PyMVPA hyperalignment tutorial data sets

1

Hi @yarikoptic and @eknahm (and anyone else who might have any insight, here) !

I recently came across the PyMVPA hyperalignment tutorial dataset and realized that it could be a great addition to a re-analysis I’m working on.

I was wondering if the original data is still available, or if we should just move forward with the preprocessed ROI-level information ? I did see that the movie (but not task) data from the associated paper is available through DataLad.

Thanks so much for your insight, and for all of your work making these great resources !

Best,

Elizabeth

3 Likes
2

Hi Elizabeth,

Quick one: that dataset from the original hyperalignment data, codenamed “monkey dog”, unfortunately we never got to share - wasn’t BIDS to start with, lack of incentive and so on. But someone (I was not even original owner/curator) might share it some day :wink: But since then other datasets used in subsequent papers, such as http://datasets.datalad.org/?dir=/openfmri/ds000241
(codename AK6, or “Animal Kingdom 6”) was shared, but I am not sure if “subject correspondence” between raiders and that one on openneuro was if you decide to cross-analyze (or well – might be an easy match task based on anatomy ;))… Also if you look at http://datasets.datalad.org/?dir=/labs/haxby – you would see 3 datasets, including raiders, with some (might be too few) subjects seems to be overlapping so you could get data with naturalistic and task in the same participants.

I have let those who might know about hyperalignment and those datasets know – they might shine more light :wink:

3

Thank you, @yarikoptic ! I appreciate your help – and the study code names :laughing:

I’ll keep an eye out for more updates, too !