ProfKin

ProfKin performs structure-based kinase selectivity profiling to predict kinase–ligand binding modes, prioritize kinase targets, and support kinase inhibitor discovery, polypharmacology analysis, and drug repositioning.

Key Features:

• KinLigDB structural repository: A manually curated database of 4,219 kinase–ligand complex structures covering 297 human kinases.
• Comprehensive complex annotations: Detailed attributes per complex include binding site types, ligand categories, interaction fingerprints, downstream molecules, and related human diseases.
• Predictive modeling: Prediction of potential binding modes of query molecules to kinases using advanced computational techniques and prioritization via interaction fingerprint analysis.
• Customizable profiling: Selection of specific KinLigDB subsets, such as kinase subfamilies or binding site types, for focused analyses.
• Kinase ranking index: Ranking of kinases based on a comprehensive index derived from predicted interactions.
• Structural superposition: Superposition of predicted binding poses onto reference kinase structures for structural comparison.
• Supplementary classification data: Provision of classification attributes and phylogenetic tree data associated with top-ranked kinases.

Scientific Applications:

• Selective inhibitor identification: Structural and predictive profiling to aid identification and prioritization of selective kinase inhibitors.
• Polypharmacology and drug repositioning: Cross-kinase profiling to support polypharmacological drug development and drug repositioning efforts.
• Early-stage drug discovery support: Use of structural data and predicted binding modes to inform early-stage target selection and off-target assessment.

Methodology:

Prediction of binding modes using advanced computational techniques, prioritization via interaction fingerprint analysis, ranking of kinases by a comprehensive index derived from predicted interactions, and superposition of predicted binding poses onto reference structures; all analyses use data from KinLigDB (4,219 curated kinase–ligand complexes covering 297 human kinases) with associated binding site, ligand, interaction fingerprint, downstream molecule, disease, classification, and phylogenetic annotations.