DEEP Computational Metadata
Process documentation
In DEEP, standardized computational analysis pipelines are documented in form of so-called Process specifications. A Process document lists required input and reference files, specifies the output generated by the pipeline and provides a step-by-step description of the computational analysis. Processes are in XML format and are rendered into a readable form when opened in a standard web browser (assuming an active internet connection). They are lightweight enough to be send via e-mail to collaboration partners.
Analysis documentation
Each run of a standardized pipeline produces an Analysis metadata document at the end. These are simple tab-separated text files listing all input, output and reference files as well as non-default parameters in form of key-value entries. Whereas a Process document has to be manually created and describes an analysis pipeline in generic form, an Analysis metadata file is created automatically. Consequently, these files are not accessible via this repository.
Reference
See the corresponding publication in Oxford Database for details: "A general concept for consistent documentation of computational analyses", Database 2015