[Newbie]Accessing anatomical and functional data in ABIDE for CPAC processing

[MeenVen]

I am totally new to neuro data analysis and interested in processing the autism abide dataset. I have installed CPAC successfully on my ubuntu. I am going through https://fcp-indi.github.io/docs/user/subject_list_config.html. In the section, 'Example File template', the file structures for the anatomical and functional templates for ABIDE are given. But I am wondering how to download these two templates ie the .nii.gz files from the ABIDE datasets. Can someone give me just one process and do a simple analysis on ABIDE data file with CPAC ?

Thank you. 

[Steven Giavasis]

Hello MeenVen,

I apologize we do not have instructions on how to get a small, quick run when you don't have data readily available- we will be creating this in the near future.

There are several ways- you could download ABIDE data directly and then create a subject list pointing to where you store it, or you could create a subject list that automatically pulls data from the FCP-INDI Amazon cloud storage. The latter is easier, especially if you just want to get up and running.

This page has some details about available data releases, as well as instructions on how to directly download data:

http://fcon_1000.projects.nitrc.org/indi/s3/index.html

However, if you'd like to pull directly from the cloud, you can simply enter S3 links as your filepath templates. For example, the Amazon S3-hosted ABIDE data release has its data stored in BIDS format, so you can tell the CPAC subject list builder to expect a BIDS data format, and give it this BIDS base directory (without the quotes):

"s3://fcp-indi/data/Projects/ABIDE/RawDataBIDS"

Note the full path to this RawDataBIDS folder, for ABIDE, and the other releases, are available on that page I linked above.

Providing this S3 link to the sublist builder will create a subject list populated with S3 links (like the one above), except to each individual subject's files. When you run CPAC with this list, it will automatically download the raw data into your working directory and continue as usual.

I hope this helps- please let me know if you need any assistance. We'll be improving our user guide soon for a more seamless experience.

Best,

Steve