Statistical analysis of methylation reveals epigenetic biomarkers for high- functioning autism spectrum disorder

Abstract

Autism spectrum disorder (ASD) is a neurological disorder that affects social behavior and learning ability. Although studies have identified certain regions of the genome (22q13.33) linked to the increased risk of ASD, there is no clear etiology, which is why psychologists theorize that ASD is a complex interaction between genetic and environmental factors. The current study was designed to identify genes with a positive correlation toward the diagnosis of high-function autism spectrum disorder (HF-ASD). Samples were obtained from National Center for Biotechnology Information (NCBI) containing 450K CpG methylation from 69 subjects diagnosed as HF-ASD or typically developing (TD). We applied Principal Component Analysis (PCA) and further utilized two sample Kolmogorov-Smirnov test (KS-test) to obtain the p-values for each patient. False Discovery Rate (FDR Correction) is computed by Benjamini-Hochberg Procedure. Based on the output of the FDR method, we identified the top ten genes that have CpGs with statistically differentially methylated. Such potential epigenetic biomarker includes WBP11, LDHB, DHX29, NUB1, C2orf67, SNAI2, MTOR, CLCN3, SYNJ2BP, and TTK. Two sample KS test and FDR Correction code is available at: https://www.kaggle.com/code/aaravsharma123/gse109905-fdr-pvalues/notebook.