Back

Non-convergent aridity adaptation despite pervasive linked selection in Eucalyptus

James, M. E.; Britton, T. G.; Mitchell, J. D.; Halliwell, B.; Holland, B.; Wright, I. J.; Ortiz-Barrientos, D.

2026-08-28 evolutionary biology
10.64898/2026.08.25.747154 bioRxiv
Show abstract

Whether independent lineages evolve similar genetic solutions when faced with the same environmental pressure is central to understanding how repeatable and predictable adaptation is. Answering this question is increasingly urgent as climate change intensifies drought and aridity worldwide, a shared pressure to which many species must independently adapt. Here we characterised genomic adaptation in Eucalyptus across three independent species pairs, each comprising two closely related lineages that have diverged from wetter into drier environments. Across all pairs we found concordant genome-wide landscapes of diversity and divergence, and strong evidence for pervasive linked selection. Because linked selection acts most strongly in the same conserved features of the genome, this shared architecture could concentrate differentiation in the same regions across lineages, creating an appearance of convergent adaptation. Despite this, the genomic regions associated with the transition to drier environments were largely non-convergent, with almost no sharing of outlier loci among pairs, demonstrating that each lineage adapted through a largely independent genetic route. Some convergence was instead evident at the level of biological function, indicating that lineages reached similar functional outcomes using different genes. We further identified large genomic islands of differentiation, which were dominated by the signature of linked selection rather than elevated divergence, though several harboured candidate genes within the drought and abscisic acid regulatory networks. Together, our results indicate that adaptation to aridity in Eucalyptus is complex and polygenic, and largely unpredictable at the level of individual loci.

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.