A Hybrid Interpolation Weighted Collaborative Filtering Method for Anti-cancer Drug Response Prediction

Abstract

Individualized therapies ask for the most effective regimen for each patient, while the patients' response may differ from each other. However, it is impossible to clinically evaluate each patient's response due to the large population. Human cell lines have harbored most of the same genetic changes found in patients' tumors, thus are widely used to help understand initial responses of drugs. Based on the more credible assumption that similar cell lines and similar drugs exhibit similar responses, we formulated drug response prediction as a recommender system problem, and then adopted a hybrid interpolation weighted collaborative filtering (HIWCF) method to predict anti-cancer drug responses of cell lines by incorporating cell line similarity and drug similarity shown from gene expression profiles, drug chemical structure as well as drug response similarity. Specifically, we estimated the baseline based on the available responses and shrunk the similarity score for each cell line pair as well as each drug pair. The similarity scores were then shrunk and weighted by the correlation coefficients drawn from the know response between each pair. Before used to find the K most similar neighbors for further prediction, they went through the case amplification strategy to emphasize high similarity and neglect low similarity. In the last step for prediction, cell line-oriented and drug-oriented collaborative filtering models were carried out, and the average of predicted values from both models was used as the final predicted sensitivity. Through 10-fold cross validation, this approach was shown to reach accurate and reproducible outcome for those missing drug sensitivities. We also found that the drug response similarity between cell lines or drugs may play important role in the prediction. Finally, we discussed the biological outcomes based on the newly predicted response values in GDSC dataset.

An Improved Anticancer Drug-Response Prediction Based on an Ensemble Method Integrating Matrix Completion and Ridge Regression

“… 17 and Zhang et al. 18 formulated a drug-response prediction as a recommender system problem, which was solved by two proposed techniques, respectively: the neighbor-based collaborative filtering with global effect removal method and the hybrid interpolation weighted collaborative filtering method; Wang et al. 19 and Guan et al.…”

Network-based Biased Tree Ensembles (NetBiTE) for Drug Sensitivity Prediction and Drug Sensitivity Biomarker Identification in Cancer

“…The dataset has been of particular interest for drug sensitivity prediction and biomarker identification efforts 1,10,16–20 . These include a number of works employing quantitative, statistical and machine learning methods such as: Cell line-similarity and drug-similarity based models 21 multilevel mixed effect models using all drug-cell line combinations 22 , quantitative structure-activity relationship (QSAR) analysis using kernelized Bayesian matrix factorization 23 , lasso and elastic net models for drug sensitivity prediction and target identification 10,24,25 , collaborative filtering based methods for drug sensitivity prediction 26,27 as well as logic models for predictor identification 28 .…”

Gene regulatory network analysis with drug sensitivity reveals synergistic effects of combinatory chemotherapy in gastric cancer

“…The three modules were Computational models that have been developed to identify candidate antitumoral molecules can be used to predict drug sensitivity and identify synergistic combinations of anti-tumoral chemotherapies 31,32 . Among these, network-based models and machine learning-based models are acknowledged as potent methodologies 33,34 . However, the data volume of the known data is limited and more accurate computational algorithms are needed.…”