Building data set

It is possible to browse and download data at study level without creating any projects. However projects allow to include

  • From one side, only the needed experiment samples (some studies contain hundreds of samples, many of those may not be relevant)
  • From another side, combine multiple studies in one downloaded archive, generalizing more data than has been published in a single work.
  • You can also select which transcriptomes to use, even if all four available are currently enabled by default.

To assemble a dataset for calculation or export, use the checkboxes in the table.

  • Study checkbox selects or deselects the entire study.
  • Sample checkboxes allow selecting only specific samples within a study.
  • When some—but not all—samples are selected, the study checkbox shows a partially selected state.

You can build datasets by ticking checkboxes, but a faster approach is to first filter (by selecting categories in the data tree or typing into the search box) and then use Select or Unselect to affect all currently visible rows.

Bulk actions respect the active filters, so only items shown by the current view are selected or unselected.

After making your selection, press Create project. A project contains all currently selected studies and samples and is the primary unit for calculation and export.

Building the data set

For comparison projects (leaves vs roots, treated vs untreated), we suggest creating multiple projects, each representing a specific case (leaves, roots, shots, etc, or treated vs untreated). This means less custom filtering code on your side as Verdanta filters for you.