Reinforced Adversarial Neural Computer for de Novo Molecular Design

Abstract

In silico modeling is a crucial milestone in modern drug design and development. Although computer-aided approaches in this field are well-studied, the application of deep learning methods in this research area is at the beginning. In this work, we present an original deep neural network (DNN) architecture named RANC (Reinforced Adversarial Neural Computer) for the de novo design of novel small-molecule organic structures based on the generative adversarial network (GAN) paradigm and reinforcement learning (RL). As a generator RANC uses a differentiable neural computer (DNC), a category of neural networks, with increased generation capabilities due to the addition of an explicit memory bank, which can mitigate common problems found in adversarial settings. The comparative results have shown that RANC trained on the SMILES string representation of the molecules outperforms its first DNN-based counterpart ORGANIC by several metrics relevant to drug discovery: the number of unique structures, passing medicinal chemistry filters (MCFs), Muegge criteria, and high QED scores. RANC is able to generate structures that match the distributions of the key chemical features/descriptors (e.g., MW, logP, TPSA) and lengths of the SMILES strings in the training data set. Therefore, RANC can be reasonably regarded as a promising starting point to develop novel molecules with activity against different biological targets or pathways. In addition, this approach allows scientists to save time and covers a broad chemical space populated with novel and diverse compounds.

Deep learning for molecular design—a review of the state of the art

“…103 To help solve these issues, Putin et al implemented a differentiable neural computer (DNC) as the generator. 103 The DNC (Graves et al, 2016) 143 is an extension of the neural Turing machine (Graves et al 2014) 144 that contains a differentiable memory cell. A memory cell allows the generator to memorize key SMILES motifs, which results in a much "stronger" generator.…”

Automated De Novo Drug Design: Are We Nearly There Yet?

“…[80] Va rious instances of chemistry-savvy generative deep networks have already been developed for de novo drug design, with ac urrent emphasis on adversarial and reinforcement learning methods. [14,81] Theoretical studies are mushrooming,b ut only ahandful of prospective applications has been performed to date.C omputer scientists are well advised to develop algorithms that can detect meaningful patterns in small data sets,w hich are characteristic of earlystage drug discovery,a nd chemists should use these tools in prospective studies.H owever,c hemists and computer scientists often seem to be disconnected. While some of the theoreticians may consider the problem of de novo design solved in principle,w eo bserve am ix of enthusiasm, healthy scepticism, and even plain denial among medicinal chemists when it comes to de novo design.…”

Human microbiome aging clocks based on deep learning and tandem of permutation feature importance and accumulated local effects

“…The renaissance of deep learning that started in 2015 resulted in unprecedented machine learning performance in image, voice, and text recognition, as well as a range of biomedical applications 29 such as drug repurposing 30 and target identification 31 . One of the most impactful applications of DL in biomedicine was in the applications of generative models to de novo molecular design [32][33][34][35][36] . In the context of aging research, these new methods can be combined for geroprotector discovery [37][38][39][40][41] .…”