Deep generative neural network for accurate drug response imputation

Abstract

Drug response differs substantially in cancer patients due to inter- and intra-tumor heterogeneity. Particularly, transcriptome context, especially tumor microenvironment, has been shown playing a significant role in shaping the actual treatment outcome. In this study, we develop a deep variational autoencoder (VAE) model to compress thousands of genes into latent vectors in a low-dimensional space. We then demonstrate that these encoded vectors could accurately impute drug response, outperform standard signature-gene based approaches, and appropriately control the overfitting problem. We apply rigorous quality assessment and validation, including assessing the impact of cell line lineage, cross-validation, cross-panel evaluation, and application in independent clinical data sets, to warrant the accuracy of the imputed drug response in both cell lines and cancer samples. Specifically, the expression-regulated component (EReX) of the observed drug response achieves high correlation across panels. Using the well-trained models, we impute drug response of The Cancer Genome Atlas data and investigate the features and signatures associated with the imputed drug response, including cell line origins, somatic mutations and tumor mutation burdens, tumor microenvironment, and confounding factors. In summary, our deep learning method and the results are useful for the study of signatures and markers of drug response.

Robust Prediction of Patient-Specific Clinical Response to Unseen Drugs From in vitro Screens Using Context-aware Deconfounding Autoencoder

“…As a comparison, the low-dimensional representations of CCLE and TCGA data are clearly separated when using original gene expression profiles or vanilla autoencoder. Thus, CODE-AE-ADV is more effective in addressing OOD problem than the embedding algorithms that are used by the state-of-the-art method VAEN [6].…”

“…We compared CODE-AE with the following base-line models that include unlabeled pre-training: VAEN [6], standard autoencoder (AE) [47], denoising autoencoder (DAE) [9], and variational autoencoder (VAE) [8] as well as representative domain adaptation methods including deep coral (CORAL) [10] and domain separation network (DSN) [11] of both MMD (DSN-MMD) and adversarial (DSN-DANN) training variants. Furthermore, we included a more recent adversarial deconfounding autoencoder (ADAE) [4] given its similar formation as DANN [48] and state-of-the-art performance in transcriptomics data sets.…”

“…It can be only applied to the scenario of few-shot learning but not zero-shot learning. Most relevant to this work, Jia et al has applied Variational Autoencoder (VAE) pre-training followed by Elastic Net supervised training (VAEN) to learn cell line models and applied them to impute in vivo drug response [6]. However, VAEN is not optimized to reliably transfer cell line data to patient samples and disentangle confounding factors [6] due to the fundamental limitation of VAE.…”

“…Lack of effective personalized chemotherapy tailored to individual patients often leads to unnecessary suffering and reduces the chances of patient’s overall survival. Our extensive studies show that CODE-AE effectively alleviates the out-of-distribution problem when transferring the cell line model to patient samples, significantly outperforms state-of-the-art methods AD-AE [4], TCRP [5], and VAEN [6] that are specifically designed for transcriptomics data as well as other popular domain adaptation methods Variational autoencoder (VAE) [8], Denoising autoencoder (DAE) [9], Deep Coral [10], and Domain Separation Network (DSN) [11] in terms of both accuracy and robustness. Using CODE-AE, we screened 50 drugs for 9,808 cancer patients.…”

Predictive Modeling of Multiplex Chemical Phenomics for Novel Cells and Patients: Applied to Personalized Alzheimer’s Disease Drug Repurposing

“…There are hundreds of characterized cell lines and patient samples with drug sensitivity data collected in different studies, such as Broad Institute Cancer Cell Line Encyclopedia (CCLE) 8 , Genomics of Drug Sensitivity in Cancer (GDSC) 9 , The Cancer Genome Atlas Program (TCGA) 10 . A great deal of effort has been devoted to developing machine learning models for predicting drug sensitivity [11][12][13][14][15] . However, the majority of previous research focused on predicting summary metrics of the drug-response curve, such as IC50 or area under the drug-response curve 16,17 .…”