Multi-Peptide: Multimodality Leveraged Language-Graph Learning of Peptide Properties DOI Creative Commons

Srivathsan Badrinarayanan,

Chakradhar Guntuboina, Parisa Mollaei

et al.

Journal of Chemical Information and Modeling, Journal Year: 2024, Volume and Issue: 65(1), P. 83 - 91

Published: Dec. 19, 2024

Peptides are crucial in biological processes and therapeutic applications. Given their importance, advancing our ability to predict peptide properties is essential. In this study, we introduce Multi-Peptide, an innovative approach that combines transformer-based language models with graph neural networks (GNNs) properties. We integrate PeptideBERT, a transformer model specifically designed for property prediction, GNN encoder capture both sequence-based structural features. By employing contrastive loss framework, Multi-Peptide aligns embeddings from modalities into shared latent space, thereby enhancing the model's predictive accuracy. Evaluations on hemolysis nonfouling data sets demonstrate Multi-Peptide's robustness, achieving state-of-the-art 88.057% accuracy prediction. This study highlights potential of multimodal learning bioinformatics, paving way accurate reliable predictions peptide-based research

Language: Английский

Machine learning methods to study sequence–ensemble–function relationships in disordered proteins DOI Creative Commons
Sören von Bülow, Giulio Tesei, Kresten Lindorff‐Larsen

et al.

Current Opinion in Structural Biology, Journal Year: 2025, Volume and Issue: 92, P. 103028 - 103028

Published: March 12, 2025

Language: Английский

Citations

1
Multi-Peptide: Multimodality Leveraged Language-Graph Learning of Peptide Properties DOI Creative Commons

Srivathsan Badrinarayanan,

Chakradhar Guntuboina, Parisa Mollaei

et al.

Journal of Chemical Information and Modeling, Journal Year: 2024, Volume and Issue: 65(1), P. 83 - 91

Published: Dec. 19, 2024

Peptides are crucial in biological processes and therapeutic applications. Given their importance, advancing our ability to predict peptide properties is essential. In this study, we introduce Multi-Peptide, an innovative approach that combines transformer-based language models with graph neural networks (GNNs) properties. We integrate PeptideBERT, a transformer model specifically designed for property prediction, GNN encoder capture both sequence-based structural features. By employing contrastive loss framework, Multi-Peptide aligns embeddings from modalities into shared latent space, thereby enhancing the model's predictive accuracy. Evaluations on hemolysis nonfouling data sets demonstrate Multi-Peptide's robustness, achieving state-of-the-art 88.057% accuracy prediction. This study highlights potential of multimodal learning bioinformatics, paving way accurate reliable predictions peptide-based research

Language: Английский

Citations

1