Posted By: Sarah Ratzel, PhD, Science Editor, AJHG
Each month, the editors of The American Journal of Human Genetics interview an author of a recently published paper. This month, we check in with Audrey Hendricks to discuss her paper, “Summix: A method for detecting and adjusting for population structure in genetic summary data”.
AJHG: What caused you to start working on this project?
AH: Over the past few years, I have been thinking a lot about how publicly available genetic summary data are incredibly useful and also not equally useful for all researchers, individuals, or studies. This is especially true for studies or individuals with admixed ancestry or ancestry that is underrepresented in genetic databases and studies. In such cases, publicly available genetic summary data are limited and the data available are often a mismatch for the ancestry of the individual or sample. This inequity makes it difficult to complete high-quality research in the very samples and individuals where more high-quality research is needed.
At the same time, I was also working to develop methods that take advantage of the surprisingly large amount of information that is retained in genetic summary data, especially when using multiple variants. I have a few candid colleagues that I like to bounce new ideas off of to check that the work is of broad interest. After talking with them about the initial idea of Summix, I knew this was a project to run with.
AJHG: What about this paper most excites you?
AH: I am really excited about the high level of accuracy and speed of Summix to estimate mixture proportions. The performance combined with Summix’s simplicity opens the door for many possible extensions and applications using only summary data such as estimating fine-scale or local ancestry, estimating mixture groups other than ancestral populations (e.g. genetic predisposition to disease), post-hoc adjustment of GWAS estimates, and clinical applications, among others.
I am also thrilled at how many undergraduate researchers made a meaningful contribution to this research. Seven in total! It was so rewarding to introduce them to statistical genetics and see them grow as researchers; many have continued onto graduate programs.
AJHG: Thinking about the bigger picture, what implications do you see from this work for the larger human genetics community?
AH: The framework of matching an external sample or genetic resource to the ancestry of a particular sample or individual, as we do in Summix, is important both for research and for precision medicine. I often think about how most people are a combination of multiple continental or finer scale ancestries. Thus, to really enable precision medicine, matching to each person or sample’s unique ancestral make-up will be important.
AJHG: What advice do you have for trainees/young scientists?
AH: I used to consult for a start-up and a piece of advice that I have kept is to fail early. Too often people (including me!) waste time working on a project only to find out much later that the most important piece isn’t possible. For a new project, identify the linchpin, the central piece. Then design a quick check to see if that central piece will work. If it does, great – keep going with the next crucial piece. If it fails, then you get to move onto the next project or the next idea to answer your question.
The piece of advice I find myself giving the most lately is to remember that your health (e.g. mental, physical, emotional) comes first. Research is important (and exciting), but without health, you and your research will suffer. I find that my best work comes when I am healthy. Find and prioritize activities that support your health. For me, it’s sleep, and getting outside.
AJHG: And for fun, tell us something about your life outside of the lab.
AH: I have two energetic kids (7 and 10). You’ll find us playing board games, making music, and enjoying the beautiful Colorado outdoors. We play soccer (I even coach), swim, and ski. My son and husband made a frisbee golf course around the neighborhood during the pandemic summer of 2020 so sometimes we’ll play a round after dinner. Since fruit season has just started, you might also find me making a pie – today was strawberry rhubarb.
Audrey Hendricks, PhD, is an associate professor in the Department of Mathematical and Statistical Sciences at the University of Colorado Denver.