Available online 24 July 2025, 101410

Author links open overlay panelDuong Thanh Tran a 1, Nhat Truong Pham a 1, Nguyen Doan Hieu Nguyen a, Leyi Wei b, Balachandran Manavalan aShow moreHighlights•

We present a framework fusing PLM-based embeddings and CCDs for peptide toxicity prediction.

•

We fuse features from multiple PLMs using cross-modal multi-head attention and Transformers.

•

We design a feature-ranking pipeline to select key descriptors and enhance model performance.

•

HyPepTox-Fuse surpasses existing methods, achieving state-of-the-art toxicity prediction.

•

HyPepTox-Fuse is accessible online, and its implementation is publicly available.

AbstractPeptide-based therapeutics hold great promise for the treatment of various diseases; however, their clinical application is often hindered by toxicity challenges. The accurate prediction of peptide toxicity is crucial for designing safe peptide-based therapeutics. While traditional experimental approaches are time-consuming and expensive, computational methods have emerged as viable alternatives, including similarity-based and machine learning (ML)-/deep learning (DL)-based methods. However, existing methods often struggle with robustness and generalizability. To address these challenges, we propose HyPepTox-Fuse, a novel framework that fuses protein language model (PLM)-based embeddings with conventional descriptors. HyPepTox-Fuse integrates ensemble PLM-based embeddings to achieve richer peptide representations by leveraging a cross-modal multi-head attention mechanism and Transformer architecture. A robust feature ranking and selection pipeline further refines conventional descriptors, thus enhancing prediction performance. Our framework outperforms state-of-the-art methods in cross-validation and independent evaluations, offering a scalable and reliable tool for peptide toxicity prediction. Moreover, we conducted a case study to validate the robustness and generalizability of HyPepTox-Fuse, highlighting its effectiveness in enhancing model performance. Furthermore, the HyPepTox-Fuse server is freely accessible at https://balalab-skku.org/HyPepTox-Fuse/ and the source code is publicly available at https://github.com/cbbl-skku-org/HyPepTox-Fuse/. The study thus presents an intuitive platform for predicting peptide toxicity and supports reproducibility through openly available datasets.Graphical abstractDownload: Download high-res image (473KB)Download: Download full-size imageKeywords

Peptide toxicity

Hybrid framework

Multi-head attention

Transformer

Deep learning

Machine learning

Natural language processing

Recommended articles

Peer review under responsibility of Xi'an Jiaotong University.

1

Both authors contributed equally to this work.

© 2025 Published by Elsevier B.V. on behalf of Xi’an Jiaotong University.