*Article* **DNA Methylation Profiles of Vegans and Non-Vegetarians in the Adventist Health Study-2 Cohort**

**Fayth L. Miles 1,2,3,4, Andrew Mashchak 1, Valery Filippov 4, Michael J. Orlich 1,2,5, Penelope Duerksen-Hughes 4, Xin Chen 4,6, Charles Wang 4,6, Kimberly Siegmund <sup>7</sup> and Gary E. Fraser 1,2,5,\***


Received: 15 October 2020; Accepted: 23 November 2020; Published: 30 November 2020

**Abstract:** We sought to determine if DNA methylation patterns differed between vegans and non-vegetarians in the Adventist Health Study-2 cohort. Genome-wide DNA methylation derived from buffy coat was profiled in 62 vegans and 142 non-vegetarians. Using linear regression, methylation of CpG sites and genes was categorized or summarized according to various genic/intergenic regions and CpG island-related regions, as well as the promoter. Methylation of genes was measured as the average methylation of available CpG's annotated to the nominated region of the respective gene. A permutation method defining the null distribution adapted from Storey et al. was used to adjust for false discovery. Differences in methylation of several CpG sites and genes were detected at a false discovery rate < 0.05 in region-specific and overall analyses. A vegan diet was associated predominantly with hypomethylation of genes, most notably methyltransferase-like 1 (*METTL1*). Although a limited number of differentially methylated features were detected in the current study, the false discovery method revealed that a much larger proportion of differentially methylated genes and sites exist, and could be detected with a larger sample size. Our findings suggest modest differences in DNA methylation in vegans and non-vegetarians, with a much greater number of detectable significant differences expected with a larger sample.

**Keywords:** DNA methylation; epigenetics; Adventist Health Study-2; vegetarian diet; linear regression; permutation
