Bollepalli, S. | Search bioRxiv

bioRxiv2018

EpiSmokEr: A robust classifier to determine smoking status from DNA methylation data

Self-reported smoking status is prone to misclassification due to under-reporting, while biomarkers like cotinine can only measure recent exposure. Smoking strongly influences DNA methylation, with current, former and never smokers exhibiting different methylation profiles. Recently, two approaches were proposed to calculate scores based on smoking-responsive DNA methylation loci, to serve as reliable indicators of long-term exposure and potential biomarkers to estimate smoking behavior. However, these two methodologies need significant improvements to make them globally applicable to all populations and to achieve an optimal classification of individuals with unknown smoking habits. To advance the practical applicability of the smoking-associated methylation signals, we used machine learning methodology to train a classifier for smoking status prediction. We show the prediction performance of our classifier on three independent whole-blood test datasets demonstrating its robustness and global applicability. Furthermore, we show the cross-tissue functionality of our classifier in tissues other than blood. Additionally, we provide the community with an R package, EpiSmokEr, facilitating implementation of our classifier to predict smoking status in future studies.

bioinformatics↗

bioRxiv2018

Smoking induces coordinated DNA methylation and gene expression changes in adipose tissue with consequences for metabolic health

Tobacco smoking is a risk factor for multiple diseases, including cardiovascular disease and diabetes. Many smoking-associated signals have been detected in the blood methylome, but the extent to which these changes are widespread to metabolically relevant tissues, and impact gene expression or cardio-metabolic health, remains unclear.\n\nWe investigated smoking-associated DNA methylation and gene expression variation in adipose tissue from 542 healthy female twins with available well-characterized cardio-metabolic phenotype profiles. We identified 42 smoking-methylation and 42 smoking-expression signals, where five genes (AHRR, CYP1A1, CYP1B1, CYTL1, F2RL3) were both hypo-methylated and up-regulated in smokers. We replicated and validated a proportion of the signals in blood, adipose, skin, and lung tissue datasets, identifying tissue-shared effects. Smoking leaves systemic imprints on DNA methylation after smoking cessation, with stronger but shorter-lived effects on gene expression. We tested for associations between the observed smoking signals and several adiposity phenotypes that constitute cardio-metabolic disease risk. Visceral fat and android/gynoid ratio were associated with methylation at smoking-markers with functional impacts on expression, such as CYP1A1, and in signals shared across tissues, such as NOTCH1. At smoking-signals BHLHE40 and AHRR DNA methylation and gene expression levels in current smokers were predictive of future gain in visceral fat upon smoking cessation.\n\nOur results provide the first comprehensive characterization of coordinated DNA methylation and gene expression markers of smoking in adipose tissue, a subset of which link to human cardio-metabolic health and may give insights into the wide ranging risk effects of smoking across the body.\n\nAuthor SummaryTobacco smoking is the strongest environmental risk factor for human disease. Here, we investigate how smoking systemically changes methylome and transcriptome signatures in multiple tissues in the human body. We observe strong and coordinated epigenetic and gene expression changes in adipose tissue, some of which are mirrored in blood, skin, and lung tissue. Smoking leaves a strong short-lived impact on gene expression levels, while methylation changes are long-lasting after smoking cessation. We investigated if these changes observed in a metabolically-relevant (adipose) tissue had impacts on human disease, and observed strong associations with cardio-metabolic disease traits. Some of the smoking signals could predict future gain in obesity and cardio-metabolic disease risk in current smokers who subsequently go on to quit smoking. Our results provide novel insights into understanding the widespread health consequence of smoking outside the lung.

genomics↗