In genome-wide association studies (GWASs), scientists compare the genomes of several individuals to detect genes associated with a specific trait, in this case tobacco or alcohol use. Until now, only a few genetic variants have been consistently linked to these behaviors. Saunders et al. used a multi-ancestry meta-regression method to meta-analyze GWAS data from 60 different cohorts including African, European, American and East Asian ancestries, and discover genes that regulate the development of tobacco and alcohol use. This GWAS method not only accounted for ancestry as a variable but also demonstrated more power for identifying the related genetic loci given the large sample size, compared with previous studies. “We have found thousands of genetic variants associated with tobacco and alcohol use behaviors, most of which have consistent effect sizes across genetic ancestry, suggesting a generally similar genetic architecture. There have been other trans-ancestry GWASs reporting similar patterns but none of this size for complex behaviors that can be so heavily influenced by environmental and cultural factors,” explains Gretchen Saunders, co-first author of the study.
For nearly 200 genetic loci, the researchers identified a variant that affected the risk of tobacco or alcohol use, thereby opening the route for follow-up studies on the functional role of these genetic variants. Although the effect size of most related variants was consistent across different ancestries, the accuracy of polygenic risk scores — the cumulative number of variants that account for a heritable risk of developing a specific trait — based on European populations was lower in non-European individuals. This highlights the importance of including many ancestries in GWASs. The discovery of several genetic loci that influence these behavioral traits is in line with observations of other traits, such as height and blood lipid levels, and are independent of cultural background or other factors that may influence smoking and drinking behaviors.
This is a preview of subscription content, access via your institution