EUVIMEDEUVIMED entdecken
Crossref

Deep cross-modal representation fusion learning for enhanced drug target affinity prediction

Essmily Simon, Sanjay Bankapur

Analytical Biochemistry · 2026-11 · 718 · 116198

DOI: 10.1016/j.ab.2026.116198 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Analytical Biochemistry, 2026-11. Verfasst von Essmily Simon und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Engineering, National Institute of Technology, Puducherry, India. Hauptthemen laut MEDLINE: Drug Discovery, Proteins, Deep Learning. Die Arbeit nennt 43 Literaturangaben. Gefördert von: Google Inc.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Prediction of Drug Target Affinity (DTA) is essential for accelerating computational drug discovery and reducing experimental costs. However, traditional experimental approaches for DTA estimation are resource-intensive and are further challenged by the structural flexibility of both drugs and target proteins. In this work, we propose the PCBERT-GAT-DFFNN-DTA model, a three-stage deep cross-modal representation fusion framework for accurate DTA prediction. In the first stage, variable-length protein sequences are transformed into contextual representations using ProtBERT to obtain fixed-size protein embeddings. Drug molecules are represented using two modalities: sequence-based embeddings generated from ChemBERT and structure-based embeddings learned from molecular graphs using a Graph Attention Network (GAT). In the second stage, each modality is processed through dedicated subnetworks to refine features and reduce dimensionality while preserving modality-specific information. In the final stage, the refined representations are fused and passed to a Deep Feed-Forward Neural Network (DFFNN) to predict drug target binding affinity. The proposed model consistently outperformed most baseline methods under the S1-S3 evaluation settings across the benchmark datasets. Under the more challenging S4 blind setting, the model achieved strong performance on the KIBA dataset and competitive results on the Davis and Metz datasets. Compared with LLMDTA, the proposed approach achieves significant improvements in R2 scores across all datasets, demonstrating its effectiveness in learning complex drug protein interactions for reliable DTA prediction.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Drug DiscoveryProteinsDeep LearningGraph Neural NetworksNeural Networks, ComputerPharmaceutical PreparationsRepresentation Machine Learning

Stichwörter der Autor:innen

Deep feed-forward neural networksDrug discoveryFeature refinementGraph attention networksProtein embeddingsSubnetwork learning

Substanzen

ProteinsPharmaceutical Preparations

Autor:innen und Institutionen

  • Department of Computer Science and Engineering, National Institute of Technology, Puducherry, India.
  • Department of Computer Science and Engineering, National Institute of Technology, Puducherry, India. Electronic address: sanjaybankapur.mit@gmail.com.

Förderung

  • Google Inc

Zitieren

APA

Simon, E., & Bankapur, S. (2026). Deep cross-modal representation fusion learning for enhanced drug target affinity prediction. Analytical Biochemistry, 718, 116198. https://doi.org/10.1016/j.ab.2026.116198

Vancouver

Simon E, Bankapur S. Deep cross-modal representation fusion learning for enhanced drug target affinity prediction. Analytical Biochemistry. 2026;718:116198. doi:10.1016/j.ab.2026.116198

BibTeX
@article{Simon2026,
  author = {Simon, Essmily and Bankapur, Sanjay},
  title = {Deep cross-modal representation fusion learning for enhanced drug target affinity prediction},
  journal = {Analytical Biochemistry},
  year = {2026},
  volume = {718},
  pages = {116198},
  doi = {10.1016/j.ab.2026.116198}
}
Kostenlos starten →