Using Deep Neural Network to Predict Drug Sensitivity of Cancer Cell Lines
High-throughput screening technology has provided a large amount of drug sensitivity data for hundreds of compounds on cancer cell lines. In this study, we have developed a deep learning architecture based on these data to improve the performance of drug sensitivity prediction. We used a five-layer deep neural network, named as DeepPredictor, that integrated both genomic features of cell lines and chemical information of compounds to predict the half maximal inhibitory concentration on the Cancer Cell Line Encyclopedia (CCLE) dataset. We demonstrated the performance of our deep model using 10-fold cross-validations and leave-one-out strategies and showed that our model outperformed existing approaches.
KeywordsCancer cell lines Drug sensitivity DeepPredictor Deep learning Predictive models
We would like to thank Isidro Cortés Ciriano at Department of Biomedical Informatics, Harvard Medical School for providing the research data and discussing with us during research. This work is supported by the National Science Fund for Excellent Young Scholars under Grant No. 61622213, the National Natural Science Foundation of China under grant No. 61772552, and the Fundamental Research Funds for the Central Universities of Central South University under grant No. 2018zzts560 and No. 2018zzts028.
- 7.Cortes-Ciriano, I., van Westen, G.J., Bouvier, G., et al.: Improved large-scale prediction of growth inhibition patterns using the NCI60 cancer cell line panel. Bioinformatics 32(1), 85–95 (2016)Google Scholar
- 8.Krizhevsky, A., Sutskever I., Hinton G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)Google Scholar
- 10.Sutskever, I., Vinyals, O., Le, Q.V.: Sequence to sequence learning with neural networks. In: Advances in Neural Information Processing Systems, pp. 3104–3112 (2014)Google Scholar