Substance Use Disorder- Ancestral Diversity is Needed in Genetic Studies
In our 5th Fresh Perspectives webinar with Amara Davis we explore the application of machine learning in medicine utilizing genetics to better understand substance use patterns among individuals. Amara interviews Dr. Alexander Hatoum, a postdoctoral fellow in the lab of Dr. Bogdan & Dr. Agrawal and an analyst for the Psychiatric Genomics Consortium Substance Use Working Group at the Washington University in St. Louis. Alex discusses two of his recent publications entitled, “The addiction risk factor: A unitary genetic vulnerability characterizes substance use disorders and their associations with common correlates” and “Ancestry may confound genetic machine learning: Candidate-gene prediction of opioid use disorder as an example”. Dr. Hatoum’s interest in substance use disorder (SUD) evolved from seeing his friends develop SUD and as such, he has been working to move the science forward.
Alex presents the importance of understanding genetic contributors of SUD (supported by family studies on heritability) and for integrating genetics with environmental factors for a more complete picture of substance use behavior. Like other conditions, such as diabetes, cardiovascular, or psychiatric, in which the interplay of biology and environment is intricate, large scale genetic studies called genome wide associated studies (GWAS) are used to identify key genes driving the underlying biology. Likewise studying the genes of persons who experience addiction can help scientists determine what biologically contributes to the vulnerability. Such knowledge could aid our understanding of substance use versus misuse to improve person-centered individualized treatment strategies. Currently, we have a poor understanding of who will respond well to available addiction medications. Identifying genetic factors contributing to SUD can potentially be applied to improve our understanding of treatment response and to help develop new medications. In persons that respond poorly to current treatments, new life-saving medications are sorely needed.
The impacted community plays an integral role in supporting these population-level studies (GWAS) for which large numbers of samples are needed. Unfortunately, our genetic databases are sorely lacking in diversity and this negatively affects scientists’ ability to make important discoveries. Due to the lack of ancestral diversity in population datasets, Alex found that when attempting to learn about opioid user disorder, the algorithms discriminated against people of non-European descent. Lack of diversity in genetic datasets hampers science and medical progress and continues to further structural inequities of minority groups. Unfortunately, the funding mechanisms to increase ancestry diversity in genetic databases are not well established. Furthermore, repair of trust by the medical community with people of color and diverse ethnic backgrounds is needed. Alex in the “thing he most would change” calls our attention and action to fix the lack of ancestral diversity in genetic datasets and create the funding mechanisms for this.
To see Amara Davis’s full interview with Alexander S. Hatoum, Ph.D., see video below or view the interview on the LMI YouTube Channel: https://youtu.be/bbRB3EDDoW4
Hatoum, A.S., Johnson, E.C., Colbert, S.M.C. et al. The addiction risk factor: A unitary genetic vulnerability characterizes substance use disorders and their associations with common correlates. Neuropsychopharmacol. (2021). https://doi.org/10.1038/s41386-021-01209-w
Hatoum AS, Wendt FR, Galimberti M, Polimanti R, Neale B, Kranzler HR, Gelernter J, Edenberg HJ, Agrawal A. Ancestry may confound genetic machine learning: Candidate-gene prediction of opioid use disorder as an example. Drug Alcohol Depend. 2021 Dec 1;229(Pt B):109115. https://doi.org/10.1016/j.drugalcdep.2021.109115