Prediction of chemical compounds properties using a deep learning model

Mykola Galushka, Chris Swain, Fiona Browne, Maurice Mulvenna, RR Bond, Darren Gray

Research output: Contribution to journalArticlepeer-review

18 Citations (Scopus)
313 Downloads (Pure)


The discovery of new medications in a cost-effective manner has become the top priority for many pharmaceutical companies. Despite decades of innovation, many of their processes arguably remain relatively inefficient. One such process is the prediction of biological activity. This paper describes a new deep learning model, capable of conducting a preliminary screening of chemical compounds in-silico. The model has been constructed using a variation autoencoder to generate chemical compound fingerprints, which have been used to create a regression model to predict their LogD property and a classification model to predict binding in selected assays from the ChEMBL dataset. The conducted experiments demonstrate accurate prediction of the properties of chemical compounds only using structural definitions and also provide several opportunities to improve upon this model in the future.

Original languageEnglish
Pages (from-to)13345–13366
Number of pages22
JournalNeural Computing and Applications
Issue number20
Early online date4 Jun 2021
Publication statusPublished (in print/issue) - 31 Oct 2021

Bibliographical note

Funding Information:
We acknowledge the contribution of Chris Swain the Founded Cambridge MedChem Consulting.

Publisher Copyright:
© 2021, The Author(s).


  • Deep neural networks
  • Machine learning
  • chemical compounds Properties


Dive into the research topics of 'Prediction of chemical compounds properties using a deep learning model'. Together they form a unique fingerprint.

Cite this