Cross-family learning transfers thermal adaptation between enzymes
- 1Chalmers University of Technology
Abstract
Thermal adaptation has evolved independently across diverse protein families, yet whether the information underlying this adaptation can be transferred between evolutionarily unrelated proteins remains unknown. Here, we demonstrate that thermostability can be learned from divergent protein families and transferred to enzymes without relying on their evolutionary ancestry. We developed THOR (THermally Optimised Representations), a generative framework that learns background-specific sequence transformations between unpaired mesophilic and thermophilic sequence populations while preserving the identity of the starting protein. Across evolutionarily divergent enzyme families, THOR transferred thermostability without exposure to thermophilic homologues during training. In malate dehydrogenase, excluded from cross-family training, zero-shot designs increased apparent melting temperature by up to 15 , with gains reaching 36 following limited family-specific fine-tuning. Majority of designed variants retained activity following prolonged thermal challenge, while selected variants exhibited nearly threefold higher catalytic turnover than wild type. Molecular dynamics simulations revealed that comparable thermal phenotypes emerge through distinct, background-dependent changes in protein flexibility and stabilising interactions. Together, these findings demonstrate that thermal adaptation encodes transferable information beyond lineage-specific evolutionary solutions, revealing evolution as a source of reusable adaptive principles for engineering phenotypes in proteins whose evolutionary lineages have never acquired them.
Related
- A rapid, machine-learning readout of cochlear neural degeneration that transfers from animals to humans
- Phage-Inducible Chromosomal Jumbo Islands (PICJIs) spreading antimicrobial resistance, host-adaptation factors and defense systems, a novel satellite family of nucleus-forming Chimalliviridae
- Structures of methanogenic schizorhodopsins reveal divergent strategies for proton transport and thermal adaptation
- Evolutionary expansion of the bacterial ompW gene family enables functional diversification during host adaptation
- The MarR-family regulator TilR links biosynthesis of enterotoxin tilimycin with physiologic adaptation in Klebsiella oxytoca
BibTeX
@misc{viknander2026crossfamily,
title = {{Cross-family learning transfers thermal adaptation between enzymes}},
author = {S. Viknander and N. Tatarakis and X. Fu and A. Diaciuc and F. Buric and A. Zelezniak},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.26.761089},
url = {https://www.biorxiv.org/content/10.64898/2026.10.26.761089v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - Cross-family learning transfers thermal adaptation between enzymes
AU - S. Viknander
AU - N. Tatarakis
AU - X. Fu
AU - A. Diaciuc
AU - F. Buric
AU - A. Zelezniak
PY - 2026
DA - 2026/10/26
PB - bioRxiv
DO - 10.64898/2026.10.26.761089
UR - https://www.biorxiv.org/content/10.64898/2026.10.26.761089v1
AB - Thermal adaptation has evolved independently across diverse protein families, yet whether the information underlying this adaptation can be transferred between evolutionarily unrelated proteins remains unknown. Here, we demonstrate that thermostability can be learned from divergent protein families and transferred to enzymes without relying on their evolutionary ancestry. We developed THOR (THermally Optimised Representations), a generative framework that learns background-specific sequence transformations between unpaired mesophilic and thermophilic sequence populations while preserving the identity of the starting protein. Across evolutionarily divergent enzyme families, THOR transferred thermostability without exposure to thermophilic homologues during training. In malate dehydrogenase, excluded from cross-family training, zero-shot designs increased apparent melting temperature by up to 15 , with gains reaching 36 following limited family-specific fine-tuning. Majority of designed variants retained activity following prolonged thermal challenge, while selected variants exhibited nearly threefold higher catalytic turnover than wild type. Molecular dynamics simulations revealed that comparable thermal phenotypes emerge through distinct, background-dependent changes in protein flexibility and stabilising interactions. Together, these findings demonstrate that thermal adaptation encodes transferable information beyond lineage-specific evolutionary solutions, revealing evolution as a source of reusable adaptive principles for engineering phenotypes in proteins whose evolutionary lineages have never acquired them.
ER -