INDI Prospective Studies
Curated, ready-to-use subsets of INDI (International Neuroimaging Data-sharing Initiative) prospective releases, prepared for the Pittsburgh Fiber Data Hub. This repository aggregates links, metadata, and selected derived artifacts (e.g., QC tables, tractography-ready files) to simplify reuse in methods and replication studies.
Access note: Source data are hosted on external portals (mostly NITRC). Most downloads require a free account and joining the 1000 Functional Connectomes Project (FCP/INDI) group. Please review each dataset’s license and acknowledgments before use.
What this repository provides
For each dataset mirrored under data-indi/pro, this repository aims to provide:
- A short dataset overview (scope, modalities, phenotyping, design).
- Pointers to upstream NITRC / INDI project pages.
-
When available, helper artifacts such as:
participants.tsvqc.tsv- Tractography-ready derivatives (
*.gqi.fz,*.qsdr.fz,.sz,.dti.dz,.gqi.dz)
All raw imaging data remain hosted by INDI / NITRC and are subject to the corresponding data use agreements.
NYU IPN — Adult Resting-State & DTI
Dataset: NYU Institute for Pediatric Neuroscience (IPN) – Adult Resting-State and Diffusion MRI Institution: Phyllis Green & Randolph Cōwen IPN, NYU School of Medicine Sample: 49 psychiatrically screened, neurotypical adults (adult subset of a larger 6–55y project) Modalities:
- 6-min resting-state fMRI (1–2 runs per session; Rest_1 always before Rest_2)
- High-resolution T1-weighted MPRAGE (defaced)
- Two 64-direction DTI scans per participant
Phenotypes: Age, gender, IQ (WASI verbal / performance / composite) License: Creative Commons Attribution–NonCommercial (CC BY-NC)
The current release represents the adult portion of a larger normative connectivity project. Most subjects have two R-fMRI runs within one visit (< 1 hour apart), making this useful for:
- Short-interval test–retest of resting-state measures
- Structure–function analyses combining DTI and R-fMRI
- Method development for motion handling and reliability
⚠️ Some participants also appear in NewYork_a (FCP classic) and NYU_Cocaine (INDI retrospective). De-duplicate before pooling.
Upstream source: Project page — http://fcon_1000.projects.nitrc.org/indi/retro/nyu_ipn.html
Download (Linux / macOS — bash)
curl -s https://api.github.com/repos/data-indi/pro/releases/tags/nyu | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO
Download (Windows PowerShell 5.x)
(Invoke-RestMethod "https://api.github.com/repos/data-indi/pro/releases/tags/nyu").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }
Enhanced NKI-RS — Multiband Test–Retest Pilot
Dataset: Enhanced Nathan Kline Institute–Rockland Sample (NKI-RS) – Multiband Imaging Test–Retest Pilot
Institution: Nathan S. Kline Institute for Psychiatric Research (NKI), Orangeburg, NY, USA
PI: Michael Milham
DOI: 10.15387/fcp_indi.corr.nki1
Sample: Adults drawn from the original NKI-RS cohort (not excluding clinical history)
Purpose: Pilot phase to evaluate the reliability and reproducibility of advanced multiband R-fMRI and high-direction DTI protocols prior to the full 1,000-participant enhanced NKI-RS rollout.
Key multiband / test–retest sequences:
- R-mfMRI: TR 645 ms, 3 mm iso, 10 min (high temporal resolution)
- R-mfMRI: TR 1400 ms, 2 mm iso, 10 min (high spatial resolution)
- R-fMRI: TR 2500 ms, 3 mm iso, 5 min (standard reference)
- DTI: 137 directions, 2 mm iso
Tasks: Visual checkerboard, eye-movement calibration, breath-hold (with stimulus design scripts).
This dataset is well suited for:
- Comparing different TRs and sampling regimes
- Evaluating multiband preprocessing strategies
- Multimodal test–retest analyses across R-fMRI and DTI
Upstream source: Enhanced NKI-RS page — http://fcon_1000.projects.nitrc.org/indi/enhanced/nki_rs.html
Typical assets in this repo:
nki-pilot.dti.dz,nki-pilot.gqi.dzparticipants.tsv,qc.tsv- Subject-level
*_dwi.*.fzand*_T1w.nii.gz
Download (Linux / macOS — bash)
curl -s https://api.github.com/repos/data-indi/pro/releases/tags/nki-pilot | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO
Download (Windows PowerShell 5.x)
(Invoke-RestMethod "https://api.github.com/repos/data-indi/pro/releases/tags/nki-pilot").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }
NKI / Rockland Sample
Dataset: Nathan Kline Institute (NKI) / Rockland Sample Institution: NKI, Orangeburg, NY, USA PIs: F. Xavier Castellanos, Bennett Leventhal, Michael Milham Coordinator: Kate Nooner Sample: Community-ascertained 4–85-year-olds, prospectively released; 207 participants total
The NKI/Rockland Sample is a large, phenotypically rich resource for lifespan and psychiatric connectomics, with deep behavioral and clinical characterization.
