# The ABCC Video This video was a result of NeuroHacakademy welcoming Eric Earl and Eric Feczko to talk about the ABCC back in 2021. Recorded by Eric Earl with feedback from the rest of the team. ## YouTube Here it is on YouTube: [youtu.be/w3SSSCMF5Pg](https://youtu.be/w3SSSCMF5Pg) ## Video Transcript 0:01 Hi, I'm Eric Earl and I have been a research associate in the OHSU DCAN lab since 2013 and a data scientist in the NIMH data science and sharing team since 2021. 0:14 I'm in this 2021 Neurohackademy video today to introduce the ABCD BIDS community collection, otherwise known as the ABCC. 0:24 I'll be covering some of the history that brought the ABCC to where it is today. 0:30 The intent to make the collection more accessible for users and contributors, how to get data access through the NDA documentation about the collection on read the docs, some of the data subsets contained in the collections BIDS folder structure and how to give back to the community using github issues. 0:50 The adolescent brain cognitive development study also known as ABCD was funded about six years ago and is the largest long term study of brain development and child health in the United States. 1:03 With 21 different sites involved collecting 11,877 participants data. 1:12 The participants start between ages 9 and 10 and are studied longitudinally for 10 years. 1:18 Each participant gets one MRI scanning session every two years. 1:23 The data analysis informatics and resource center or DAIRC turns the MRI data around and releases it to the NIMH data archive or NDA as quick as they can, they even share quality control notes of the as acquired MRI data, the ABCC is centered around standardized versions of the as acquired MRI data using the brain imaging data structure or bids, which was introduced about five years ago. 1:52 Anyone should be able to get an NDA account and then request collection 3165 access. 2:00 I would call that a user. 2:02 We feel that anyone should then be welcome to give back any derivative data they make as a contributor and that process should be easy. 2:15 Now, I'm going to tell you how to get access to the data. 2:18 So sip your coffee, grab your notebook. 2:21 Here it comes. 2:24 You can go to this website or you can Google nda.nih.gov SOP dash 04, then read through the SOP dash 04, data access permission request. 2:42 It basically says you can request access for an institution or an individual. 2:48 We recommend the institutional way so you can request many people have access all at once and it's renewable once a year. 2:57 Here's the short version of the procedure in the top right corner, click my account and then request account, log in and click the data permissions tab then under NDA permission groups, find ABCD click the actions on the right to request access, download and fill out the data, use certification or DUC. 3:24 It asks for things like a description of your group, a statement about how you want to use the data, the people needing access and some signatures for the people. 3:34 Just request everybody that might use or even see the data in a separate spreadsheet or word document and attach it, then get it signed. 3:45 You'll need a signing official at your institution to sign it along with your principal investigator. 3:51 And that was the short version now on to the collection. 3:57 This is our read the docs website. 3:59 We keep our docs as up to date as we can. 4:02 And if you the user find something inconsistent or out of date, you can let us know more on that. 4:09 Later. 4:09 When I talk about raising github issues, there are six major sections of our docs release notes is a section on what each release has in it and some corrections that have happened over time inputs is a section covering the BIDS inputs, both what is available and how they were generated pipeline talks. 4:30 Primarily about the ABCD BIDS pipeline used for the initial release derivatives is primarily for explanations of what each derivative data subset is made up of recommendations is the page about how we recommend you use the data set. 4:46 You may even want to read this section first as it covers a lot of ground like the matched subject groups, quality control data types. 4:55 And more finally, the useful links page recaps the website links and references we mentioned throughout the docs. 5:04 And if you still have questions past that, please ask us by posting on the github issues. 5:11 Now, all about the data. 5:14 Thanks to our ABCD DICOM to BIDS code which uses DICOM to BIDS to call DCM2NII X. 5:22 We have standard BIDS, input data for T1 and T2 weighted anatomical images, diffusion weighted images, field maps and functional data for all four ABCD task types. 5:36 Those tasks are resting state, the monetary incentive delay task, the stop signal task and the N back task. 5:46 We only keep the scans that past DAIRC QC. 5:50 We understand that 10 megabytes of files per session will cost you 100 gigabytes to download and store across 10,000 MRI sessions. 6:01 So that's why we always attempt to make the collection's derivatives as small but useful of data subset packages that they can be right now. 6:11 There are just derivatives from our ABCD BIDS pipeline and utilities but coming soon are fmriprep and QSIprep derivatives as well. 6:20 Those extra derivatives will be released as soon as the NDA's quality assurance checks are all passed. 6:26 And here's the short list of other derivatives being included in the next release just around the corner. 6:32 Individual specific brain network labels based on both a template matching approach. 6:38 And an infomap community detection approach, all six task-based FMRI level one analyses and the three level two analyses which are ready for third level or group wise analyses. 6:54 Finally, github issues provides you a way to reach us all at once that is both transparent and public and still provides notifications. 7:04 It's like a public help desk for just our NDA collection. 7:09 If you want to make a good github issue, follow this advice. 7:13 Keep it concise but descriptive in your title. 7:17 Have a sufficient description that allows us to reproduce what you did or know what you mean. 7:22 Exactly. 7:23 Without uncertainty, feel free to send along suggestions, observations or even documentation edits. 7:31 Every contribution helps and that's it. 7:36 Thanks for watching. 7:38 We made it to the end and I'm pretty sure we're all still standing or sitting or whatever state of motion you're in. 7:46 I'm putting some key links here for you to reference later so you can take a screenshot. 7:50 Now. 7:51 There's the read the doc site, the NDA collection itself, our open science framework site and the github repo where you can report issues. 8:00 That's it. 8:01 Bye.