Scientific
Data is a Springer Nature journal focused on data publication. The main article
type is a Data Descriptor, which depicts rigorously described, reusable
datasets. Each Data Descriptor is accompanied by metadata represented in
the ISA-Tab format, which is a tabular format following the
Investigation/Study/Assay model.
The ISA-explorer tool [1] uses the information in the Scientific Data
ISA-Tab metadata files to facilitate dataset discovery. The tool allows
users to filter datasets by different criteria: for example, a specific
design type or data stored at a particular repository.. The filters can
be combined in a boolean search allowing users to easily and quickly
discover specific types of data.
The ISA-explorer also offers a generic search box, allowing for
keyword-based searches of the ISA-Tab metadata files. Additionally, the
tool allows browsing of dataset specific information such as related
publications and a visualisation of the distribution of the
characteristics of the samples involved in the study generating the
data.
We have some improvements already planned; for example, showing overall
statistics, supporting richer query searches considering the semantics
of the data (e.g. query expansion based on synonyms) and relying on
richer knowledge representations (such as linkedISA [2]).
[1]
http://blogs.nature.com/scientificdata/2015/12/17/isa-explorer/
[2] Alejandra González-Beltrán, Eamonn Maguire, Susanna-Assunta Sansone
and Philippe Rocca-Serra. linkedISA: semantic representation of ISA-Tab
experimental metadata. BMC Bioinformatics 2014, 15(Suppl 14):S4.
http://dx.doi.org/10.1186/1471-2105-15-S14-S4