Related Experiment Video
Updated: Oct 4, 2025
09:34
A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data
Published on: September 25, 2021
4.1K
Graph Neural Networks as a Potential Tool in Improving Virtual Screening Programs
Luiz Anastacio Alves1, Natiele Carla da Silva Ferreira1, Victor Maricato1
1Laboratory of Cellular Communication, Oswaldo Cruz Institute - Fiocruz, Rio de Janeiro, Brazil.
Frontiers in Chemistry
|February 7, 2022
Summary
Graph neural networks (GNNs) offer a promising approach to enhance virtual screening (VS) for drug discovery. This deep learning method could significantly improve the identification of potential drug candidates from natural products, overcoming limitations of current VS techniques.
Area of Science:
- Computational chemistry and cheminformatics
- Artificial intelligence in drug discovery
- Pharmacology and medicinal chemistry
Background:
- Traditional drug discovery yields a low success rate, with less than 1% of researched drugs reaching the market.
- Virtual screening (VS) offers cost and time efficiencies but has not significantly improved the number of approved drugs.
- Current VS methods rely on physics and quantum mechanics, with limited success in identifying viable drug candidates.
Purpose of the Study:
- To explore the potential of graph neural networks (GNNs) as a tool to enhance virtual screening.
- To evaluate the application of GNNs for natural product drug discovery.
- To discuss the advantages, disadvantages, and future prospects of GNNs in drug discovery.
Main Methods:
- Review of existing literature on virtual screening methodologies.
- Analysis of graph neural network (GNN) applications in computational chemistry.
- Discussion of GNN integration with traditional VS algorithms and standalone use.
Main Results:
- Graph neural networks (GNNs) show potential to improve the accuracy and efficiency of virtual screening.
- GNNs can be applied to natural product libraries, expanding the scope of drug discovery.
- Current limitations include challenges with spatial coordinates and the need for adequate datasets.
Conclusions:
- GNNs represent a powerful deep learning approach that could revolutionize drug discovery.
- Overcoming current obstacles is crucial for realizing the full potential of GNNs in identifying novel therapeutics.
- The integration of GNNs may significantly increase the success rate of bringing new drugs to market.
