Network-based drug sensitivity prediction

Abstract

Background Drug sensitivity prediction and drug responsive biomarker selection on high-throughput genomic data is a critical step in drug discovery. Many computational methods have been developed to serve this purpose including several deep neural network models. However, the modular relations among genomic features have been largely ignored in these methods. To overcome this limitation, the role of the gene co-expression network on drug sensitivity prediction is investigated in this study. Methods In this paper, we first introduce a network-based method to identify representative features for drug response prediction by using the gene co-expression network. Then, two graph-based neural network models are proposed and both models integrate gene network information directly into neural network for outcome prediction. Next, we present a large-scale comparative study among the proposed network-based methods, canonical prediction algorithms (i.e., Elastic Net, Random Forest, Partial Least Squares Regression, and Support Vector Regression), and deep neural network models for drug sensitivity prediction. All the source code and processed datasets in this study are available at https://github.com/compbiolabucf/drug-sensitivity-prediction. Results In the comparison of different feature selection methods and prediction methods on a non-small cell lung cancer (NSCLC) cell line RNA-seq gene expression dataset with 50 different drug treatments, we found that (1) the network-based feature selection method improves the prediction performance compared to Pearson correlation coefficients; (2) Random Forest outperforms all the other canonical prediction algorithms and deep neural network models; (3) the proposed graph-based neural network models show better prediction performance compared to deep neural network model; (4) the prediction performance is drug dependent and it may relate to the drug’s mechanism of action. Conclusions Network-based feature selection method and prediction models improve the performance of the drug response prediction. The relations between the genomic features are more robust and stable compared to the correlation between each individual genomic feature and the drug response in high dimension and low sample size genomic datasets.

Pan-Cancer Prediction of Cell-Line Drug Sensitivity Using Network-Based Methods

“…In this study, we developed a network-based method for predicting the drug sensitivity of pan-cancer cell lines in the GDSC database. Several studies have proposed network-based methods for drug response prediction on single omics data [20,21], whereas the current study used multi-modal genomic and cheminformatic data. The CDCN modeling introduced by Wei et al [26] and its extended method [19] also used genomic and cheminformatic data [26].…”

“…Sensitivity and resistance scores were then computed for each cell-line drug pair [20]. Ahmed et al employed a network-based feature selection method using a gene co-expression network [21]. The resulting output was then used in neural network models for drug-response prediction.…”

An overview of machine learning methods for monotherapy drug response prediction

“…Ahmed et al. [ 74 ] employed two different GNN methods to use GE and a gene co-expression network, i.e. a network representing the correlation between the expression of gene pairs, to construct a GNN.…”

The International Conference on Intelligent Biology and Medicine (ICIBM) 2020: Data-driven analytics in biomedical genomics

“…In "Network-based Drug Sensitivity Prediction", Ahmed et al [4] explored network-based methods for drug sensitivity prediction, including a newly developed method that first used gene coexpression networks to extract representative features and then used graph-based neural network models for drug response prediction. Applying this method to RNA-seq data in non-small cell lung cancer cell lines with treatments of 50 different drugs, this study demonstrated that combining network-based feature selection with graph-based prediction methods, they were able to improve the performance of predicting drug sensitivity and response.…”