Utilizing graph machine learning within drug discovery and development.

Area of Science:

• Biomedical Machine Learning
• Computational Chemistry
• Drug Discovery and Development

Background:

• Graph machine learning (GML) is increasingly utilized in pharmaceutical and biotechnology sectors.
• GML excels at modeling complex biomolecular structures and relationships.
• It facilitates the integration of diverse data types, including multi-omic datasets.

Purpose of the Study:

• To provide a multidisciplinary academic-industrial review of GML in drug discovery and development.
• To summarize current GML applications across the drug development pipeline.
• To identify emerging trends and future potential of GML in this field.

Main Methods:

• Introduction of key terms and GML modeling approaches.
• Chronological review of GML applications through the drug development pipeline.
• Summarization of existing literature on GML for target identification, molecular design, and drug repurposing.

Main Results:

• GML is being applied to target identification and the design of small molecules and biologics.
• Significant progress has been made in drug repurposing using GML.
• Emerging milestones include GML-identified repurposed drugs advancing to in vivo studies.

Conclusions:

• Graph machine learning is a rapidly developing field with significant promise for drug discovery.
• GML's ability to model complex biological data positions it as a key future technology.
• The successful application of GML in drug repurposing suggests its broader impact on pharmaceutical R&D.