Back

Accelerated Evolution of Context-Dependent Gene Expression in Primate Immune Cells

Durvasula, A.; Longtin, A.; Arner, A. M.; Köhler, K.; Rivas-Gonzalez, I.; Tung, J.; Housman, G.; Lea, A. J.

2026-09-24 evolutionary biology
10.64898/2026.09.23.753790 bioRxiv
Show abstract

Regulatory variation is widely recognized as a driver of evolution within and between species, including in the primate lineage. Evidence for potentially adaptive regulatory evolution in primates comes from both sequence-based tests for selection and cross-species comparisons of gene expression. However, comparative studies of gene expression have tended to focus on baseline cellular states, despite growing evidence that context-dependent gene regulation is important. Here, we investigate the evolution of gene expression in lymphoblastoid cell lines (LCLs) from 6 primate lineages (humans, chimpanzees, bonobos, western gorillas, Sumatran orangutans, and rhesus macaques) using data from both control and 5 perturbed cellular conditions (n=22-25 individuals per condition). Analyzing baseline expression data, we find that genes involved in immunity often exhibit lineage-specific transcriptional shifts, and that these genes occur near regions with lineage-specific sequence acceleration and lineage-specific epigenomic states. However, the subset of genes that respond to in vitro perturbations have evolved at significantly higher rates than environmentally-insensitive genes, a finding we replicate in an independent four-species dataset of primary white blood cells (n=26). Together, our findings support the argument that focusing on baseline states alone provides a biased view of gene regulatory evolution, and suggest that studying gene regulation in dynamic contexts can enrich our understanding of primate evolution.

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.