All screens contained within a publication or pre-publication article are referred to as a dataset in ORCS. A dataset may contain a single screen or many screens performed under a variety of conditions. Additionally, these screens may be reported in the literature in a single large data file that contains scores for multiple screens or may be reported in separate files.
To accommodate these differences ORCS allows multiple file uploads for all publications including a “supplemental file”, a completely unmodified file directly associated with the publication and provided in the same format as originally published by the author, and a “screen data file” which is a file that has been formatted for parsing by ORCS.
These files can then generate as many screen score sets as appropriate. For example, a single data file that reports scores for five screens performed in five different cell lines could be used to generate five separate screen score sets for a given dataset.