Investigating the Effect that Polygenic Risk Score has on Heritability of Mental Disorders
Beta EEG is an endophenotype for AUD. This summer, I worked with the COGA databank in order to gain a better understanding as to what causes alcoholism. This project used PRS calculated from the COGA databank in order to see the way they corresponded with Beta EEG values. For the most part, high PRS led to high Beta EEG, however, 2 specific families did not fit this model, 20059 and 30018.
Families 20059 and 30018 both had high heritability linkage peaks on chromosomes 6 and 8 respectively. Essentially, this means that we would expect these two families to have high PRS. However, this is not the case. These families have below average genome wide PRS and chromosome specific PRS for having high EEG. This is interesting as previously; polygenic risk was thought to be the primary cause of high EEG (and therefore AUD). Further analysis found that there are no rare variants of large effect causing such high heritability on the chromosomes. These findings lead to the idea that there are also protective alleles that, if missing, can lead to AUD.
Comments
Methods
Hi Nicky, you have a really interesting topic here. I'm curious to know more about your methods. Did you speak to any families with a history of alcoholism to gain insight before performing your research? Ideally, how would you like this research to be used in the future?
Hi Nicky, Really cool…
Hi Nicky,
Really cool project. As per my understanding - polygenic risk scores are very representative of a select diversity of populations (mostly europeans) if I'm not wrong. I realized your sample size fits in well with this availability. I was just wondering how your methods and results would apply to diverse populations of different race/ethnicity etc.
Replying to Zoe and Seth
Hi Zoe, I did not speak with families with a history of alcoholism prior to performing my research. Ideally, this research can help influence not only the way the genetics of alcoholism is studied but also the genetics of other disorders. As of now, there is not much research on or equipment to track protective alleles and hopefully, the data from families 20059 and 30018 can help expand the field.
Hi Seth, you are correct in your understanding of PRS only working for specific populations, however, that is not to say it only works for Europeans. Rather, PRS can be calculated for any ethnicity, as long as their specific markers are being looked for. That being said, a large downside of PRS is that it does not work in datasets where the individuals come from diverse ethnic backgrounds.
Nice presentation, Nicky!
Nice presentation, Nicky!