ISA File Types

In this guide we introduce the different ISA file types used in the ARC.

The ARC builds on the ISA Abstract Model for metadata annotation. Briefly, the ISA model comes with a hierarchy (ISA: Investigation - Study - Assay) that aligns well with most projects in (plant) biology labs. It allows to group multiple assays to one study, and multiple studies to one investigation.

Your ARC has one isa.investigation.xlsx workbook at its root (i.e. every ARC collects the data to one investigation). Each study or assay that you add to your ARC contains one isa.study.xlsx or isa.assay.xlsx, respectively.

The most user-intuitive format of the ISA metadata framework is ISA-Tab. As the name suggest, it's a tabular format. Hence, you can view the files in a spread-sheet program of choice.

The major difference between the ISA workbooks is their read-direction:

• isa.investigation.xlsx is read top-to-bottom (keys on the left, values extending to the right)
• isa.study.xlsx and isa.assay.xlsx left-to-right (keys on top, value extending to the bottom)

The isa.investigation.xlsx allows to store metadata relevant on the investigation-level (e.g. title, date, contributor and publication details of the investigation). In addition, it functions as a "registry" to your ARC.

Each study (isa.study.xlsx) and assay (isa.assay.xlsx) of your ARC as well as a summary of metadata contained in them are registered and listed in the isa.investigation.xlsx.

💡 When opening the isa.investigation.xlsx for the first time, it may be necessary to widen the first column to make the entries visible.

The output of one study or assay can function as the input to another study or assay. By using the same unique identifiers across your isa.study.xlsx and isa.assay.xlsx workbooks, respectively, you can communicate how the experimental processes and workflows connect.

By linking files stored in your ARC (e.g. raw data files in a dataset folder), you can let others know which experimental workflow was followed to produce these data files.

DataPLANT Support