SELFormerMM: multimodal molecular representation learning via SELFIES, structure, text, and knowledge graph integration
Bioinformatics · 2026
DOI: 10.1093/bioinformatics/btag451 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Erva Ulusoy und 3 weiteren Autor:innen. Beteiligte Einrichtung: Biological Data Science Lab, Department of Computer Engineering, Hacettepe University, Ankara, 06800, Turkey. Hauptthemen laut MEDLINE: Computational Biology, Drug Discovery, Software. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Scientific and Technological Research Council of Turkey, Erva Ulusoy's conference travel expenses to present this work at ECCB 2026.
Abstract (Originalsprache)
Motivation Molecular representation learning is central to computational drug discovery. However, most existing models rely on single-modality inputs, such as molecular sequences or graphs, which capture only limited aspects of molecular behaviour. Yet unifying these modalities with complementary resources, such as textual descriptions and biological interaction networks, into a coherent multimodal framework remains non-trivial, hindering the development of more informative and biologically grounded representations. Results We introduce SELFormer Multimodal (SELFormerMM), a multimodal molecular representation learning framework that integrates SELFIES notations with structural graphs, textual descriptions and knowledge graph–derived biological interaction data. By aligning these heterogeneous views, SELFormerMM effectively captures complementary signals that unimodal approaches often overlook. Our performance evaluation has revealed that SELFormerMM outperforms structure-, sequence- and knowledge-based models on multiple molecular property prediction tasks. Ablation analyses further indicate that effective cross-modal alignment and modality coverage improve the model’s ability to exploit complementary information. Overall, integrating SELFIES with structural, textual and biological context enables richer molecular representations and provides a promising framework for hypothesis-driven drug discovery. Availability and implementation SELFormerMM is available as a programmatic tool at https://github.com/HUBioDataLab/SELFormerMM. Pre-trained and fine-tuned model checkpoints, multimodal pre-training and finetuning datasets including SMILES, textual descriptions and knowledge graph, and their precomputed single-modality embedding files are available at https://huggingface.co/datasets/HUBioDataLab/SELFormerMM.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Biological Data Science Lab, Department of Computer Engineering, Hacettepe University, Ankara, 06800, Turkey.
- Department of Bioinformatics, Graduate School of Health Sciences, Hacettepe University, Ankara, 06100, Turkey.
- Department of Health Informatics, Institute of Informatics, Hacettepe University, Ankara, 06800, Turkey.
ORCID-Profile: Tunca Doğan
Förderung
- Scientific and Technological Research Council of Turkey
- Erva Ulusoy's conference travel expenses to present this work at ECCB 2026
Zitieren
APA
Ulusoy, E., Bostancı, Ş., Deniz, B. E., & Doğan, T. (2026). SELFormerMM: multimodal molecular representation learning via SELFIES, structure, text, and knowledge graph integration. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag451
Vancouver
Ulusoy E, Bostancı Ş, Deniz BE, Doğan T. SELFormerMM: multimodal molecular representation learning via SELFIES, structure, text, and knowledge graph integration. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag451
BibTeX
@article{Ulusoy2026,
author = {Ulusoy, Erva and Bostancı, Şevval and Deniz, Bora Engin and Doğan, Tunca},
title = {SELFormerMM: multimodal molecular representation learning via SELFIES, structure, text, and knowledge graph integration},
journal = {Bioinformatics},
year = {2026},
doi = {10.1093/bioinformatics/btag451}
}
EUVIMED