Back

Trans-ancestry meta-analysis improves performance of genetic scores for multiple adiposity-related traits in East Asian populations

Fairhurst-Hunter, Z.; Lin, K.; Millwood, I. Y.; Pozarickij, A.; Chen, T.-T.; Torres, J. M.; Luan, J.-a.; Kartsonaki, C.; Gan, W.; Mahajan, A.; Du, H.; Sohoni, R.; Guo, Y.; Sansome, S.; Yang, L.; Yu, C.; Chen, Y.; Lv, J.; Hemani, G.; Koido, M.; Kamatani, Y.; Spracklen, C. N.; Gordon-Larsen, P.; Koprulu, M.; Meng, X.; Kuchenbaecker, K.; Fatumo, S.; Bhatta, L.; Brumpton, B.; Alegre-Diaz, J.; Kuri-Morales, P.; Tapia-Conyer, R.; Graham, S. E.; Willer, C. J.; Neville, M. J.; Karpe, F.; Graff, M.; North, K. E.; Loos, R. J. F.; Haiman, C. A.; Peters, U.; Buskye, S.; Gignoux, C. R.; Wojcik, G.; Lin, Y.

2022-07-07 genetic and genomic medicine
10.1101/2022.07.05.22277254 medRxiv
Show abstract

Genome-wide association studies (GWAS) in predominately European-ancestry (EUR) populations have identified numerous genetic variants associated with adiposity-related traits. An emerging challenge is the limited transferability of genetic scores constructed based on GWAS results from one ancestry for trait prediction in other ancestries. We performed trans-ancestry meta-analysis (TAMA) for eight adiposity-related traits using genetic data from 96,124 East Asian (EAS) and 443,359 EUR individuals. We identified >1400 genomic regions significantly associated with one or more traits. Despite EAS comprising only [~]20% of the study population, genetic scores constructed from the trans-ancestry (TA) results accounted for between 30% and 79% more variation in the adiposity traits in EAS compared with scores derived from the EUR GWAS alone. Furthermore, TA scores also modestly improved variance explained in African/African American, Hispanic and South Asian populations. Our findings highlight the utility of TAMA for increasing variance explained by genetic scores across populations of different ancestries.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.