Applying deep learning to iterative screening of medium-sized molecules for protein-protein interaction-targeted drug discovery.
Yugo ShimizuTomoki YonezawaYu BaoJunichi SakamotoMariko YokogawaToshio FuruyaMasanori OsawaKazuyoshi IkedaPublished in: Chemical communications (Cambridge, England) (2023)
We combined a library of medium-sized molecules with iterative screening using multiple machine learning algorithms that were ligand-based, which resulted in a large increase of the hit rate against a protein-protein interaction target. This was demonstrated by inhibition assays using a PPI target, Kelch-like ECH-associated protein 1/nuclear factor erythroid 2-related factor 2 (Keap1/Nrf2), and a deep neural network model based on the first-round assay data showed a highest hit rate of 27.3%. Using the models, we identified novel active and non-flat compounds far from public datasets, expanding the chemical space.
Keyphrases
- protein protein
- machine learning
- nuclear factor
- deep learning
- drug discovery
- neural network
- small molecule
- toll like receptor
- big data
- high throughput
- artificial intelligence
- image quality
- healthcare
- oxidative stress
- electronic health record
- convolutional neural network
- mental health
- cancer therapy
- rna seq
- computed tomography
- drug delivery
- inflammatory response
- single cell
- drug induced