Advances in de novo drug design: from conventional to machine learning methods

Varnavas D. Mouchlis, Antreas Afantitis, Angela Serra, Michele Fratello, Anastasios G. Papadiamantis, Vassilis Aidinis, Iseult Lynch, Dario Greco, Georgia Melagraki

Research output: Contribution to journalReview articlepeer-review

443 Downloads (Pure)


De novo drug design is a computational approach that generates novel molecular structures from atomic building blocks with no a priori relationships. Conventional methods include structure‐based and ligand‐based design, which depend on the properties of the active site of a biological target or its known active binders, respectively. Artificial intelligence, including machine learning, is an emerging field that has positively impacted the drug discovery process. Deep reinforcement learning is a subdivision of machine learning that combines artificial neural networks with reinforcement‐learning architectures. This method has successfully been employed to develop novel de novo drug design approaches using a variety of artificial networks including recurrent neural networks, convolutional neural networks, generative adversarial networks, and autoencod-ers. This review article summarizes advances in de novo drug design, from conventional growth algorithms to advanced machine‐learning methodologies and highlights hot topics for further de-velopment.

Original languageEnglish
Article number1676
Number of pages22
JournalInternational Journal of Molecular Sciences
Issue number4
Publication statusPublished - 7 Feb 2021

Bibliographical note

Funding Information:
Funding: V.D.M. and A.A. acknowledge support from: CONCEPT/0618/0031, ENTER‐ PRISES/0916/14 and ENTERPRISES/0618/0122 projects, which were co‐funded by the European Re‐ gional Development Fund and the Republic of Cyprus through the Research and Innovation Foun‐ dation. DG received support from the Academy of Finland (grant agreement 322761). This work was supported via the H2020 EU research infrastructure for nanosafety project NanoCommons (Grant Agreement No. 731032), the EU H2020 nanoinformatics project NanoSolveIT (Grant Agree‐ ment No. 814572).

Publisher Copyright:
© 2021 by the authors. Licensee MDPI, Basel, Switzerland.


  • Artificial intelligence
  • Artificial neural networks
  • Autoencoders
  • Convolutional neural networks
  • De novo drug design
  • Deep reinforcement learning
  • Generative adversarial networks
  • Machine learning
  • Recurrent neural networks

ASJC Scopus subject areas

  • Catalysis
  • Molecular Biology
  • Spectroscopy
  • Computer Science Applications
  • Physical and Theoretical Chemistry
  • Organic Chemistry
  • Inorganic Chemistry


Dive into the research topics of 'Advances in de novo drug design: from conventional to machine learning methods'. Together they form a unique fingerprint.

Cite this