IntAct is an open-source, open-data molecular interaction database and toolkit that provides curated binary interaction evidences from the literature or direct data submissions. As of September 2009, IntAct contains over 200,000 curated binary interaction evidences from 4,035 scientific publications. The database now offers a two-tiered view of interaction data, allowing users to iteratively develop complex queries using detailed annotations and hierarchical controlled vocabularies. Results are presented in a simplified tabular format, with specialized views enabling deeper exploration of interaction details. IntAct follows a deep curation model, capturing detailed experimental information from publications. It uses controlled vocabularies such as the PSI-MI ontology to describe technical details, binding sites, and protein tags. Interacting molecules are systematically mapped to stable identifiers from public databases. Binding sites are cross-referenced with the InterPro database. IntAct provides a comprehensive curation approach, with detailed annotations, synonyms, and Gene Ontology annotations for each complex. The IntAct data model stores n-ary interactions, which may involve one or more interactors. To facilitate analysis, binary interactions are generated from complexes using the spoke model. IntAct also provides targeted datasets on specific curation topics, such as nucleoside diphosphate kinases, cyanobacteria, and Alzheimer's disease. These datasets can be searched or downloaded from the website. IntAct continues to provide curated complexes as a dictionary of complex nomenclature, with each complex having a distinct name and entry. The function, synonyms, and physical properties of each complex are described, along with links to external databases. IntAct allows direct data deposition from authors as part of the publication process. Data is kept confidential until acceptance or publication. A new website has been developed to meet increased user demands, offering improved functionalities for searching, browsing, and visualizing molecular interaction data. The website includes a simple search facility, a Molecular Interaction Query Language (MIQL), and a dedicated chemical structure search module. The IntAct website also provides browsing capabilities by selected topics, such as proteins, chemicals, and nucleic acids. It includes a 'Lists' tab to display participants in interactions and a 'Interaction Details' tab to show detailed information about specific interactions. The 'Molecule View' tab provides detailed information about proteins, including links to external resources. IntAct supports network visualization through a simple static graph and interoperability with external tools like Cytoscape. The database is implemented using Java and open-source libraries, with all software being free and open-source. IntAct is part of the IMEx consortium, which aims to reduce redundant curation and increase the coverage of published interaction data. IntAct continues to develop new features, including the MITAB 2.6 format and a new curation tool. It also supports dynamic networks, allowing users to explore dynamically changing interaction networks. IntAct aims to increase the proportion of interaction data directly depositedIntAct is an open-source, open-data molecular interaction database and toolkit that provides curated binary interaction evidences from the literature or direct data submissions. As of September 2009, IntAct contains over 200,000 curated binary interaction evidences from 4,035 scientific publications. The database now offers a two-tiered view of interaction data, allowing users to iteratively develop complex queries using detailed annotations and hierarchical controlled vocabularies. Results are presented in a simplified tabular format, with specialized views enabling deeper exploration of interaction details. IntAct follows a deep curation model, capturing detailed experimental information from publications. It uses controlled vocabularies such as the PSI-MI ontology to describe technical details, binding sites, and protein tags. Interacting molecules are systematically mapped to stable identifiers from public databases. Binding sites are cross-referenced with the InterPro database. IntAct provides a comprehensive curation approach, with detailed annotations, synonyms, and Gene Ontology annotations for each complex. The IntAct data model stores n-ary interactions, which may involve one or more interactors. To facilitate analysis, binary interactions are generated from complexes using the spoke model. IntAct also provides targeted datasets on specific curation topics, such as nucleoside diphosphate kinases, cyanobacteria, and Alzheimer's disease. These datasets can be searched or downloaded from the website. IntAct continues to provide curated complexes as a dictionary of complex nomenclature, with each complex having a distinct name and entry. The function, synonyms, and physical properties of each complex are described, along with links to external databases. IntAct allows direct data deposition from authors as part of the publication process. Data is kept confidential until acceptance or publication. A new website has been developed to meet increased user demands, offering improved functionalities for searching, browsing, and visualizing molecular interaction data. The website includes a simple search facility, a Molecular Interaction Query Language (MIQL), and a dedicated chemical structure search module. The IntAct website also provides browsing capabilities by selected topics, such as proteins, chemicals, and nucleic acids. It includes a 'Lists' tab to display participants in interactions and a 'Interaction Details' tab to show detailed information about specific interactions. The 'Molecule View' tab provides detailed information about proteins, including links to external resources. IntAct supports network visualization through a simple static graph and interoperability with external tools like Cytoscape. The database is implemented using Java and open-source libraries, with all software being free and open-source. IntAct is part of the IMEx consortium, which aims to reduce redundant curation and increase the coverage of published interaction data. IntAct continues to develop new features, including the MITAB 2.6 format and a new curation tool. It also supports dynamic networks, allowing users to explore dynamically changing interaction networks. IntAct aims to increase the proportion of interaction data directly deposited