Core imaging protocol per participant:
- 10-min resting-state fMRI (R-fMRI)
- 6-direction DTI
- 64-direction DTI (2 mm iso)
- High-resolution MPRAGE (structural T1)
- Short-sequence MPRAGE
- T2-weighted structural scan
All participants undergo semi-structured psychiatric interviews plus a comprehensive neuropsychological and behavioral battery, enabling:
- Brain–behavior modeling across the lifespan
- Dimensional analyses of psychopathology
- Method development in large, heterogeneous samples
Upstream source: Enhanced NKI-RS main page — http://fcon_1000.projects.nitrc.org/indi/enhanced/nki_rs.html
This repo: Provides selected derivatives (when available) and helper tables to simplify working with the NKI/Rockland releases.
Download (Linux / macOS — bash)
curl -s https://api.github.com/repos/data-indi/pro/releases/tags/nki | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO
Download (Windows PowerShell 5.x)
(Invoke-RestMethod "https://api.github.com/repos/data-indi/pro/releases/tags/nki").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }
Beijing Normal University — Short TR Sample
Dataset: Beijing Normal University – Short TR Sample Institution: State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, China Sample: 28 healthy college-aged volunteers Design: Within-subject comparison of two TR conditions in R-fMRI
Imaging protocol:
- Resting-state fMRI (long TR): TR = 2.0 s, 8 min, 240 volumes
- Resting-state fMRI (short TR): TR = 0.4 s, 8 min, 1200 volumes
- T1-weighted MPRAGE (defaced)
- 64-direction DTI (2 mm iso)
- Demographics: age, sex
Participants completed both R-fMRI TR conditions in a single session, supporting:
- Direct comparison of short vs long TR for connectivity measures
- Evaluation of temporal sampling effects, SNR, and reliability
- Combined structure/function analyses with DTI
Upstream source: Project page — http://fcon_1000.projects.nitrc.org/indi/retro/BeijingShortTR.html
Note: Data were released in multiple LiteNIFTI batches (BeijingShortTR.001.001 – BeijingShortTR.009.001) totaling 28 subjects.
Download (Linux / macOS — bash)
curl -s https://api.github.com/repos/data-indi/pro/releases/tags/bnu | jq -r '.assets[].browser_download_url' | xargs -n1 -P4 curl -LO
Download (Windows PowerShell 5.x)
(Invoke-RestMethod "https://api.github.com/repos/data-indi/pro/releases/tags/bnu").assets | ForEach-Object { Invoke-WebRequest $_.browser_download_url -OutFile (Split-Path $_.browser_download_url -Leaf) }
How to cite
For any analysis using these data:
- Cite the original dataset (and DOI where provided), plus key publications from each project.
- Include required funding acknowledgments from the upstream project pages.
- When you use derivatives or helper files from this repository, please add:
“Processed artifacts and curation were prepared by the Pittsburgh Fiber Data Hub (
data-indi/pro).”
Licensing
Unless otherwise specified on the upstream pages, these datasets are distributed under Creative Commons Attribution–NonCommercial (CC BY-NC):
- You may share and adapt the material for non-commercial research, with proper attribution.
- Always check each dataset’s NITRC / INDI page for dataset-specific conditions or additional restrictions.
Known overlaps / deduplication
-
NYU IPN (adult subset) contains participants who also appear in NewYork_a (FCP classic) and NYU_Cocaine (INDI retrospective).
- If you combine these datasets, perform subject-level deduplication to avoid double-counting individuals.
Similar overlaps may exist between the NKI/Rockland Sample and related NKI multiband test–retest datasets; check subject IDs if pooling.
Disclaimer
- All data originate from independent sites and are redistributed as-is.
- Acquisition parameters, scanner models, and instructions (eyes open/closed, fixation, etc.) vary across datasets.
-
Users are responsible for:
- Performing appropriate quality control (e.g., motion, artifacts).
- Choosing preprocessing pipelines suitable for multiband and short-TR acquisitions.
- Respecting all data use agreements and licensing terms.
Repository: data-indi/pro — INDI Prospective Data Sharing Samples curated by the Pittsburgh Fiber Data Hub.