Journal of Chemical Information and Modeling · 2020 · 83 citations · 39 references
PharmacyMedicinal ChemistryDrug DesignDrug DiscoveryMachine LearningChemical Library DesignChemical Space ExplorationComputational Drug DiscoveryDe Novo Drug DesignMolecular ModelingMolecular SimilarityGated Recurrent Unit NetworksLead Compound OptimizationMolecular GenerationRecurrent Neural Network DesignRecurrent Neural NetworksSimilarity MetricsTrainingTransfer Learning For Molecular Property Prediction
molecular design. While a variety of different methods are available, it is still a challenge to assess and compare their performance. A particularly promising approach for automated drug design is to use recurrent neural networks (RNNs) as SMILES generators and train them with the learning procedure called "transfer learning". This involves first training the initial model on a large generic data set of molecules to learn the general syntax of SMILES, followed by fine-tuning on a smaller set of molecules, coming from, e.g., a lead optimization program. To create a well-performing transfer learning application which can be automated, it is important to understand how the size of the second data set affects the training process. In addition, extensive postfiltering using similarity metrics of the molecules generated after transfer learning should be avoided, as it can introduce new biases toward the selection of drug candidates. Here, we present results from the application of a gated recurrent unit cell (GRU)-RNN to transfer learning on data sets of varying sizes and complexity. Analysis of the results has allowed us to provide some general guidelines for transfer learning. In particular, we show that data set sizes containing at least 190 molecules are needed for effective GRU-RNN-based molecular generation using transfer learning. The methods presented here should be applicable generally to the benchmarking of other deep learning methodologies for molecule generation.
39
On Information and Sufficiency
S. Kullback, R. A. Leibler · The Annals of Mathematical Statistics · 1951 · 20K citations · Full text
Extended-Connectivity Fingerprints
David Rogers, Mathew Hahn · Journal of Chemical Information and Modeling · 2010 · 7.6K citations
PubChem Substance and Compound databases
Sunghwan Kim, Paul Thiessen, Evan Bolton et al. · Nucleic Acids Research · 2015 · 5.5K citations · Full text
Pharmacy, Biochemistry, Chemistry +16