Selected article for: "latent representation and representation space"

Author: Kaifu Gao; Duc Duy Nguyen; Rui Wang; Guo-Wei Wei
Title: Machine intelligence design of 2019-nCoV drugs
  • Document date: 2020_2_4
  • ID: 1qniriu0_9
    Snippet: In the present approach, novel molecules are initially designed at the molecule generator using a three-step procedure. In the first step, the latent space representation of a seed molecule is evaluated for their drug-like properties, such as binding affinities, solubility (logS), partition coefficient, similarity, etc., via pre-trained DNNs. The DNN consists of two hidden layers with 1024 neurons in each layer. In our second step, evaluation res.....
    Document: In the present approach, novel molecules are initially designed at the molecule generator using a three-step procedure. In the first step, the latent space representation of a seed molecule is evaluated for their drug-like properties, such as binding affinities, solubility (logS), partition coefficient, similarity, etc., via pre-trained DNNs. The DNN consists of two hidden layers with 1024 neurons in each layer. In our second step, evaluation results ({Å· i (X)} ) are compared with a set of target values ({y i0 }) via a loss function,

    Search related documents:
    Co phrase search for related documents
    • bind affinity and present approach: 1
    • etc similarity and molecule generator: 1
    • etc similarity and partition coefficient: 1
    • etc similarity partition coefficient and molecule generator: 1
    • etc similarity partition coefficient and partition coefficient: 1
    • evaluation result and pre trained dnn: 1
    • evaluation result and present approach: 1