Molecular Generation for Desired Transcriptome Changes With Adversarial Autoencoders
Rim Shayakhmetov, Maksim Kuznetsov, Alexander Zhebrak, Artur Kadurin, Sergey Nikolenko, Alexander Aliper, Daniil Polykovskiy
Abstract
Gene expression profiles are useful for assessing drugs' efficacy and side-effects. In this paper, we propose a new generative model that infers drug molecules that could induce a desired change in gene expression. Our model—Bidirectional Adversarial Autoencoder—explicitly separates cellular processes captured in gene expression changes into two feature sets: related and unrelated to the drug incubation. The model uses related features to produce a drug hypothesis. We validate our model on the LINCS L1000 dataset by generating molecular structures in the SMILES format for the desired transcriptional response. In the experiments, we show that the proposed model can generate novel molecular structures that could induce a given gene expression change, or predict gene expression difference after incubation of a given molecular structure.