Is there any compilation of the metadata from most different neuro data repositories?

,
1

I’m trying to create a searchable spreadsheet/csv with the metadata of almost all available datasets, and wondering whether anything similar has been done before.

For example, I’d like to have a csv with an entry for most available datasets. The entry would include wherever applicable modality, dataset size, organism, n_subjects, ages, condition(s), duration, brain region, license, authors, paper, maybe a few other things. This would allow for search through these datasets without needing to check many different repositories, and differentiate from KnowledgeSpace by showing metadata in a condensed view that can be aggregated to pivot tables, etc. It seems there are some csvs of metadata around, e.g. there are ones from OpenNeuro and Brain Library, but they crucially don’t include dataset size. So it takes quite a while just to check if a single type of data exists, much less how much data of several different specifications.

Anyone know of other approximations of this task or tool that have been collected?

Thanks,
Connor

2

I have tsv for open BIDS datalad datasets (openneuro and datalad superdatasets):

It does include (amongst other things):

  • modality
  • dataset size
  • n_subjects
  • task names but not conditions
  • authors

I could try to update the listing with species though for the vast majority it will be unspecified of those datasets.
emphasized text
Age is non standardized across datasets so it is mess: for this I would suggest checking neurobagel query tool: Query Tool
Should allow to get the age of subset of participants from openneuro.

3

Awesome, thanks! This is very useful, just to have sizes at least. I wonder if there are other metadata lists like this languishing out there.

4

FYI I will be updating the listing so it contains

  • license
  • references and links
  • duration of fmri scans

For the open BIDS datasets I have checked, I only found 3 that mentioned explicitly the species studied in a reliable fashion. So I won’t bother with that for now.