SMART (Simple Modular Architecture Research Tool) is a web resource for identifying and annotating protein domains and analyzing domain architectures. Version 9 includes 68 new models for diverse recombinase families, providing comprehensive coverage of mobile element recombinases. The underlying protein databases have been synchronized with UniProt, Ensembl, and STRING, increasing the number of annotated domains. The vector-based protein display engine has been updated to use latest web technologies, and domain architecture analysis components have been optimized. The database now contains over 1.1 million proteins from a non-redundant database. A 'genomic' analysis mode includes proteins from completely sequenced genomes, with data synchronized with STRING version 11. The protein visualization engine now uses interactive SVG for visualization, allowing zooming and saving as vector graphics. Expanded protein interaction data is available for over 21 million proteins. Domain architecture analysis functions allow users to access proteins with specific domain combinations, and data can be exported as FASTA files or phylogenetic trees. The phylogenetic tree export is compatible with the Interactive Tree of Life (iTOL) version 5. The database and web server have been optimized for performance, with improved processing speed and reduced response times. SMART data is freely available to academic users through EMBLem. Funding was provided by de.NBI and EMBL.SMART (Simple Modular Architecture Research Tool) is a web resource for identifying and annotating protein domains and analyzing domain architectures. Version 9 includes 68 new models for diverse recombinase families, providing comprehensive coverage of mobile element recombinases. The underlying protein databases have been synchronized with UniProt, Ensembl, and STRING, increasing the number of annotated domains. The vector-based protein display engine has been updated to use latest web technologies, and domain architecture analysis components have been optimized. The database now contains over 1.1 million proteins from a non-redundant database. A 'genomic' analysis mode includes proteins from completely sequenced genomes, with data synchronized with STRING version 11. The protein visualization engine now uses interactive SVG for visualization, allowing zooming and saving as vector graphics. Expanded protein interaction data is available for over 21 million proteins. Domain architecture analysis functions allow users to access proteins with specific domain combinations, and data can be exported as FASTA files or phylogenetic trees. The phylogenetic tree export is compatible with the Interactive Tree of Life (iTOL) version 5. The database and web server have been optimized for performance, with improved processing speed and reduced response times. SMART data is freely available to academic users through EMBLem. Funding was provided by de.NBI and EMBL.