Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks

Abstract

In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active toward a given biological target, we propose to fine-tune the model with small sets of molecules, which are known to be active against that target. Against Staphylococcus aureus, the model reproduced 14% of 6051 hold-out test molecules that medicinal chemists designed, whereas against Plasmodium falciparum (Malaria), it reproduced 28% of 1240 test molecules. When coupled with a scoring function, our model can perform the complete de novo drug design cycle to generate large sets of novel molecules for drug discovery.

Cheminformatics in Drug Discovery, an Industrial Perspective

“…Convolutional networks that are used for image analysis . Recurrent neural networks are used for de novo molecular design, Auto‐encoder networks that can also be used for de novo molecular design ,…”

“…It has been shown in the last years that neural network architectures trained on SMILES strings can learn the SMILES grammar and generate novel molecules. Two different types of architectures have mainly been utilized autoencoders, and recurrent neural networks (RNNs) ,. Example architectures are given are given in Figure .…”

iQSPR in XenonPy: A Bayesian Molecular Design Algorithm

“…The sampling efficiency of iQSPR-X is highly influenced by the reliability of the evaluator that predicts the material properties for any given chemical structure. [18,[34][35][36][37][38][39][40][41] In this study, we applied a specific type of transfer learning using pre-trained neural networks. XenonPy currently provides 140,000 pretrained neural networks for the prediction of physical, chemical, electronic, thermodynamic, and mechanical properties of small organic molecules, polymers, and inorganic crystalline materials, with models for 15, 18, and 12 properties of these material types, respectively.…”

“…In particular, to broaden the search space, ML methods that use probabilistic language models based on deep neural networks (DNNs) have proliferated intensively since 2017. Promising examples have included various types of varia-tional autoencoders, [12][13][14][15] generative adversarial networks, [16] recurrent neural networks, [17,18] and so on. [11] Models trained to recognize chemically realistic structures are then used to refine chemical strings in the molecular design calculation.…”

Hunting for Organic Molecules with Artificial Intelligence: Molecules Optimized for Desired Excitation Energies

“…It has been brought closer to reality by recent advances on machine learning algorithms for de novo molecule design, that do not need handcrafted chemical rules [1][2][3][4][5] . Figure 1 illustrates our AI-assisted chemistry platform to develop new molecules.…”