MINT

MINT catalogs protein–protein interactions (PPI) curated from peer‑reviewed scientific literature to provide a standardized resource of molecular interaction data.

Key Features:

• Dataset content (Sep 2011): Approximately 235,000 binary interactions derived from over 4,750 publications.
• Protein–protein interactions (PPI): Focused curation and representation of binary protein–protein interactions reported in the literature.
• IMEx membership and PSI‑MI compliance: Participation in the International Molecular Exchange consortium (IMEx) and adherence to the Proteomics Standards Initiative's Molecular Interaction Ontology (PSI‑MI) for annotation and data representation.
• Manual curation and annotation: Manual extraction, curation, and annotation of interaction data from peer‑reviewed publications according to defined curation policies.
• Confidence scoring algorithm: Implementation of an algorithm to assign confidence scores to individual interactions.
• Data integration with IntAct (EMBL‑EBI): Transfer and integration of manually curated MINT records into the IntAct database hosted by EMBL‑EBI, contributing to combined IMEx datasets.

Scientific Applications:

• Molecular interaction analysis: Use of curated PPI data to analyze interaction networks underlying biological processes and disease mechanisms.
• Interoperability and cross‑database analyses: Integration with IMEx and IntAct enabling comparative and aggregated analyses across molecular interaction resources.
• Data reliability assessment: Use of assigned confidence scores to evaluate the reliability of reported interactions in downstream studies.

Methodology:

Manual curation and annotation of interactions from peer‑reviewed literature; representation using PSI‑MI ontology; and application of an algorithm to assign confidence scores to interactions; integration of curated records into IntAct (EMBL‑EBI).