecular Signatures Database. The Hallmark (H) gene sets represent well-defined biological states or processes, whereas the GO (C5) set consists of genes annotated by the same ontology term.