Search bioRxiv⌕ Search

Biology subjects

Bormke, H.

Publications and source records attributed to Bormke, H..

2 recordsLinked to original sources

Deep sequence-function mapping reveals computationally unpredictable epistatic trajectories that drive laboratory evolution

The black-box nature of directed evolution has been constraining advances in enzyme engineering and the wider application of engineered biocatalysts. While conventional deep sequence-function mapping has delivered insights into sequence space architecture, it lacks the ability to analyse adaptive landscapes and fully grasp the long-term impact of epistasis on the accessibility of evolutionary trajectories. Here, we introduce Deep Evolutionary Trajectory Mapping (dETM) which tracks the winners of a multi-round evolutionary campaign to map catalytic rates across the adaptive landscape of an imine reductase. We identified founder variants as stepping stones towards fitness peaks and recognise positive epistasis as their competitive advantage over alternative lineages. State-of-the-art machine learning models could not predict the resulting winning trajectories, which highlights the importance of decoding design principles for successful evolution experimentally. Permissive selection and ultrahigh-throughput screening emerge as foundations of derisking landscape traversal by keeping multiple paths accessible and avoiding diminishing returns.

biochemistry↗

Unsupervised protein language models learn patterns of enzyme function

While enormous amounts of sequence information have become available, assignment of sequence to a particular enzymatic function has remained elusive. Here we describe a framework that drives a general protein language model to find a target reaction without specific training, using an initial bridgehead protein. At the heart of this framework is PLM-clust, an algorithm that employs k-means on top of protein language model embeddings to convert sequence space into functional reservoirs of latent space, and samples from these clusters based on accelerated zero-shot scoring. We demonstrate PLM-clust in a recursive discovery process (with enzyme hit rates quickly rising to >90%), segmenting isofunctional reservoirs and exploring them in greater detail. This approach - exemplified for glycosyl hydrolases (a xylanase, >100-fold activity increase) and for imine reductases (IREDs, >100-fold increase in catalytic promiscuity profiles) - reliably brings about novel enzymes that are proficient at the catalytic task at hand, reaching deeply into sequence space with a majority of residues exchanged.

synthetic biology↗