Groundbreaking study co-authored by Penn State researcher finds new genetic clues on alcohol and tobacco use
Airdate: December 15, 2022
Family history or one’s genetics are often a factor in how much a person drinks alcohol, suffers from alcohol use disorder or uses tobacco. It’s one factor but there are others, including the environment a person grew up in or lives in. The largest study of its kind that surveyed more than 3.4 million people and included over a hundred scientists found more than 1,900 genetic variants that had not previously been linked to substance use behaviors.
The study’s co-author is Dajiang Liu (Ph.D), a Professor & Vice Chair for Research, Department of Public Health Sciences, Interim Director, Biomedical Informatics and Artificial Intelligence, Co-Chair, Bioinformatics and Genomics Graduate Program, Penn State College of Medicine, who appeared on The Spark Thursday.
Dr. Liu was asked how much of a role genes play in drinking and smoking,”Genetics plays an important role in many of our behaviors, including smoking and drinking. The number actually varies from studies on it’s not completely conclusive, but generally speaking, it accounts for 50 to 60% of the intra-individual differences of smoking and drinking behaviors.”
How can the data be used? Liu said,”The DNA is a fundamental building block. So there’s a lot of things we could learn from genetics. One thing is we can build our prediction models that can forecast how likely somebody is getting addicted. So we were with a large sample size and with the machine learning and artificial intelligence methods, we were able to build prediction models that takes people’s DNA as input. And then we can predict how likely this person may develop addiction or substance use behaviors later in their life. So DNA is something that rarely changes in our life course. So we can actually use those information to predict a disease risk long before it happens so that we can actually take preventive measures to help those people. Another thing is another useful use of the research is that it can help us develop medications. So some people may carry mutations that make them immune to addiction behaviors, but a lot of the other people may not be that lucky and may not have those genetic variants. So but we can actually develop medications that mimic the mutations, those beneficial mutations, which can potentially help people, you know, broad populations against those disease or, you know, harmful behaviors.
Dr. Liu indicated the research was unique in that it included subjects with non-European ancestry,”A lot of the current genetic research are still primarily focused on samples of European ancestry. But people realize that the genetics that we uncover from studying samples of European ancestry may not be transferable to minority populations. And obviously for scientific reasons, for probably health reasons and for political reasons as well, studying diverse ancestry is critical. And so our study is actually one of the most ancestrally, diverse genetic studies and smoking drinking addiction so far. One thing that we find particularly intriguing why we study, why we include the non-European population is that one, we derive our risk prediction models using samples of European ancestry. It can predict really nicely when we use them in European ancestry populations of European ancestry. But why we apply the same risk prediction model in another ancestry, like African-Americans, for example, those rate risk prediction models have much reduced accuracy. So that actually suggested that in order to improve the application, the research of genetics across different populations, increasing diversity is the only way is it’s a must and it’s the only way that we can achieve that.”