Deargen Molecule Equalizer

Deargen Molecule Equalizer

Early research into the metabolic properties of candidates in the process of drug development is becoming a key strategy to reduce the cost and time of drug development. MolEQ designs novel drug candidates by optimizing ADMET (absorption, distribution, metabolism, excretion, toxicity) of compounds to the desired disease target. MolEQ technology has a feature that multiple properties of a substance can be optimized by giving a weight differently according to the molecular design goals, based on affinity numerical information from the DearDTI. The optimized features are shown as efficacy, toxicity, Synthetic Accessibility (SA), and Quantitative Estimate of Drug-likeness (QED), etc. Other models typically proceed with optimization by learning only the properties of one substance such as QED. As a result, MolEQ has optimized several properties of the compound at once, so it has an advantage that the results can be directly applied to a drug development research.

*Please note that member solutions are often customizable to meet the needs of individual enterprise end users.



  • A novel small-molecule lead optimization model
    MolEQ is constructed on a module that predicts the drug-protein binding affinity based on molecule transformer (MT), convolutional neural network (CNN), and recurrent neural network (RNN), a toxicity prediction module, and an ADME prediction module
  • Optimization model with multiple property modulation
    Uses a chemical structural manipulation model known as molecule deep Q-network (MolDQN). It takes in to account Quantitative estimate of drug-likeness (QED) and binding affinity, toxicity and ADME (Absoprtion, Distribution, Metabolism, and Excretion)


KoreaIntel® Xeon ScalableCSP - Amazon Web ServicesCSP - Microsoft AzureTensorFlowHealthcareModels can be trained - data input only requiredModels can be trained - requires labeled dataAI Software/SaaSLinux OtherOtherDrug DiscoveryDeep Learning