Genome-wide DNA methylation profiles of lymphoblastoid cell lines from sex-matched sibling pairs discordant for autism diagnosis

The goals of this study were to: 1) investigate changes in DNA methylation in lymphoblastoid cell lines (LCL) from individuals with a specific subtype of ASD and their respective non-autistic siblings, 2) relate the differentially methylated genes to ASD-relevant pathways and functions, and 3) explore the potential for class determination (case-control) biomarkers based on methylation signatures. DNA from LCL of 21 sibling pairs (42 samples in all) were analyzed for methylated regions using the MeDIP-chip protocol. Each sample was represented by MeDIP-enriched DNA and input DNA which were each fragmented, labeled and separately hybridized to Affymetrix Human Promoter 1.0R GeneChips (84 arrays in all) following the Affymetrix protocols. Data from the scanned genechips were first analyzed for methylation-enriched regions using Partek's workflow for tiling arrays, which resulted in MAT scores representing methylation-enriched (positive MAT scores) and depleted (negative MAT scores) across the genome as well as determination of average probe intensities across the promoter regions. The intensities were then imported into Multi-experiment View (MeV) microarray analysis software for further analyses, These additional data analyses to identify differentially methylated genes for class prediction included Significance Analysis of Microarrays (SAM), Pavlidis Template Matching (PTM), Uncorrelated Shrunken Centroids (USC), and Support Vector Machines (SVM) which were all incorporated into the MeV software package.