Learning deep representations of enzyme thermal adaptation

TY - JOUR

T1 - Learning deep representations of enzyme thermal adaptation

AU - Li, Gang

AU - Buric, Filip

AU - Zrimec, Jan

AU - Viknander, Sandra

AU - Nielsen, Jens

AU - Zelezniak, Aleksej

AU - Engqvist, Martin K.M.

N1 - Funding Information: GL and JN have received funding from the European Union's Horizon 2020 research and innovation program under the Marie Skłodowska‐Curie program, project PAcMEN (Grant agreement no. 722287). JN also acknowledges funding from the Novo Nordisk Foundation (Grant no. NNF10CC1016517), the Knut and Alice Wallenberg Foundation. The study was supported by SciLifeLab funding and the Swedish Research Council (Vetenskapsrådet) starting Grant no. 2019‐05356. AZ was supported by the Marius Jakulis Jason Foundation and JZ by the Slovenian Research Agency (ARRS) Grant no. J2‐3060 and Public Scholarship, Development, Disability and Maintenance Fund of the Republic of Slovenia Grant no. 11013‐9/2021‐2. The computations were performed on resources at Chalmers Center for Computational Science and Engineering (C3SE) provided by the Swedish National Infrastructure for Computing (SNIC). Mikael Öhman and Thomas Svedberg at C3SE are acknowledged for technical assistance. Funding Information: GL and JN have received funding from the European Union's Horizon 2020 research and innovation program under the Marie Skłodowska-Curie program, project PAcMEN (Grant agreement no. 722287). JN also acknowledges funding from the Novo Nordisk Foundation (Grant no. NNF10CC1016517), the Knut and Alice Wallenberg Foundation. The study was supported by SciLifeLab funding and the Swedish Research Council (Vetenskapsrådet) starting Grant no. 2019-05356. AZ was supported by the Marius Jakulis Jason Foundation and JZ by the Slovenian Research Agency (ARRS) Grant no. J2-3060 and Public Scholarship, Development, Disability and Maintenance Fund of the Republic of Slovenia Grant no. 11013-9/2021-2. The computations were performed on resources at Chalmers Center for Computational Science and Engineering (C3SE) provided by the Swedish National Infrastructure for Computing (SNIC). Mikael Öhman and Thomas Svedberg at C3SE are acknowledged for technical assistance. Publisher Copyright: © 2022 The Authors. Protein Science published by Wiley Periodicals LLC on behalf of The Protein Society.

PY - 2022/12

Y1 - 2022/12

N2 - Temperature is a fundamental environmental factor that shapes the evolution of organisms. Learning thermal determinants of protein sequences in evolution thus has profound significance for basic biology, drug discovery, and protein engineering. Here, we use a data set of over 3 million BRENDA enzymes labeled with optimal growth temperatures (OGTs) of their source organisms to train a deep neural network model (DeepET). The protein-temperature representations learned by DeepET provide a temperature-related statistical summary of protein sequences and capture structural properties that affect thermal stability. For prediction of enzyme optimal catalytic temperatures and protein melting temperatures via a transfer learning approach, our DeepET model outperforms classical regression models trained on rationally designed features and other deep-learning-based representations. DeepET thus holds promise for understanding enzyme thermal adaptation and guiding the engineering of thermostable enzymes.

AB - Temperature is a fundamental environmental factor that shapes the evolution of organisms. Learning thermal determinants of protein sequences in evolution thus has profound significance for basic biology, drug discovery, and protein engineering. Here, we use a data set of over 3 million BRENDA enzymes labeled with optimal growth temperatures (OGTs) of their source organisms to train a deep neural network model (DeepET). The protein-temperature representations learned by DeepET provide a temperature-related statistical summary of protein sequences and capture structural properties that affect thermal stability. For prediction of enzyme optimal catalytic temperatures and protein melting temperatures via a transfer learning approach, our DeepET model outperforms classical regression models trained on rationally designed features and other deep-learning-based representations. DeepET thus holds promise for understanding enzyme thermal adaptation and guiding the engineering of thermostable enzymes.

KW - bioinformatics

KW - deep neural networks

KW - enzyme catalytic temperatures

KW - optimal growth temperatures

KW - protein thermostability

KW - transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85143127960&partnerID=8YFLogxK

U2 - 10.1002/pro.4480

DO - 10.1002/pro.4480

M3 - Article

C2 - 36261883

AN - SCOPUS:85143127960

SN - 0961-8368

VL - 31

JO - Protein Science

JF - Protein Science

IS - 12

M1 - e4480

ER -