Pathology of Genetically Engineered and Other Mutant Mice

Databases fall into two main categories; those where data are gathered, expertly curated, and presented by the database project itself, often integrating data from several sources and analytical tools. The second type of database acts as an archival platform to which data are uploaded by users, often with minimal curation. There are several examples of hybrids of various proportions where uploads are solicited by the database, curated to a different degree, and presented with core data sourced and validated by the database staff. Curated databases may also be long‐term repositories or archives of legacy datasets. Data resources are generally funded out of project/program grants and core institutional funding. The degree of currency and the depth of curation often depends on funding and funding sustainability. Metrics and indicators for database description, and to a degree for assessing database quality, can be found in ELIXIR [1], as the criteria developed for qualification as an ELIXIR core database are widely applicable. A good source of rapidly accessible core information about databases is the annual database issue of Nucleic Acids Research . Databases present their core activities and modes of access in the papers but also publish updates as significant developments are achieved. It is always useful to browse this issue when looking for new sources of data.

The move to make scientific data openly discoverable, accessible, and useable has led to the formalization of the principles of open data, known as FAIRsharing [2]. Data has to be free, accessible, interoperable, and reusable. Increasingly databases are trying to comply with the FAIR semantic and access technical requirements as part of data discovery and interoperability, and many databases now provide access computationally through an Application Programming Interface (API) as well as through Hypertext Markup Language (HTML), together with complete data dumps and data reports. Searching using standardized semantics and ontologies provides very powerful ways of searching databases, and for databases such as MouseMine it is possible to use very powerful SPARQL Protocol and RDF (Resource Description Framework) Query Language (SPARQL) queries.

Along with standardization of database structures and access goes formalization of standards, essential for interoperability and data aggregation. Different communities have established specific standards for metadata, minimal information (MI) for reporting (MI standards), and other data structures. These standards and the characteristics of databases can now be found in databases of databases such as the very useful FAIRsharing database. FAIRsharing lists and describes standards used in databases but also the type of data and importantly the species. Currently, FAIR sharing lists 90 databases either dedicated to the mouse or with specifically mouse data [2].

The Re3data repository [3] lists 57 mouse data resources, interestingly with limited overlap with those from FAIRsharing, indicating the usefulness of looking at as many metaresources as possible. Of course, many databases, particularly small but often valuable ones, are not listed in these resource repositories. It is one aim in this chapter to introduce some of these smaller databases that are nevertheless very valuable.

Table 2.1 Websites with information useful for phenotyping mutant laboratory mice.

Many archival databases and some reference resources are very dependent on users to curate, correct uploads, and to give them added value for a community. Databases run for the use of specific communities can provide a high degree of added value at low cost and contribute heavily to scientific sustainability. Image databases especially fall into this category and in particular need new material from experts in the field to constantly expand the resource and accurately curate the material. Pathologists can use these resources as legacy tools for extra photomicrographs they take while working up new models or defining spontaneous diseases in colonies. Such additional images can be referenced in the primary publication as being publicly available on the websites, which usually is very well received by journals and reviewers on future grant applications. Many of the current public repositories will not accept files of the size of zoomable histopathology images and limited user uploads. This, together with payment to commercial repositories is a specific problem for very image intensive histopathology and cell biology data.