KUALA

KUALA prioritizes kinase inhibitor repurposing candidates using machine learning models trained on a curated set of molecular descriptors.

Key Features:

• Machine learning models: Employs machine learning to predict active kinase–ligand interactions.
• Curated molecular descriptors: Uses a curated set of molecular descriptors to characterize ligands.
• Multi-target priority score: Computes a multi-target priority score to rank ligands across multiple kinases.
• Repurposing threshold: Establishes a repurposing threshold to distinguish candidates suitable for repositioning.
• Binding-site similarity and poly-pharmacology: Exploits kinase binding-site similarity to assess poly-pharmacology and repurposing potential.
• Kinase scope: Focuses on the protein kinase family (encoded by over 500 genes).
• Output dataset: Produces a comprehensive dataset of evaluated kinase–ligand pairs with associated scores.

Scientific Applications:

• Kinase inhibitor repositioning: Identifies existing molecules with potential activity against alternative kinases for repurposing.
• Prioritization for experimental follow-up: Ranks kinase–ligand pairs to guide experimental validation.
• Poly-pharmacology analysis: Facilitates analysis of multi-target interactions across kinases by leveraging binding-site similarities.
• Support for kinase drug discovery: Provides a scored dataset to support kinase-targeted drug discovery and pharmacology studies.

Methodology:

Applies machine learning to curated molecular descriptors to identify active kinase ligands, computes multi-target priority scores and a repurposing threshold, and outputs a scored dataset of kinase–ligand pairs.