Adversarial Threshold Neural Computer for Molecular de Novo Design
| Autor: | Alex Zhavoronkov, Alexander Aliper, Yan A. Ivanenkov, Arip Asadulaev, Anastasia V. Aladinskaya, Quentin Vanhaelen, Evgeny Putin |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
0301 basic medicine
Discriminator Similarity (geometry) Artificial neural network Computer science business.industry Pharmaceutical Science Machine Learning 03 medical and health sciences 030104 developmental biology Filter (video) Drug Discovery Molecular Medicine Reinforcement learning Neural Networks Computer Artificial intelligence business Cluster analysis Block (data storage) Generator (mathematics) |
| Zdroj: | Molecular Pharmaceutics. 15:4386-4397 |
| ISSN: | 1543-8392 1543-8384 |
| DOI: | 10.1021/acs.molpharmaceut.7b01137 |
| Popis: | In this article, we propose the deep neural network Adversarial Threshold Neural Computer (ATNC). The ATNC model is intended for the de novo design of novel small-molecule organic structures. The model is based on generative adversarial network architecture and reinforcement learning. ATNC uses a Differentiable Neural Computer as a generator and has a new specific block, called adversarial threshold (AT). AT acts as a filter between the agent (generator) and the environment (discriminator + objective reward functions). Furthermore, to generate more diverse molecules we introduce a new objective reward function named Internal Diversity Clustering (IDC). In this work, ATNC is tested and compared with the ORGANIC model. Both models were trained on the SMILES string representation of the molecules, using four objective functions (internal similarity, Muegge druglikeness filter, presence or absence of sp3-rich fragments, and IDC). The SMILES representations of 15K druglike molecules from the ChemDiv collection... |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|