NIMH course for reproducibility in neuroimaging.

Links :

Course materials site is rendered here.

NIMH's Data Science and Sharing Team will be conducting a workshop on open and reproducible neuroscience the week of March 13th. Over the course of four mornings (8:30a to 12:30p Mon, Wed, Thur, & Fri) we will provide hands on training on:

• Using Python and Github to produce & share reproducible processing pipelines
• Structuring and reporting your data using community-recognized standards (COBIDAS, BIDS, EQUATOR)
• Accessing and contributing to share repositories
• Using pre-print servers (biorxiv.org)
• Accessing NIMH dedicated HPC resources

Mar 13, 15, 16, 17th, 8:30 am – 12:30 pm (Note Mar 14th is NIH Pi Day)

Note that space in the course is limited. Our priority is to aim to maximize the number of labs represents across the NIH Neuroscience Intramural Research Program.

Course Organizers: Peter Bandettini, SFIM, FMRIF Adam Thomas, DSST, FMRIF John Lee, DSST, FMRIF Rick Reynolds, SSCC Paul Taylor, SSCC Sara Kimmich, SFIM Matthew Brett

The course will:

• Take place March 13th, and 15th-17th at NIH.
• Draw heavily from the content of software/data carpentry
• Similar to the carpentries it will put emphasis on hands-on acquisition of programming skills.
• As with data carpentry it will maintain a single dataset throughout to give a sense of how all the tools tie together for an analysis.
• Be divided into 4 days (described in more depth in [google doc](https://docs.google.com/document/d /1RtLaNrbFtXLmj53_dGmolqh0iGRxseQ5d6LkG-ojv28/edit?usp=sharing)):

The course materials

The course materials are maintained in a similar manner to software carpentry from which we have imported this repository. A separate repository will be maintained for each day of the course. Jekyll is used to render the repository as a website. Changes pushed to the repository will update the website. In order to push said changes, clone this repository, make your changes, and then push the changes back to the remote. The content for the main page is in index.md. The content for the individual lessons is in _episodes/X.md. Until the content has been edited it may bear no relation to the actual course material that day. We will use the basic structure of the repository/website for convenience though. And there is some overlap in the material...

Testing the website locally

If you want to generate the website locally for testing purposes there are guides on software carpentry linked below.

• As described here you will need ruby (with package manager npm) and jekyll (with kramdown), and python 3 with the pyyaml module installed.

• You runmake serve to serve the website locally and then point your browser to the appropriate port.

• More generally when you have make installed typing make in the base directory will provide a list of helpful commands.

• If you're completely lost you can work you're way through this this description of how software carpentry lesson materials are created/maintained.

• Once setup the repository can generate a website locally. For example to host on localhost:11091 execute the command:

  jekyll serve --watch --port 11091&