The Moving Goalposts: Modeling Evolving Polygenic Architecture to Resolve Conflicts in Ancient Selection Studies

There is tension in the literature between studies showing recent polygenic selection on traits like height, body fat, and cognition, and critics who argue that applying modern polygenic scores to ancient populations is methodologically problematic because genetic architecture likely changed over time. This idea aims to resolve that by treating genetic architecture as an evolving trait. Computational models would be developed to allow allele effect sizes to change over time in response to shifting environments and genetic backgrounds. For example, an allele increasing body fat might have been beneficial pre-agriculture but neutral or harmful after farming began. Incorporating this flexibility would enable re-analysis of ancient DNA time-series data for a more accurate picture of selection, moving beyond simple allele frequency tracking to a nuanced understanding of gene-trait relationships shaped by recent evolution.

References:

1. Pervasive findings of directional selection realize the promise of ancient DNA to elucidate human adaptation. Ali Akbari, Allison R. Barton, S. Gazal, Zheng Li, Mohammadreza Kariminejad, Annabel Perry, Yating Zeng, Alissa Mittnik, Nick Patterson, Matthew Mah, Xiang Zhou, Alkes L. Price, Eric S. Lander, R. Pinhasi, N. Rohland, Swapan Mallick, D. Reich (2024). bioRxiv.
2. Eight thousand years of natural selection in Europe. I. Mathieson, Iosif Lazaridis, N. Rohland, Swapan Mallick, B. Llamas, Joseph K. Pickrell, H. Meller, M. Guerra, Johannes Krause, David, Anthony, Dorcas Brown, Carles Lalueza Fox, Alan Cooper, K. Alt, W. Haak, N. Patterson, David Reich (2015).
3. Quantitative Human Paleogenetics: What can Ancient DNA Tell us About Complex Trait Evolution?. Evan K. Irving-Pease, R. Muktupavela, M. Dannemann, F. Racimo (2021). Frontiers in Genetics.