Back

Pleiotropy and repeated mutation drove convergent domestication of grain amaranth

Winkler, T. S.; Kadner, M.; Graf, C.; Lentz, R.; Martinez, J.; Singh, A.; Stetter, M. G.

2026-02-27 evolutionary biology
10.64898/2026.02.26.708151 bioRxiv
Show abstract

Humans reshaped environments for domesticated plants, yet the extent of human agency in the domestication process remains unresolved. Color variation has been used as evidence for an intended process by humans, as the trait has been mainly linked to visual properties. In the ancient pseudocereal grain amaranth major domestication traits have not changed, but the seed color changed repeatedly from dark to white during multiple domestication processes. Here, we show that the transition to white seed color was of major ecological significance beyond human preference through mechanistic links to seed germination. We identify major metabolomic and transcriptomic changes in the proanthocyanidin pathway leading to loss of seed color and dormancy in domesticated amaranth. Despite the potentially large mutational target size of the proanthocyanidin pathway, the repeated mutation in a single gene led to the domestication phenotype. Multiple independent knock-out mutations in this gene reveal the repeated selection for the loss of seed dormancy in different parts of the Americas. Competition experiments show that the pleiotropic link between seed color and seed dormancy provides a competitive advantage to white seeds in agricultural environments. The reduction of proanthocyanidin content in the seed is observed in numerous crops, including rice and beans, suggesting a common ecological adaptation after early humans altered their environment through changing lifestyles. Our findings establish a direct mechanistic link between individual mutations, pathway-level biochemical reconfiguration, and ecological fitness, illustrating how plants adapted to human-modified environments largely without intentional human agency.

Matching journals

The top 4 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.