SMART is a web resource for protein domain identification and annotation, providing extensive information on protein domain architectures. The current version includes over 1200 manually curated protein domains, with more than 200 new models since the last update. The underlying protein databases have been synchronized with UniProt, Ensembl, and STRING, resulting in over 100 million annotated domains and protein features. The 'Genomic' mode has been expanded to include 2031 species, compared to 1133 in the previous release. The protein annotation pages have been completely redesigned, featuring a new vector-based display engine for protein schematics, which can be exported as high-resolution bitmap images. The taxonomic tree displays have been significantly improved, with an integrated search engine for easy navigation. The new version also includes post-translational modification data and detailed orthology information from PTMcode and eggNOG databases. Biological pathway and protein interaction data have been expanded and synchronized with the latest versions of Pathways Explorer and STRING databases. The new tabbed annotation interface allows users to access subsets of information as requested. The backend of SMART is a relational database powered by PostgreSQL, storing all annotation data and pre-calculated protein analyses. The database has been restructured to handle the increased number of annotated domains and features, which caused performance issues. SMART remains a popular and widely used resource for protein domain analysis. Funding for open access is provided by the European Molecular Biology Laboratory.SMART is a web resource for protein domain identification and annotation, providing extensive information on protein domain architectures. The current version includes over 1200 manually curated protein domains, with more than 200 new models since the last update. The underlying protein databases have been synchronized with UniProt, Ensembl, and STRING, resulting in over 100 million annotated domains and protein features. The 'Genomic' mode has been expanded to include 2031 species, compared to 1133 in the previous release. The protein annotation pages have been completely redesigned, featuring a new vector-based display engine for protein schematics, which can be exported as high-resolution bitmap images. The taxonomic tree displays have been significantly improved, with an integrated search engine for easy navigation. The new version also includes post-translational modification data and detailed orthology information from PTMcode and eggNOG databases. Biological pathway and protein interaction data have been expanded and synchronized with the latest versions of Pathways Explorer and STRING databases. The new tabbed annotation interface allows users to access subsets of information as requested. The backend of SMART is a relational database powered by PostgreSQL, storing all annotation data and pre-calculated protein analyses. The database has been restructured to handle the increased number of annotated domains and features, which caused performance issues. SMART remains a popular and widely used resource for protein domain analysis. Funding for open access is provided by the European Molecular Biology Laboratory.