ACS Central Science· 2017Q1
Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks
- 1,678citations
- Q1SCImago
- 2017year
Short summary
Recurrent neural networks (RNNs) can generate novel drug-like molecules by learning from existing molecular structures, achieving 14-28% reproduction of test molecules against specific biological targets.
AI-generated from the title and abstract; the full text is not read.
Key points
- Recurrent neural networks (RNNs) are trained as generative models for molecular structures.
- Generated molecules' properties correlate well with training data.
- Fine-tuning RNNs with active molecule sets enriches generated libraries for specific targets.
- Model reproduced 14% of test molecules against S. aureus and 28% against P. falciparum.
- RNNs coupled with scoring functions can perform the complete de novo drug design cycle.
AI-generated from the title and abstract; the full text is not read.
Abstract
High Resolution Image Download MS PowerPoint Slide 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.
The authors' abstract, as published at the source. ACS Central Science, 2017 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Computational Theory and Mathematics
Computational Theory and MathematicsComputer Science