Author(s)

ZiWei Ning

Affiliation(s)

School of Computer and Control Engineering, Northeast Forestry University, Harbin 150040, Heilongjiang, China.

Corresponding Author

ZiWei Ning

ABSTRACT

The treatment of cancer and other complex diseases require sophisticated therapeutic strategies due to their intricate pathophysiological mechanisms and resistance profiles. While combination therapy has emerged as a cornerstone of precision medicine, many existing computational approaches primarily focus on integrating only simple modalities such as molecular graphs and SMILES sequences, which limits their ability to model the interaction dynamics in an interpretable and biologically meaningful manner. To address these challenges, we propose SGDSynergy that introduces several key innovations in multi-modal fusion. Specifically, in the drug feature extraction stage, we adopt a CLIP-style contrastive learning mechanism to align molecular representations derived from SMILES sequences and molecular graphs, enabling the  generation of semantically enriched and modality-aware drug embeddings. For cell line representation, SGDSynergy incorporates external features derived from DDIs data and heterogeneous graph structures, which significantly enhance the biological context of cancer cell line embeddings. Moreover, in the synergy evaluation module, we introduce a Bayesian attention mechanism integrated with MC-dropout to probabilistically weigh synergistic features, thereby reducing overfitting and explicitly quantifying prediction uncertainty. These integrated innovations allow SGDSynergy to achieve superior accuracy in predicting drug combination effects, especially in capturing non-linear dynamics and rare synergy patterns, offering a biologically informed and clinically actionable framework for designing personalized combination therapies and overcoming treatment resistance.

KEYWORDS

Drug combination; Drug-drug interaction; Transformer; heterogeneou graph; CLIP-style contrastive learning

CITE THIS PAPER

ZiWei Ning. SGDSynergy: leveraging multimodal data for accurate drug combination prediction. Bioinformatics and Computational Medicine. 2025, 2(1): 7-17. DOI: https://doi.org/10.61784/bcm3003.

REFERENCES

[1] Guo Y, Hu H, Chen W, et al. SynergyX: A multi-modality mutual attention network for interpretable drug synergy prediction. Briefings in Bioinformatics, 2024, 25(2).

[2] Rafiei F, Zeraati H, Abbasi K, et al. DeepTraSynergy: Drug combinations using multimodal deep learning with transformers. Bioinformatics, 2023, 39(8).

[3] Pang Y, Chen Y, Lin M, et al. MMSyn: A new multimodal deep learning framework for enhanced prediction of synergistic drug combinations. Journal of Chemical Information and Modeling, 2024, 64(9): 3689–3705.

[4] Zhang L, Yang Q, Wang H. Reinforcement learning for combinatorial optimization in drug synergy prediction. Nature Machine Intelligence, 2023, 5(4): 412–425. https://doi.org/10.1038/s42256-023-00644-2.

[5] Chen D, Wang X, Zhu H, et al. Predicting anticancer synergistic drug combinations based on multi-task learning. BMC Bioinformatics, 2023, 24(1): 448.

[6] Shan W, Shen C, Luo L, et al. Multi-task learning for predicting synergistic drug combinations based on auto-encoding multi-relational graphs. iScience, 2023, 26(10).

[7] Zhao X, Xu J, Shui Y, et al. PermuteDDS: A permutable feature fusion network for drug–drug synergy prediction. Journal of Cheminformatics, 2024, 16(1): 41.

[8] Zhang C, Zang T, Zhao T. KGE-UNIT: Toward the unification of molecular interactions prediction based on knowledge graph and multi-task learning on drug discovery. Briefings in Bioinformatics, 2024, 25(2).

[9] Misra I, Shrivastava A, Gupta A, et al. Cross-stitch networks for multi-task learning. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016: 3994–4003. https://doi.org/10.1109/CVPR.2016.433.

[10] Ziaee SS, Rahmani H, Tabatabaei M, et al. DCGG: Drug combination prediction using GNN and GAE. Progress in Artificial Intelligence, 2024, 13(1): 17–30.

[11] Zhang G, Gao Z, Yan C, et al. KGANSynergy: Knowledge graph attention network for drug synergy prediction. Briefings in Bioinformatics, 2023, 24(3).

[12] Li H, Zou L, Kowah JAH, et al. Predicting drug synergy and discovering new drug combinations based on a graph autoencoder and convolutional neural network. Interdisciplinary Sciences: Computational Life Sciences, 2023, 15: 316–330.

[13] Alaa M, van der Schaar A, Gonzalez J. Bayesian uncertainty quantification for drug response prediction. Journal of Machine Learning Research, 2022, 23(112): 1–35. https://doi.org/10.5555/3586586.3586601.

[14] Wang T, Wang R, Wei L. AttenSyn: An attention-based deep graph neural network for anticancer synergistic drug combination prediction. Journal of Chemical Information and Modeling, 2024, 64(7): 2854–2862.

[15] Zhang P, Tu S. MGAE-DC: Predicting the synergistic effects of drug combinations through multi-channel graph autoencoders. PLoS Computational Biology, 2023, 19(3).

[16] Wu L, Gao J, Zhang Y, et al. A hybrid deep forest-based method for predicting synergistic drug combinations. Cell Reports Methods, 2023, 3(2).

[17] Shen X, Li Z, Liu Y, et al. PEB-DDI: A task-specific dual-view substructural learning framework for drug–drug interaction prediction. IEEE Journal of Biomedical and Health Informatics, 2024, 28(1): 569–579.

[18] Gu J, Bang D, Yi J, et al. A model-agnostic framework to enhance knowledge graph-based drug combination prediction with drug–drug interaction data and supervised contrastive learning. Briefings in Bioinformatics, 2023, 24(5).

[19] Karras T, Aittala M, Laine S. 3D-STMG: Diffusion models for 3D molecular generation. In: Advances in Neural Information Processing Systems (NeurIPS), 2022: 25430–25443. https://doi.org/10.48550/arXiv.2210.09431.

[20] Zhang H, Ek CH, Rattray M, et al. SynBa: Improved estimation of drug combination synergies with uncertainty quantification. Bioinformatics, 2023, 39(Suppl 1): 121–130.

[21] Liu M, Srivastava G, Ramanujam J, et al. Augmented drug combination dataset to improve the performance of machine learning models predicting synergistic anticancer effects. Scientific Reports, 2024, 14(1): 1668.

[22] Huangfu X, Zhang C, Li H, et al. SNSynergy: Similarity network-based machine learning framework for synergy prediction towards new cell lines and new anticancer drug combinations. Computational Biology and Chemistry, 2024, 110: 108054.

[23] Wang X, Zhu H, Chen D, et al. A complete graph-based approach with multi-task learning for predicting synergistic drug combinations. Bioinformatics, 2023, 39(6): btad351.

[24] El Khili MR, Memon SA, Emad A. MARSY: A multitask deep-learning framework for prediction of drug combination synergy scores. Bioinformatics, 2023, 39(4).

[25] Zhou K, Yang J, Manning CD. PromptTuning: Multimodal prompt engineering for drug discovery. In: Proceedings of the AAAI Conference on Artificial Intelligence, 2023: 12987–12995. https://doi.org/10.1609/aaai.v37i10.26432.

[26] Gan Y, Huang X, Guo W, et al. Predicting synergistic anticancer drug combination based on low-rank global attention mechanism and bilinear predictor. Bioinformatics, 2023, 39(10): btad607.

[27] Pan Y, Ren H, Lan L, et al. Review of predicting synergistic drug combinations. Life, 2023, 13(9).

[28] Lakizadeh A, Babaei M. Detection of polypharmacy side effects by integrating multiple data sources and convolutional neural networks. Molecular Diversity, 2022: 1–11.

[29] Besharatifard M, Vafaee F. A review on graph neural networks for predicting synergistic drug combinations. Artificial Intelligence Review, 2024, 57(3): 49.

[30] Rossi E, Chamberlain B, Bronstein M. Temporal graph networks for drug combination evolution analysis. ACM Transactions on Knowledge Discovery from Data, 2023, 17(3): 1–25. https://doi.org/10.1145/3578368.3580001.

[31] Alam W, Tayara H, Chong KT. Unlocking the therapeutic potential of drug combinations through synergy prediction using graph transformer networks. Computational Biology and Medicine, 2024, 170: 108007.

[32] El-Hafeez TA, Shams MY, Elshaier YAMM, et al. Harnessing machine learning to find synergistic combinations for FDA-approved cancer drugs. Scientific Reports, 2024, 14.

[33] Ru Z, Wu Y, Shao J, et al. A dual-modal graph learning framework for identifying interaction events among chemical and biotech drugs. Briefings in Bioinformatics, 2023, 24(5).

[34] Zhang L, Qin Y, Chen M. SMILESynergy: Anticancer drug synergy prediction based on transformer pre-trained model. Journal of Biomedical Engineering, 2023, 40(3): 544–551.

[35] Rani P, Dutta K, Kumar V. Drug synergy model for malignant diseases using deep learning. Journal of Bioinformatics and Computational Biology, 2023, 21(3): 2350014:1–2350014:18.

[36] Li T, Shetty S, Kamath A, et al. CancerGPT for few-shot drug pair synergy prediction using large pretrained language models. npj Digital Medicine, 2024, 7(1).

[37] Preuer K, Lewis RPI, Hochreiter S, et al. DeepSynergy: Predicting anti-cancer drug synergy with deep learning. Bioinformatics, 2018, 34(9): 1538–1546.

[38] Wang X, Zhu H, Jiang Y, et al. PRODeepSyn: Predicting anticancer synergistic drug combinations by embedding cell lines with protein–protein interaction network. Briefings in Bioinformatics, 2022, 23(2).

[39] Prada-Gracia D, Huerta-Ypez S, Moreno-Vargas LM. Application of computational methods for anticancer drug discovery, design, and optimization. Boletín Médico del Hospital Infantil de México, 2016, 73(6): 411–423.

[40] Chen T, Guestrin C. XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, CA, USA, 2016: 785–794.