Predicting free energy changes(DDG)is essential for enhancing our understanding of protein evolution and plays a pivotal role in protein engineering and pharmaceutical development.While traditional methods offer valua...Predicting free energy changes(DDG)is essential for enhancing our understanding of protein evolution and plays a pivotal role in protein engineering and pharmaceutical development.While traditional methods offer valuable insights,they are often constrained by computational speed and reliance on biased training datasets.These constraints become particularly evident when aiming for accurate DDG predictions across a diverse array of protein sequences.Herein,we introduce Pythia,a self-supervised graph neural network specifically designed for zero-shot DDG predictions.Our comparative benchmarks demonstrate that Pythia outperforms other self-supervised pretraining models and force field-based approaches while also exhibiting competitive performance with fully supervised models.Notably,Pythia shows strong correlations and achieves a remarkable increase in computational speed of up to 105-fold.We further validated Pythia’s performance in predicting the thermostabilizing mutations of limonene epoxide hydrolase,leading to higher experimental success rates.This exceptional efficiency has enabled us to explore 26 million high-quality protein structures,marking a significant advancement in our ability to navigate the protein sequence space and enhance our understanding of the relationships between protein genotype and phenotype.In addition,we established a web server at https://pythia.wulab.xyz to allow users to easily perform such predictions.展开更多
基金supported by the National Key R&D Program of China(grant no.2023YFA0916000)the National Natural Science Foundation of China(32225002,32170033,and 32422001)+2 种基金the Key Research Program of Frontier Sciences(ZDBS-LYSM014)the Biological Resources Program(KFJ-BRP-009 and KFJ-BRP-017-58)from the Chinese Academy of Sciences,the Informatization Plan of Chinese Academy of Sciences(CAS-WX2021SF-0111)the Youth Innovation Promotion Association CAS(2022086).
文摘Predicting free energy changes(DDG)is essential for enhancing our understanding of protein evolution and plays a pivotal role in protein engineering and pharmaceutical development.While traditional methods offer valuable insights,they are often constrained by computational speed and reliance on biased training datasets.These constraints become particularly evident when aiming for accurate DDG predictions across a diverse array of protein sequences.Herein,we introduce Pythia,a self-supervised graph neural network specifically designed for zero-shot DDG predictions.Our comparative benchmarks demonstrate that Pythia outperforms other self-supervised pretraining models and force field-based approaches while also exhibiting competitive performance with fully supervised models.Notably,Pythia shows strong correlations and achieves a remarkable increase in computational speed of up to 105-fold.We further validated Pythia’s performance in predicting the thermostabilizing mutations of limonene epoxide hydrolase,leading to higher experimental success rates.This exceptional efficiency has enabled us to explore 26 million high-quality protein structures,marking a significant advancement in our ability to navigate the protein sequence space and enhance our understanding of the relationships between protein genotype and phenotype.In addition,we established a web server at https://pythia.wulab.xyz to allow users to easily perform such predictions.
