drug-repurposing

To identify drug repurposing candidates, this skill matches disease-specific transcriptomic signatures against known drug perturbation profiles. It generates a ranked list of compounds by scoring connectivity, genetic associations, and network proximity using pandas and scipy for statistical enrichment.

Systematic drug repositioning is achieved through four multi-strategy workflows: signature-based, target-based, network-based, and genetics-based analysis. These methods integrate clinical evidence to prioritize approved compounds for new therapeutic indications.

Yes, networkx and gseapy are utilized for protein-protein interaction network analysis and statistical enrichment. This computes graph-based proximity between disease targets and drug nodes to find existing compounds that reverse disease signatures.

You need a disease-specific biological signature and genetic associations to match with drug perturbation profiles. The skill processes this input to score connectivity and clinical evidence, yielding a ranked list of therapeutic candidates.

Yes, the approach integrates computational predictions with clinical and safety data. This evidence integration prioritizes candidate drugs by combining network proximity scores with known safety profiles to find new uses for existing drugs.

Graph-based proximity analysis requires numpy, pandas, scipy, gseapy, and networkx. Matplotlib and seaborn are also included for visualizing the statistical enrichment and network results during the computational drug repurposing workflow.