Computational Analysis of Novel Drug Opportunities (CANDO)

Computational Analysis of Novel Drug Opportunities (CANDO) analyzes drug-protein interactions across proteomes to predict drug similarities, repurpose existing therapeutics, and prioritize novel chemical entities for specific indications.

Key Features:

• Multitarget Analysis: Evaluates drugs by their interactions with multiple biological targets to capture multitarget theory and phenotypic outcomes.
• Drug-Proteome Interaction Scoring: Implements an in-house protocol to rapidly score large numbers of drug-protein interactions and compute drug-proteome interaction signatures.
• Benchmarking Protocols: Provides benchmarking protocols for shotgun drug discovery and repurposing to evaluate relationships between drugs and approved indications.
• Drug Prediction via Consensus Scoring: Generates compound predictions by consensus scoring to identify compounds most similar to known therapeutics for specific conditions.
• Support for Novel Chemical Entities: Offers tools to compare and rank novel chemical entities against existing drug-proteome interaction signatures.
• Machine Learning Modules: Integrates machine learning modules that support benchmarking and the prediction of putative drug candidates.

Scientific Applications:

• Large-scale drug-protein interaction analysis: Enables proteome-wide assessment of drug-target interaction patterns to inform therapeutic hypotheses.
• Multitarget mechanism exploration: Supports investigation of multitarget effects relevant to complex or multifaceted disease mechanisms.
• Drug repurposing: Facilitates identification of existing drugs with similarity to approved therapeutics for new indications.
• Novel candidate prioritization: Aids ranking and selection of novel chemical entities as potential drug candidates based on interaction signatures.

Methodology:

CANDO is implemented as a Python package and uses an in-house rapid interaction scoring protocol to compute drug-proteome interaction signatures, applies consensus scoring and benchmarking protocols for prediction and evaluation, and incorporates integrated machine learning modules to support benchmarking and candidate prediction.