NARPS Open Pipelines

[JLefortBesnard]

Title

NARPS Open Pipelines

Short description and the goals for the OHBM BrainHack

The goal of the NARPS Open Pipelines project is to create a codebase reproducing the 70 pipelines of the NARPS study (Botvinik-Nezer et al., 2020) and share this as an open resource for the community. We base our reproductions on the original descriptions provided by the teams and test the quality of the reproductions by comparing our results with the original results published on NeuroVault.

The Neuroimaging Analysis Replication and Prediction Study (NARPS-Botvinik-Nezer et al., 2020) aimed to provide the first scientific evidence on the variability of results across analysis teams in neuroscience. 70 teams were asked to analyze the same dataset (task-fMRI with 108 participants) and then, to provide their methods and results to be later analyzed and compared.

We would like to focus on these tasks during the Brainhack:

• Start writing the pseudo code of a pipeline based on the knowledge of participants (i.e.: which fMRI analysis software they are used to, whether they use Python or not, ...)

• Start reproducing the pipeline from the pseudo code,

• Improving documentation and accessibility of the project.

Link to the Project

https://github.com/Inria-Empenn/narps_open_pipelines

Image/Logo for the OHBM brainhack website

https://github.com/Inria-Empenn/narps_open_pipelines/blob/main/assets/images/project_illustration.png

Project lead

Lefort-Besnard jlefortbesnard jlefortbesnard

Main Hub

Seoul

Link to the Project pitch

No response

Other hubs covered by the leaders

• Seoul
• Hybrid (Asia / Pacific)
• Hybrid (Europe / Middle East / Africa)
• Hybrid (Americas)

Skills

Necessary:

• task-fMRI preprocessing or task-fMRI analysis

Useful but not necessary:

• Programming (any language)

• Nipype (Python) or SPM or AFNI or FSL

• GLM

• Basic knowledge of git and GitHub

Recommended tutorials for new contributors

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7404612/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6319393/
https://fmriprep.org/en/stable/
https://andysbrainbook.readthedocs.io/en/latest/fMRI_Short_Course/fMRI_04_Preprocessing.html
https://brainlife.io/docs/tutorial/fmri-preprocessing-tutorial/
https://school-brainhack.github.io/modules/git_github/

Good first issues

To start with, participants could draft pseudo code outlining the data preprocessing and analysis procedures, mirroring the approach taken by one of the 70 NARPS teams. Initially, they would replicate this for a pipeline that has already been completely reproduced. Following this, they would repeat the process for a pipeline that has not yet been reproduced.

The pseudo code will specify the preprocessing and analysis steps without delving into the actual coding with Nipype, or at most using command lines from SPM/FSL/Afni for executing these specific steps.

Twitter summary

Help the neuroimaging community improve reproducibility by joining our hackathon project! The goal of the NARPS Open Pipelines project is to create a codebase reproducing the 70 task-fmri pipelines of the NARPS study (Botvinik-Nezer et al., 2020) and share this as an open resource for the community.
#OHBMHackathon #Brainhack #OHBM2024

Short name for the Discord chat channel (~15 chars)

NARPS_Open_Pipelines

Please read and follow the OHBM Code of Conduct

• I agree to follow the OHBM Code of Conduct during the hackathon

[sina-mansour]

@JLefortBesnard many thanks for submitting this project for the 2024 Brainhack!

As you have indicated the project will also be made available in a hybrid format, we encourage recording a short project pitch to be made available to online participants. Please let us know if you wish to add a link to the project pitch before finalizing the project submission. (we can provide a link to the recorded pitch on the brainhack website)

[sina-mansour]

I have approved the project to be listed on the brainhack website. Please feel free to let us know if you record a pitch video at a later time and you like that to be included for online/hybrid participants.