Computationally repurposing drugs for breast cancer subtypes using a network-based approach.

Area of Science:

• Biomedical Informatics
• Computational Biology
• Drug Discovery

Background:

• 'De novo' drug discovery is expensive, slow, and high-risk.
• Drug repurposing offers a faster, cheaper, and more efficient alternative.
• Emerging large-scale biological data enables computational drug repurposing (CDR).

Purpose of the Study:

• To address challenges in traditional CDR, particularly capturing complex, non-linear drug-disease-gene associations.
• To develop a network-based integration approach for CDR.
• To identify potential therapeutic drugs for various breast cancer subtypes using CDR.

Main Methods:

• Integrated heterogeneous biomolecular, chemical, bioactivity, genomic, and phenotypic data.
• Applied a network-based machine learning approach to capture complex relationships.
• Utilized the network approach to identify drugs for specific breast cancer subtypes.

Main Results:

• Successfully captured complex, non-linear associations among drugs, genes, and diseases.
• Identified single and pairs of approved or experimental drugs with potential therapeutic effects.
• The identified drugs show promise for treating different breast cancer subtypes.

Conclusions:

• The proposed network-based integration approach enhances CDR by effectively modeling complex biological relationships.
• This method facilitates the discovery of novel therapeutic strategies for breast cancer.
• Further clinical validation is required to confirm the efficacy of the identified drugs.