InterProScan 5 is a new version of the widely used protein function prediction software. It is based on a Java-based architecture and provides a flexible and stable system that can use multiprocessor machines and/or conventional clusters for scalable distributed data analysis. The software is freely available for download from the EMBI-EBI FTP site and the open source code is hosted at Google Code. InterProScan 5 has been redeveloped to address the shortcomings of previous versions, including difficulties in installation, scalability, and extension of functionality. The new version allows for the analysis of millions of sequences and provides a more efficient and scalable system. The software includes new features such as the ability to function on a massive scale, the use of a single 'Master' process for managing analysis steps, and the ability to run on a wide range of computing environments. The software also includes new analysis algorithms and features, such as Phobius, which is a combined transmembrane topology and signal peptide predictor. InterProScan 5 also provides new output formats, including GFF3, SVG, and XML, and supports web services for lookup of precomputed results. The software is primarily designed to run on the command line, but users can also obtain a graphical view of their results using output format options. The InterProScan 5 distribution can be downloaded for local installation from the EMBL-EBI FTP site, and users can also search their sequences using a hosted version via the InterPro Web site or SOAP and REST web services. The source code is available from a Google Code repository, along with documentation for both developers and end-users. InterProScan 5 is a widely used tool for protein and nucleotide sequence analysis, and it allows users to obtain a better overview of what their results mean by adding valuable information from the InterPro database. The new implementation of InterProScan aims to address all of the shortcomings of the previous versions of InterProScan without losing important functionality. The full reimplementation of the software into a robust, data model-driven, Java-based architecture gives considerable improvements in scalability and usability, with the new version much easier to install and configure. New features have been added, including the provision of an additional analysis algorithm (Phobius); new (GFF3, SVG) and improved (XML, HTML) output formats; new web services for lookup of precomputed results and the ability to infer the potential membership of proteins in pathways (via InterPro entry mappings to KEGG, MetaCyc, UniPathway, etc.). The hosting of the source code at Google Code also signifies a significant step toward InterProScan becoming a fully open source software development project.InterProScan 5 is a new version of the widely used protein function prediction software. It is based on a Java-based architecture and provides a flexible and stable system that can use multiprocessor machines and/or conventional clusters for scalable distributed data analysis. The software is freely available for download from the EMBI-EBI FTP site and the open source code is hosted at Google Code. InterProScan 5 has been redeveloped to address the shortcomings of previous versions, including difficulties in installation, scalability, and extension of functionality. The new version allows for the analysis of millions of sequences and provides a more efficient and scalable system. The software includes new features such as the ability to function on a massive scale, the use of a single 'Master' process for managing analysis steps, and the ability to run on a wide range of computing environments. The software also includes new analysis algorithms and features, such as Phobius, which is a combined transmembrane topology and signal peptide predictor. InterProScan 5 also provides new output formats, including GFF3, SVG, and XML, and supports web services for lookup of precomputed results. The software is primarily designed to run on the command line, but users can also obtain a graphical view of their results using output format options. The InterProScan 5 distribution can be downloaded for local installation from the EMBL-EBI FTP site, and users can also search their sequences using a hosted version via the InterPro Web site or SOAP and REST web services. The source code is available from a Google Code repository, along with documentation for both developers and end-users. InterProScan 5 is a widely used tool for protein and nucleotide sequence analysis, and it allows users to obtain a better overview of what their results mean by adding valuable information from the InterPro database. The new implementation of InterProScan aims to address all of the shortcomings of the previous versions of InterProScan without losing important functionality. The full reimplementation of the software into a robust, data model-driven, Java-based architecture gives considerable improvements in scalability and usability, with the new version much easier to install and configure. New features have been added, including the provision of an additional analysis algorithm (Phobius); new (GFF3, SVG) and improved (XML, HTML) output formats; new web services for lookup of precomputed results and the ability to infer the potential membership of proteins in pathways (via InterPro entry mappings to KEGG, MetaCyc, UniPathway, etc.). The hosting of the source code at Google Code also signifies a significant step toward InterProScan becoming a fully open source software development project.