This is the authors' abstract. We don't add key points for this paper.
Machine learning-based modelling assisting functional-oriented enzyme engineering is normally built on a predefined sequence space. Efficiently defining the determinant amino acid positions upon which the combinatorial mutational library is constructed is still a challenge in protein science. Herein, we present a comprehensive investigation of modifying a recombinant DNA polymerase for efficient incorporating unnatural nucleotides via fine-tuning of functional dynamics, including the identification of key sites/regions outside known active center, machine learning-assisted mutant screening, and the underlying mechanism of kinetics boosting. By using hundreds of training points and only dozens of machine-learning predicted samples, we found that the highly engineered enzyme’s catalytic efficiency can be further improved by at least one order of magnitude by specific mutation of two sites on finger subdomain, 485I and 451L. Contrary to the position 485 which is known to dominate local conformation of B-family DNA polymerases, position 451 is a new active site discovered by our approach. The synergy of 485I and 451L on kinetics boosting is driven by an enhanced dynamical-cooperative regulation mechanism of the state distribution of 'open/closed' conformations and the dynamics of the nucleotide of template DNA for substrate-pairing. These findings extended the boundaries of the functional dynamics engineering of complex protein enzymes and have implications in guiding the design of DNA polymerases for a broad range of biotechnological applications.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Lyu et al. (2022) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: