Improving Drug Response Prediction using Dual Similarity Regularization

Document Type : Research Article

Authors

1 Department of Computer Engineering, Sanandaj Branch, Islamic Azad University, Sanandaj, Iran.

2 Department of Computer Engineering, ShQ.C., Islamic Azad University, Shahr-e Qods, Iran.

3 Department of Computer Engineering, Malard Branch, Islamic Azad University, Tehran, Iran.

Abstract

Personalized medicine aims to identify effective anticancer therapies tailored to individual patients, a core goal of precision oncology. Despite significant advances, achieving reliable and accurate drug response prediction remains challenging due to the complexity and heterogeneity of pharmacogenomic data. Motivated by the principle that similar cell lines exhibit similar responses to similar drugs, we propose an enhanced matrix factorization framework incorporating a novel dual similarity regularization strategy. The proposed Dual Similarity-Regularized Matrix Factorization (DSRMF) model constrains the latent representations of cell lines and drugs to preserve biological and chemical similarity relationships, ensuring that similar entities occupy proximate positions in the latent space while dissimilar ones remain distant. The model integrates two-dimensional (2D) and three-dimensional (3D) chemical structural features to construct a refined drug similarity matrix and was trained and validated on processed datasets from the Genomics of Drug Sensitivity in Cancer (GDSC) and the Cancer Cell Line Encyclopedia (CCLE). Experimental results demonstrate that DSRMF achieves robust predictive performance, with an average Pearson correlation coefficient (PCC) of approximately 0.96 and a root mean square error (RMSE) of 0.30, indicating a strong correlation between predicted and observed drug responses. These findings confirm that incorporating dual similarity regularization and heterogeneous biological information enhances both predictive accuracy and interpretability. Overall, DSRMF advances drug response modeling and provides a scalable framework for integrating multi-dimensional biological data to improve personalized cancer treatment strategies.

Keywords

Main Subjects

References

[1] Paul Geeleher and Nancy J. Cox and R. Stephanie Huang. Cancer biomarker discovery is improved by accounting for variability in general levels of drug sensitivity in pre-clinical models. Genome Biology. 17: 190, 2016. [DOI ]
[2] Francesco Iorio and Theo A. Knijnenburg and Daniel J. Vis and Graham R. Bignell and Michael P. Menden and Michael Schubert and others. A landscape of pharmacogenomic interactions in cancer. Cell. 166: 740--754, 2016. [DOI ]
[3] Jordi Barretina and Giordano Caponigro and Nicolas Stransky and others. The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature. 483: 603--607, 2012. [DOI ]
[4] Mathew J. Garnett and Elena J. Edelman and Sonja J. Heidorn and others. Systematic identification of genomic markers of drug sensitivity in cancer cells. Nature. 483: 570--575, 2012. [DOI ]
[5] Lindsay C. Stetson and Taylor Pearl and Yanwen Chen and Jill S. Barnholtz-Sloan. Computational identification of multi-omic correlates of anticancer therapeutic response. BMC Genomics. 15(Suppl 7): S2, 2014. [DOI ]
[6] Amrita Basu and Ramkrishna Mitra and Han Liu and Stuart L. Schreiber and Paul A. Clemons. RWEN: response-weighted elastic net for prediction of chemosensitivity of cancer cell lines. Bioinformatics. 34: 3332--3339, 2018. [DOI ]
[7] Muhammad Ammad-ud-din and Elisabeth Georgii and Mehmet Gönen and Tuomo Laitinen and Olli Kallioniemi and Krister Wennerberg and Antti Poso and Samuel Kaski. Integrative and personalized QSAR analysis in cancer by kernelized Bayesian matrix factorization. Journal of Chemical Information and Modeling. 54: 2347--2359, 2014. [DOI ]
[8] Nai-Na Guan and Yan Zhao and Chao-Chao Wang and Jian-Qiang Li and Xin Chen and Xu Piao. Anticancer drug response prediction in cell lines using weighted graph regularized matrix factorization. Molecular Therapy - Nucleic Acids. 17: 164--174, 2019. [DOI ]
[9] Kris Matlock and Carlos De Niz and Raziur Rahman and Souparno Ghosh and Ranadip Pal. Investigation of model stacking for drug sensitivity prediction. BMC Bioinformatics. 19(Suppl 7): 71, 2018. [DOI ]
[10] Antti Honkela and Mrinal Das and Arttu Nieminen and Onur Dikmen and Samuel Kaski. Efficient differentially private learning improves drug sensitivity prediction. Biology Direct. 13: 1, 2018. [DOI ]
[11] Mehmet Gönen and Adam A. Margolin. Drug susceptibility prediction against a panel of drugs using kernelized Bayesian multitask learning. Bioinformatics. 30: i556--i563, 2014. [DOI ]
[12] Han Yuan and Ivan Paskov and Hristo Paskov and Alvaro J. González and Christina S. Leslie. Multitask learning improves prediction of cancer drug sensitivity. Scientific Reports. 6: 31618, 2016. [DOI ]
[13] Duong Van and Huy Pham. Drug Response Prediction by Globally Capturing Drug and Cell Line Information in a Heterogeneous Network. Journal of Molecular Biology. 430: 2993--3004, 2018. [DOI ]
[14] Prakash Shivakumar and Michael Krauthammer. Structural similarity assessment for drug sensitivity prediction in cancer. BMC Bioinformatics. 10(Suppl 9): S17, 2009. [DOI ]
[15] Sebo Kim and Varsha Sundaresan and Lei Zhou and Tamer Kahveci. Integrating domain specific knowledge and network analysis to predict drug sensitivity of cancer cell lines. PLoS One. 11: e0162173, 2016. [DOI ]
[16] Soufiane Mourragui and Marco Loog and Mark A. van de Wiel and Marcel J. T. Reinders and Lodewyk F. A. Wessels. PRECISE: a domain adaptation approach to transfer predictors of drug response from pre-clinical models to tumors. Bioinformatics. 35: i510--i519, 2019. [DOI ]
[17] Zuoli Dong and Naiqian Zhang and Chun Li and Haiyun Wang and Yun Fang and Jun Wang and Xiaoqi Zheng. Anticancer drug sensitivity prediction in cell lines from baseline gene expression through recursive feature selection. BMC Cancer. 15: 489, 2015. [DOI ]
[18] Elisabetta Fersini and Enza Messina and Francesco Archetti. A p-median approach for predicting drug response in tumour cells. BMC Bioinformatics. 15: 353, 2014. [DOI ]
[19] Lin Wang and Xiaozhong Li and Louxin Zhang and Qiang Gao. Improved anticancer drug response prediction in cell lines using matrix factorization with similarity regularization. BMC Cancer. 17: 513, 2017. [DOI ]
[20] Chun Wei Yap. PaDEL-descriptor: An open source software to calculate molecular descriptors and fingerprints. Journal of Computational Chemistry. 32: 1466--1474, 2011. [DOI ]
[21] Naiqian Zhang and Haiyun Wang and Yun Fang and Jun Wang and Xiaoqi Zheng and X. Shirley Liu. Predicting anticancer drug responses using a dual-layer integrated cell line-drug network model. PLoS Computational Biology. 11: e1004498, 2015. [DOI ]
[22] Isidro Cortés-Ciriano and Gerard J. P. van Westen and Guillaume Bouvier and Michael Nilges and John P. Overington and Andreas Bender and Thérèse E. Malliavin. Improved large-scale prediction of growth inhibition patterns using the NCI60 cancer cell line panel. Bioinformatics. 32: 85--95, 2016. [DOI ]
[23] Muhammad Ammad-ud-din and Elisabeth Georgii and Mehmet Gönen and Tuomo Laitinen and Olli Kallioniemi and Krister Wennerberg and others. Integrative and personalized QSAR analysis in cancer by kernelized Bayesian matrix factorization. Journal of Chemical Information and Modeling. 54: 2347--2359, 2014. [DOI ]
[24] Jordi Barretina and others. The Cancer Cell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity. Nature. 483: 603--607, 2012. [DOI ]
[25] Wanjuan Yang and others. Genomics of Drug Sensitivity in Cancer (GDSC): a resource for therapeutic biomarker discovery in cancer cells. Nucleic Acids Research. 41: D955--D961, 2013. [DOI ]
[26] Brent M. Kuenzi and Jisoo Park and Samson H. Fong and Kyle S. Sanchez and John Lee and Jason F. Kreisberg and Jianzhu Ma and Trey Ideker. Predicting drug response and synergy using a deep learning model of human cancer cells. Cancer Cell. 38: 672--684.e6, 2020. [DOI ]
[27] Yoosup Chang and Hyejin Park and Hyun-Jin Yang and Seungju Lee and Kwee-Yum Lee and Tae Soon Kim and Jongsun Jung and Jae-Min Shin. Cancer drug response profile scan (CDRscan): a deep learning model that predicts drug effectiveness from cancer genomic signature. Scientific Reports. 8: 8857, 2018. [DOI ]
[28] Thin Nguyen and Hang Le and Thomas P. Quinn and Tri Nguyen and Thuc Duy Le and Svetha Venkatesh. GraphDTA: predicting drug-target binding affinity with graph neural networks. Bioinformatics. 37: 1140--1147, 2021. [DOI ]
[29] Tianyi Ma and Qianmu Liu and Hui Li and Minghan Zhou and Rui Jiang and Xiuzhen Zhang. DualGCN: a dual graph convolutional network model to predict cancer drug response. BMC Bioinformatics. 23(Suppl 4): 129, 2021. [DOI ]
[30] Brent M. Kuenzi and Jisoo Park and Samson H. Fong and Kyle S. Sanchez and John Lee and Jason F. Kreisberg and Jianzhu Ma and Trey Ideker. Predicting Drug Response and Synergy Using a Deep Learning Model of Human Cancer Cells. Cancer Cell. 38(5): 672--684.e6, 2020. [DOI ]
[31] Minghui Li and Yifei Wang and Ruiyuan Zheng and Xinghua Shi and Yixin Li and Feng-Xu Wu and Jianxin Wang. DeepDSC: A Deep Learning Method to Predict Drug Sensitivity of Cancer Cell Lines. IEEE/ACM Transactions on Computational Biology and Bioinformatics. 18: 575--582, 2021. [DOI ]
[32] Chong Wang and David M. Blei. Collaborative topic modeling for recommending scientific articles. Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining. 448--456, 2011. [DOI ]
[33] Vikas Sindhwani and Partha Niyogi and Mikhail Belkin. Beyond the point cloud: from transductive to semi-supervised learning. Proceedings of the 22nd international conference on Machine learning. 824--831, 2005. [DOI ]
[34] Pengyong Li and Zhengxiang Jiang and Tianxiao Liu and Xinyu Liu and Hui Qiao and Xiaojun Yao. Improving drug response prediction via integrating gene relationships with deep learning. Briefings in Bioinformatics. 25: bbae153, 2024. [DOI ]
[35] Rui Miao and Bing-Jie Zhong and Xin-Yue Mei and Xin Dong and Yang-Dong Ou and Yong Liang and Hao-Yang Yu and Ying Wang and Zi-Han Dong. A semi-supervised weighted SPCA- and convolution KAN-based model for drug response prediction. Frontiers in Genetics. 16: 1532651, 2025. [DOI ]
[36] Huimin Luo and Chunli Zhu and Jianlin Wang and Ge Zhang and Junwei Luo and Chaokun Yan. Prediction of drug–disease associations based on reinforcement symmetric metric learning and graph convolution network. Frontiers in Pharmacology. 15: 1337764, 2024. [DOI ]
[37] Deepa Kumari and Dhruv Agrawal and Arjita Nema and Nikhil Raj and Subhrakanta Panda and Jabez Christopher and Jitendra Kumar Singh and Sachidananda Behera. A study on improving drug-drug interactions prediction using convolutional neural networks. Applied Soft Computing. 166: 112242, 2024. [DOI ]
[38] Hui Liu and Feng Wang and Jian Yu and Yong Pan and Cheng Gong and Lei Zhang and Lihua Zhang. DBDNMF: A dual branch deep neural matrix factorization method for drug response prediction. PLoS Computational Biology. 20: e1012012, 2024. [DOI ]
[39] Minwoo Pak and Dongmin Bang and Inyoung Sung and Sun Kim and Sunho Lee. DGDRP: drug-specific gene selection for drug response prediction via re-ranking through propagating and learning biological network. Frontiers in Genetics. 15: 1441558, 2024. [DOI ]
Journal of Computing and Security
Volume 12, Issue 2 - Serial Number 2
September 2025
Pages 93-105

Files

History

  • Receive Date: 29 March 2025
  • Revise Date: 01 November 2025
  • Accept Date: 28 December 2025

Share

How to cite

Statistics