Methods Mol Biol. 2025;2947:139-150. doi: 10.1007/978-1-0716-4662-5_7.
ABSTRACT
In the field of bioinformatics, automated function prediction (AFP) for proteins is a significant issue. We propose a computational framework based on a protein language model to provide accurate functional predictions for proteins. By integrating multiple component methods, our approach effectively improves prediction performance. Additionally, we have developed a user-friendly online platform that allows users to obtain prediction results simply by submitting protein sequences, freely available at https://dmiip.sjtu.edu.cn/ng3.0/?returning=true . We provide a detailed guide on how to use the web server and correctly interpret the prediction results. Finally, through a practical example, we demonstrate the superior performance of NetGO 3.0 in predicting protein functions, further showcasing the potential of this framework for protein functional annotation.
PMID:40728611 | DOI:10.1007/978-1-0716-4662-5_7
