iMFP-LG: Identify Novel Multi-functional Peptides Using Protein Language Models and Graph-based Deep Learning

2024/11/25 by Jiawei Luo, Kejuan Zhao, Junjie Chen +7 · 1 voice · 1 citation
Biochemistry, Genetics and Molecular Biology · #Chemical Synthesis and Analysis #Machine Learning in Bioinformatics #vaccines and immunoinformatics approaches

paper · pdf · doi:10.1093/gpbjnl/qzae084

openalex created_date 2024/11/25 · openalex publication_date 2024/11/25 · openalex updated_date 2026/08/01

Abstract

Functional peptides are short amino acid fragments that have a wide range of beneficial functions for living organisms. The majority of previous studies have focused on mono-functional peptides, but an increasing number of multi-functional peptides have been discovered. Although there have been enormous experimental efforts to assay multi-functional peptides, only a small portion of millions of known peptides has been explored. The development of effective and accurate techniques for identifying multi-functional peptides can facilitate their discovery and mechanistic understanding. In this study, we presented iMFP-LG, a method for multi-functional peptide identification based on protein language models (pLMs) and graph attention networks (GATs). Our comparative analyses demonstrated that iMFP-LG outperformed the state-of-the-art methods in identifying both multi-functional bioactive peptides and multi-functional therapeutic peptides. The interpretability of iMFP-LG was also illustrated by visualizing attention patterns in pLMs and GATs. Regarding the outstanding performance of iMFP-LG on the identification of multi-functional peptides, we employed iMFP-LG to screen novel peptides with both anti-microbial and anti-cancer functions from millions of known peptides in the UniRef90 database. As a result, eight candidate peptides were identified, among which one candidate was validated to process both anti-bacterial and anti-cancer properties through molecular structure alignment and biological experiments. We anticipate that iMFP-LG can assist in the discovery of multi-functional peptides and contribute to the advancement of peptide drug design.

Citations

• Temporins, Antimicrobial Peptides from the European Red Frog Rana temporaria
• Basic local alignment search tool
• Multi-Label Learning with Deep Forest
• Pfam: The protein families database in 2021
• Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences
• ProtTrans: Toward Understanding the Language of Life Through Self-Supervised Learning
• SignalP 6.0 predicts all five types of signal peptides using protein language models
• SignalP 6.0 predicts all five types of signal peptides using protein language models
• Large language models generate functional protein sequences across diverse families
• Evolutionary-scale prediction of atomic-level protein structure with a language model

Cited by

• Geometric deep learning assists protein engineering. Opportunities and Challenges

Discussions

• 🧬 A #GPB study introduces iMFP-LG, a deep learning framework combining protein language models and graph attention networks for identifying multi-functional peptides. doi.org/10.1093/gpbj... #DeepL [bsky, 0 points, 0 comments]

Related

• pLMFPPred: a novel approach for accurate prediction of functional peptides integrating embedding from pre-trained protein language model and imbalanced learning
• PepGB: Facilitating peptide drug discovery via graph neural networks
• Multi-Peptide: Multimodality Leveraged Language-Graph Learning of Peptide Properties
• PepTriX: A Framework for Explainable Peptide Analysis through Protein Language Models
• A general language model for peptide function identification