Knowledge-based approaches to drug discovery for rare diseases

Area of Science:

• Biomedical Informatics
• Drug Discovery
• Computational Biology

Background:

• The traditional drug discovery process is inefficient for rare diseases.
• There is a need for novel approaches to identify therapeutics for rare conditions.

Purpose of the Study:

• To discuss advances in biomedical knowledge mining for rare disease drug discovery.
• To evaluate the role of machine learning and knowledge graphs in identifying rare disease therapeutics.

Main Methods:

• Review of current chemogenomics data relevant to rare diseases.
• Application of machine learning (ML) and biomedical knowledge graph mining.
• Case study using chordoma as an example.

Main Results:

• Biomedical knowledge mining, ML, and knowledge graphs show promise for rare disease drug discovery.
• These methods can expedite the identification of viable drug candidates.
• A chordoma case study demonstrates the effectiveness of these approaches.

Conclusions:

• Knowledge graph mining and artificial intelligence (AI) are powerful tools for drug discovery.
• Wider adoption of AI and knowledge graphs can accelerate the development of treatments for rare and common diseases.