Dr.VAE: Drug Response Variational Autoencoder
Rampasek, Ladislav, Hidru, Daniel, Smirnov, Petr, Haibe-Kains, Benjamin, Goldenberg, Anna
We present two deep generative models based on Variational Autoencoders to improve the accuracy of drug response prediction. Our models, Perturbation Variational Autoencoder and its semi-supervised extension, Drug Response Variational Autoencoder (Dr.VAE), learn latent representation of the underlying gene states before and after drug application that depend on: (i) drug-induced biological change of each gene and (ii) overall treatment response outcome. Our VAE-based models outperform the current published benchmarks in the field by anywhere from 3 to 11% AUROC and 2 to 30% AUPR. In addition, we found that better reconstruction accuracy does not necessarily lead to improvement in classification accuracy and that jointly trained models perform better than models that minimize reconstruction error independently.
Jul-6-2017
- Country:
- North America > Canada > Ontario > Toronto (0.15)
- Genre:
- Research Report > New Finding (0.46)
- Industry:
- Technology: