EUVIMEDEUVIMED entdecken
Crossref

NeuroPpred-PSCG: A Multimodal Framework Using ProtT5 and Structural Features for Neuropeptide Prediction Based on Gated Cross-Attention Mechanism and Multiscale Gated Convolution

Shengli Zhang iD, Xinyi Zhang

Journal of Chemical Information and Modeling · 27.07.2026 · 66 (14) · 8501-8514

DOI: 10.1021/acs.jcim.6c01123 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, online erschienen am 13.07.2026. Verfasst von Shengli Zhang und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics and Statistics, Xidian University, Xi'an710071, P. R. China. Hauptthemen laut MEDLINE: Neuropeptides, Deep Learning, Computational Biology. Die Arbeit nennt 45 Literaturangaben. Gefördert von: Natural Science Basic Research Program of Shaanxi Province, Xidian University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As fundamental signaling molecules, neuropeptides are involved in diverse physiological regulatory activities. Abnormal expression patterns have been closely linked to neurological and metabolic disorders, which underscores the urgent need for precise and rapid identification methods. Here, we develop NeuroPpred-PSCG, a novel multimodal deep learning framework. At its core, the framework employs both a bidirectional cross-attention mechanism and a gated fusion module to integrate global semantic insights from the protein language model (ProtT5) with local conformational data derived from secondary structure. Furthermore, it utilizes a multiscale gated convolutional network to enhance hierarchical contextual information. On the independent test set, NeuroPpred-PSCG surpasses current leading methods on most metrics, achieving ACC of 94.8%, SN of 95.1%, F1-score of 94.8%, and MCC of 0.896, reflecting robust overall discriminative capability. Compared to the previous best-performing model, NeuroPpred-MSN, our method shows improvements of 1.2% and 0.024 in ACC and MCC, respectively, and most notably, a significant 2.8% gain in SN. Ablation studies, robustness analysis, and interpretability analysis further substantiate the effectiveness of each component, as well as the overall stability and interpretability of NeuroPpred-PSCG. The source code and data sets have been made publicly accessible at https://github.com/xinyizhang186/NeuroPpred-PSCG

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

NeuropeptidesDeep LearningComputational BiologyProtein ConformationModels, MolecularPrediction Algorithms

Substanzen

Neuropeptides

Autor:innen und Institutionen

  • School of Mathematics and Statistics, Xidian University, Xi'an710071, P. R. China.

ORCID-Profile: Shengli Zhang

Förderung

  • Natural Science Basic Research Program of Shaanxi Province
  • Xidian University

Zitieren

APA

Zhang, S., & Zhang, X. (2026). NeuroPpred-PSCG: A Multimodal Framework Using ProtT5 and Structural Features for Neuropeptide Prediction Based on Gated Cross-Attention Mechanism and Multiscale Gated Convolution. Journal of Chemical Information and Modeling, 66(14), 8501-8514. https://doi.org/10.1021/acs.jcim.6c01123

Vancouver

Zhang S, Zhang X. NeuroPpred-PSCG: A Multimodal Framework Using ProtT5 and Structural Features for Neuropeptide Prediction Based on Gated Cross-Attention Mechanism and Multiscale Gated Convolution. Journal of Chemical Information and Modeling. 2026;66(14):8501-8514. doi:10.1021/acs.jcim.6c01123

BibTeX
@article{Zhang2026,
  author = {Zhang, Shengli and Zhang, Xinyi},
  title = {NeuroPpred-PSCG: A Multimodal Framework Using ProtT5 and Structural Features for Neuropeptide Prediction Based on Gated Cross-Attention Mechanism and Multiscale Gated Convolution},
  journal = {Journal of Chemical Information and Modeling},
  year = {2026},
  volume = {66},
  number = {14},
  pages = {8501--8514},
  doi = {10.1021/acs.jcim.6c01123}
}
Kostenlos starten →