NEW YORK – With association analyses that included millions of participants, an international team tracked down a large collection of genetic loci linked to tobacco or alcohol use, bringing in cross-ancestry clues to fine-map the associations and find candidate causal variants.
"[W]e found that a large majority of associated genetic variants showed homogeneous effect size estimates across diverse ancestries, suggesting that the genetic variants associated with substance use affect such individuals similarly," co-senior and co-corresponding authors Scott Vrieze, a psychology researcher at the University of Minnesota, and Dajiang Liu, a public health sciences researcher at Penn State College of Medicine, and their colleagues wrote in a paper published Wednesday in Nature.
The researchers brought together array-based genotypes for 3.4 million individuals with European, African, African American, or East Asian ancestry for genome-wide association and GWAS meta-analyses focused on tobacco use measures — ranging from smoking initiation and smoking cessation to the number of cigarettes smoked per day or age at which individuals began smoking — as well as alcohol use, measured by the number of alcoholic drinks participants reportedly consumed each week.
All told, the team tracked down 3,823 tobacco- or alcohol use-associated variants falling at 2,143 loci. These included more than 1,300 genetic loci that coincided with smoking initiation and 128 loci coinciding with smoking cessation. More than 240 variants found at 140 loci were linked to the number of cigarettes smoked each day, while the age that individuals started smoking corresponded to 33 genomic loci. Almost 500 loci associated with the number of alcoholic drinks that the participants partook of each week.
The multi-ancestry approach not only made it possible to find risk variants that would have been missed otherwise, the researchers explained, but also helped in fine-mapping steps that focused in on almost 600 loci and prioritized genes from pathways involved in everything from neurogenesis, neuronal development, and neuronal differentiation to synaptic function.
Likewise, the team highlighted genes that are expressed in specific brain cell types, bringing in additional insights from a trans-ancestry transcriptome-wide association analysis to link smoking and drinking phenotypes to genes expressed in the brain and other body tissues.
Overall, the researchers saw similar heritability estimates and variant effect sizes across the ancestry groups considered, prompting them to suggest that the "genetic architecture underlying substance use is not markedly different across ancestries." Even so, they explained, polygenic risk scores developed from GWAS data had better performance in individuals with European ancestry than in participants with other ancestral backgrounds.
"[P]olygenic risk scores developed in one ancestry performed poorly in others, highlighting the continued need to increase sample sizes of diverse ancestries to realize any potential benefit of polygenic prediction," the authors wrote, adding that "increases in genetic diversity and consideration of environmental moderators, including cultural factors, will continue to add to our understanding of the genetic architecture of both substance use and related behaviors and diseases."