Back

Analytical and computational solution for the estimation of SNP-heritability in biobank-scale and distributed datasets

Qi, G.-A.; Zhang, Q.-X.; Kang, J.; Li, T.; Xu, X.; Zhang, Z.; Fan, Z.; Liu, S.; Chen, G.-B.

2024-09-24 genetics
10.1101/2024.09.20.614017 bioRxiv
Show abstract

Estimation of heritability has been a routine in statistical genetics, in particular with the increasing sample size such as biobank-scale data and distributed datasets, the latter of which has increasing concerns of privacy. Recently a randomized Haseman-Elston regression (RHE-reg) has been proposed to estimate SNP-heritability, and given sufficient iteration (B) RHE-reg can tackle biobank-scale data, such as UK Biobank (UKB), very efficiently. In this study, we present an analytical solution that balances iteration B and RHE-reg estimation, which resolves the convergence of the proposed RHE-reg in high precision. We applied the method for 81 UKB quantitative traits and estimated their SNP-heritability and test statistics precisely. Furthermore, we extended RHE-reg into distributed datasets and demonstrated their utility in real data application and simulated data. The software for estimating SNP-heritability for biobank-scale data is released: https://github.com/gc5k/gear2.

Matching journals

The top 6 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.