Reports can be exported in JSON format. An exported JSON file contains 4 main elements:
- header. Contains information about the version of the JSON exporter and front page elements included in the report (the front page elements are visible in the PDF export of the report).
- data. Contains the actual data found in the report (sections, subsections, figures, tables, text).
- metadata. Contains information about metadata files the report referenced to.
- history. Contains information about the history of the report (as seen in the "Show history" view).
The data section contains nested elements following the structure of the report:
- The keys of sections (and subsections, etc) are formed from the section (and subsection, etc) title, with special characters replaced. For example, the section "Counted fragment by type (total)" is exported to an element with the key "counted_fragments_by_type_total".
- A section is made of the section title, the section number, and all other elements that are nested in it (e.g., other subsections, figures, tables, text).
- Figures, tables and text are exported to elements with keys "figure_n", "table_n" and "text_n", n being the number of the elements of that type in the report.
- Figures contain information about the titles of the figure, x axis, and y axis, as well as legend and data. This data is originally available in the Workbench by double clicking on a figure in a report and using the "Show Table" view.
- The names of table columns are transformed to keys in a similar way to section titles.
Once exported, the JSON file can be parsed and further processed. For example, using R and the package jsonlite, reports from different samples can be jointly analyzed. This enables easy comparison of any information present in the original reports across samples.
Note that the tool Combine Reports tool already provides a similar functionality, but the JSON export allows for more flexibility in what is compared across samples.