cando.py: Open Source Software for Predictive Bioanalytics of Large Scale Drug-Protein-Disease Data

Area of Science:

• Computational biology
• Pharmacology
• Bioinformatics

Background:

• Traditional drug discovery targets single biological pathways.
• The multitarget theory suggests drugs interact with multiple targets for therapeutic effects.
• Large-scale drug-protein interaction analysis offers insights into disease mechanisms and potential drug candidates.

Purpose of the Study:

• To present a Python package for analyzing drug-proteome and drug-disease relationships.
• To facilitate drug similarity assessment and predict novel therapeutics.
• To support drug discovery and repurposing through computational analysis.

Main Methods:

• Implementation of the Computational Analysis of Novel Drug Opportunities (CANDO) platform in Python.
• Rapid scoring of billions of drug-protein interactions to calculate drug-proteome signature similarity.
• Development of benchmarking protocols for drug discovery and repurposing, including consensus scoring for drug predictions.

Main Results:

• The CANDO package enables rapid drug similarity assessment and analysis of drug-proteome interactions.
• It provides tools for benchmarking drug discovery and repurposing efforts.
• Machine learning modules are integrated for benchmarking and predicting potential drug candidates.

Conclusions:

• The CANDO Python package offers a robust platform for analyzing complex drug-target interactions.
• It facilitates efficient drug repurposing and the identification of novel therapeutic candidates.
• The package is accessible via GitHub, Conda, and a dedicated website for broader adoption in drug discovery research.