With the advent of deep generative models in computational chemistry, in-silico drug design is undergoing an unprecedented transformation. Although deep learning approaches have shown potential in generating compounds with desired chemical properties, they disregard the cellular environment of target diseases. Bridging systems biology and drug design, we present a reinforcement learning method for de novo molecular design from gene expression profiles. We construct a hybrid Variational Autoencoder that tailors molecules to target-specific transcriptomic profiles, using an anticancer drug sensitivity prediction model (PaccMann) as reward function. Without incorporating information about anticancer drugs, the molecule generation is biased toward compounds with high predicted efficacy against cell lines or cancer types. The generation can be further refined by subsidiary constraints such as toxicity. Our cancer-type-specific candidate drugs are similar to cancer drugs in drug-likeness, synthesizability, and solubility and frequently exhibit the highest structural similarity to compounds with known efficacy against these cancer types.

Bridging systems biology and drug design, we propose a deep learning framework for de novo discovery of molecules tailored to bind with given protein targets. Our methodology is exemplified by the task of designing antiviral candidates to target SARS-CoV-2 related proteins. Crucially, our framework does not require fine-tuning for specific proteins but is demonstrated to generalize in proposing ligands with high predicted binding affinities against unseen targets. Coupling our framework with the automatic retrosynthesis prediction of IBM RXN for Chemistry, we demonstrate the feasibility of swift chemical synthesis of molecules with potential antiviral properties that were designed against a specific protein target. In particular, we synthesize an antiviral candidate designed against the host protein angiotensin converting enzyme 2 (ACE2); a surface receptor on human respiratory epithelial cells that facilitates SARS-CoV-2 cell entry through its spike glycoprotein. This is achieved as follows. First, we train a multimodal ligand–protein binding affinity model on predicting affinities of bioactive compounds to target proteins and couple this model with pharmacological toxicity predictors. Exploiting this multi-objective as a reward function of a conditional molecular generator that consists of two variational autoencoders (VAE), our framework steers the generation toward regions of the chemical space with high-reward molecules. Specifically, we explore a challenging setting of generating ligands against unseen protein targets by performing a leave-one-out-cross-validation on 41 SARS-CoV-2-related target proteins. Using deep reinforcement learning, it is demonstrated that in 35 out of 41 cases, the generation is biased towards sampling binding ligands, with an average increase of 83% comparing to an unbiased VAE. The generated molecules exhibit favorable properties in terms of target binding affinity, selectivity and drug-likeness. We use molecular retrosynthetic models to provide a synthetic accessibility assessment of the best generated hit molecules. Finally, with this end-to-end framework, we synthesize 3-Bromobenzylamine, a potential inhibitor of the host ACE2 protein, solely based on the recommendations of a molecular retrosynthesis model and a synthesis protocol prediction model. We hope that our framework can contribute towards swift discovery of de novo molecules with desired pharmacological properties.

The identification of new targeted and personalized therapies for cancer requires the fast and accurate assessment of the drug efficacy of potential compounds against a particular biomolecular sample. It has been suggested that the integration of complementary sources of information might strengthen the accuracy of a drug efficacy prediction model. Here, we present a web-based platform for the Prediction of AntiCancer Compound sensitivity with Multimodal Attention-based Neural Networks (PaccMann). PaccMann is trained on public transcriptomic cell line profiles, compound structure information and drug sensitivity screenings, and outperforms state-of-the-art methods on anticancer drug sensitivity prediction. On the open-access web service (https://ibm.biz/paccmann-aas), users can select a known drug compound or design their own compound structure in an interactive editor, perform in-silico drug testing and investigate compound efficacy on publicly available or user-provided transcriptomic profiles. PaccMann leverages methods for model interpretability and outputs confidence scores as well as attention heatmaps that highlight the genes and chemical sub-structures that were more important to make a prediction, hence facilitating the understanding of the model’s decision making and the involved biochemical processes. We hope to serve the community with a toolbox for fast and efficient validation in drug repositioning or lead compound identification regimes.

No abstract

Reliable identification of molecular biomarkers is essential for accurate patient stratification. While state-of-the-art machine learning approaches for sample classification continue to push boundaries in terms of performance, most of these methods are not able to integrate different data types and lack generalization power, limiting their application in a clinical setting. Furthermore, many methods behave as black boxes, and we have very little understanding about the mechanisms that lead to the prediction. While opaqueness concerning machine behavior might not be a problem in deterministic domains, in health care, providing explanations about the molecular factors and phenotypes that are driving the classification is crucial to build trust in the performance of the predictive system. We propose Pathway-Induced Multiple Kernel Learning (PIMKL), a methodology to reliably classify samples that can also help gain insights into the molecular mechanisms that underlie the classification. PIMKL exploits prior knowledge in the form of a molecular interaction network and annotated gene sets, by optimizing a mixture of pathway-induced kernels using a Multiple Kernel Learning (MKL) algorithm, an approach that has demonstrated excellent performance in different machine learning applications. After optimizing the combination of kernels to predict a specific phenotype, the model provides a stable molecular signature that can be interpreted in the light of the ingested prior knowledge and that can be used in transfer learning tasks.