Issues Identified with Data Release 2.0.1
Issues have been identified with ABCD Data Release 2.0.1 that impact some of the neuroimaging and genetics data. See below for details.
fMRI:
Due to incorrect post-processing, all task and resting-state fMRI data obtained on Phillips scanners should be excluded from all analyses. This includes tabulated and minimally processed data. This issue does not affect other modalities (sMRI, dMRI) or raw DICOM data (Fast Track). Philips fMRI data have been removed from DEAP and RDS. They will be corrected and made available with Data Release 3.0.
Philips fMRI minimally processed and tabulated data can be excluded by using the ABCD MRI Info instrument (abcd_mri01) and exclude by: mri_info_manufacturer = “Philips Medical Systems”. An R script is available at https://github.com/ABCD-STUDY/fMRI-cleanup to remove Philips fMRI data from tabulated data.
We recommend that all users re-run fMRI related analyses without the Philips fMRI data. For published studies, we encourage users to confirm findings without the Philips fMRI data and provide a corrigendum with the updated results.
Genetics:
1) We found that 111 participants have incorrect genetic data. We will resolve this issue in Data Release 3.0. Meanwhile, we recommend excluding these data from genetic analyses. PGUIDs are provided to registered users via the NIMH Data Archive.
2) The Family ID and relationship between participants (rel_family_id and rel_relationship) in NDA instrument acspsw03 as well as in the RDS file is based on information obtained from guardians at enrollment and is not genetically informed. Currently, the Genetics Working Group, Coordinating Center (CC) and the DAIRC are collaboratively creating more comprehensive measures of family membership to replace Family ID for future releases to enhance analyses accounting for family structure, including twin-based analyses.
Detailed information regarding known issues in Data Release 2.0.1 is available on the NIMH Data Archive.
ABCD Data Release 2.0.1 available now on the NIMH Data Archive
Due to reporting compilation and processing errors in 2.0 Data Release, a 2.0.1 Fix Release has been issued. Please ensure curated data (datasheets, minimally processed data) from the original Data Release 2.0 are replaced with data from 2.0.1 Fix Release. The following release notes were updated to reflect these changes:
- NDA 2.0.1 Release Notes ABCD README FIRST
- NDA 2.0.1 Release Notes Imaging Instruments
- NDA 2.0.1 Changes and Known Issues Fix Release 2.0.1
- NDA 2.0.1 Diffusion Magnetic Resonance Imaging
- NDA 2.0.1 Task-Based Functional Magnetic Resonance Imaging
- NDA 2.0.1 Mental Health
- NDA 2.0.1 Genetics
Registered users can obtain more information from https://nda.nih.gov/study.html?id=721, and access updated data via Option One or Option Two using the NDA Query Tool – https://nda.nih.gov/general-query.html. For downloading only updated minimally processed imaging data, add the Minimally Processed “Release 2.0.1” to the Workspace via Option Two.
Issues Identified with Data Release 2.0
Problems were identified with the data in 11 imaging-related files (5 diffusion tensor imaging (DTI) files and 6 fMRI Monetary Incentive Delay (MID) task files). See below for more detail. Corrected files will be re-uploaded to the NIMH Data Archive and available soon to users.
DTI subcortical and cortical ROIs and white fiber tracts (Part 1)
DTI subcortical and cortical ROIs and white matter fiber tracts (Part 4)
dMRI DTI Destrieux Parcellations Part 1
dMRI DTI Destrieux Parcellations Part 2
dMRI DTI Full Part 2
Task fMRI MID Average SEM Destriex Parcellations Part 2
Task fMRI MID Run 1 Beta Weights Destrieux Parcellations Part 2
Task fMRI MID Run 2 Beta Weights Destrieux Parcellations Part 2
Task fMRI MID Average Beta Weights Destrieux Parcellations Part 2
Task fMRI MID Run 1 SEM Destrieux Parcellations Part 2
Task fMRI MID Run 2 SEM Destrieux Parcellations Part 2
An error was also discovered in imaging data collected from Siemens scanners between September 2017 and December 2017 where structural images are flipped left-right. These data will be updated in a patch release later this year.
Detailed information regarding known issues in Data Release 2.0 is available on the NIMH Data Archive.
ABCD Data Release 2.0 available now on the NIMH Data Archive!
The second annual curated ABCD data release 2.0 is available now on the NIMH Data Archive. Data Release 2.0 includes baseline data on the full participant cohort, ages 9-10 years. This release contains much of the same type of data as in Data Release 1.1, and also includes genotypic data for the first time. Smokescreen genotyping array data are available on almost 11,000 participants. These include common variations, as well as variations associated with addiction, smoking behavior and nicotine metabolism.
ABCD Study data will be released annually. The next data release will be in early summer 2020 and will include the first longitudinal data from the 6-month and 1-year follow-up assessments.
For more information about Data Release 2.0, visit the ABCD data collection on the NIMH Data Archive. Full details are included in the Release Notes for Data Release 2.0.
Researchers new to the NIMH Data Archive system interested in gaining access to the data can create an account here and follow the instructions to request access. Returning researchers can log into their account using the Login button.
Click here for full baseline data demographics (n = 11,878 participants; 48% female; 52% male)
