Work with us
How a project with CatalystNeuro runs, what the lab provides, and how to include the work in a grant application.
Project process
Most of our work is converting a lab's data to Neurodata Without Borders (NWB) and publishing it on the DANDI Archive. We have built conversion pipelines for 66 labs, 32 of them with published dandisets, and the code for each is public. Software engineering and AI projects follow the same pattern with a different deliverable.
- 01
Introductory call
A free 30-minute conversation about the data, the funder's requirements, and the timeline. We say plainly whether the work needs us, or whether NeuroConv or NWB GUIDE will get the lab there on its own.
- 02
Scope of work
After a kick-off meeting we send a written scope of work: which datasets and data streams are converted, what is published on the DANDI Archive and when, and any software or analysis work beyond the conversion.
- 03
Point person and data transfer
The lab names one technical point person who can answer questions about the data or route them to whoever can. We set up a transfer, usually through Globus, of the raw acquisition files and the outputs of the lab's processing pipeline.
- 04
Conversion and review
We build the conversion as an open-source repository, ask detailed questions about metadata and synchronization as they come up, and review sample files with the lab before converting everything.
- 05
Publication and handover
We upload to DANDI, embargoed or public, write example notebooks showing how to stream and analyze the data, and train the lab to run the pipeline on new sessions.
Lab responsibilities
- A point person with time to answer questions about the data
- Access to the raw data and the processed outputs
- A paper, preprint, or methods description if one exists
- Where the session metadata lives: subjects, sexes, ages, start times
The point person does not need to write code. The engagement guide lists the questions we ask at the start, and the Globus guide covers the transfer.
Grant applications
Our work has been funded through 17 grants and contracts from 10 funders, including the NIH, the Simons Foundation, the Michael J. Fox Foundation, and the Kavli Foundation (funded projects). If you are writing a data management and sharing plan or a budget that includes data standardization, contact us before the deadline and we will help you describe the work and estimate its size.
For the policy side of the plan, the NWB and DANDI guide explains what the NIH Data Management and Sharing Policy expects and why DANDI meets the repository criteria.
Ownership and licensing
Everything we write for a lab is released as open-source software under a permissive license, so there are no licensing fees and nothing to renew. The lab can copy, modify, and run the pipeline without us.