Sook-Lei Liew (sliew@usc.edu), James M. Finley (jmfinley@pt.usc.edu), Keith Lohse (Keith.Lohse@health.utah.edu)
This github repository has information and materials for the American Society for Neurorehabilitation Meeting - Symposium on Reliability and Reproducibility in Neurorehabilitation. Code is now updated! If you're new to github, you can click the "clone or download" button on this repository to download a zip file.
Also, if you plan to attend, please help us tailor this symposium to your needs by filling out this super quick survey: https://forms.gle/wesLsNyKwG9h1mTY9 - Thanks!
Each section of this symposium will use a mix of powerpoint and hands-on programming examples. During the hands-on portion, participants can ‘walk through’ code that we will make available for download prior to the session, and get a feel for executing the code in real time along with the presenter. Alternatively, participants can opt to just watch the presenter execute the code, or they can partner with someone else. This is meant to provide participants with breadth, and a little depth, to understand tools available for reproducible science. Because of this, we will use three different programming languages/environments (Python, Matlab, R) so that participants can see the different interfaces and learn about some of their strengths. This session is meant to be as interactive as possible to give participants an opportunity to get familiar with code that they can adapt for their own purpose after the session - but participants are also welcome to observe if they are most comfortable with that.
Please install the following software (listed below) from good, reliable wifi! :) You should budget a few hours of time for installing all the software. You can also download some code and datasets to play with (indicated below).
12:15 - 1 pm Troubleshooting - Keith, James, and Lei will be on hand to help troubleshoot and answer any install questions
1 - 1:15 Introduction to Reliability and Reproducibility in Neurorehab Research (Lei)
1:15 - 1:45 Data Management - Best practices and a hands-on tutorial in Python (Lei)
1:45 - 2:15 Data Analysis - Best practices and a hands-on tutorial in Matlab (James)
2:15 - 2:45 Data Visualization - Best practices and a hands-on tutorial in R (Keith)
2:45 - 3 Questions and Discussion
Keep in touch with us on twitter and ask us any questions - let’s keep this conversation going! - @NPNLatUSC, @JamesMFinley, @Keith_Lohse
Install Python version 3.7 and Jupyter Notebook using the Anaconda distribution: https://www.anaconda.com/distribution/
If for some reason you install Python separately (highly unadvised!), you should:
-install Jupyter Notebook from here: https://jupyter.org/install
-also install these additional libraries, which should be included if you used the Anaconda distribution (e.g., you shouldn’t need to do anything else), but if you get an error, you will need to use pip install for the following:
Pandas - https://pandas.pydata.org/pandas-docs/stable/install.html
Scipy/Numpy - https://scipy.org/install.html
More on pip install (which you should run from a terminal): https://pip.pypa.io/en/stable/
Version - 2016a and above (Download from matlab page: https://www.mathworks.com/downloads/web_downloads/) Note: Unlike Python and R, Matlab costs money. But most universities have a licensing agreement. You can contact your department IT to see if it is available.
Any version of R will likely be satisfactory, but attendees are encouraged to install the latest version, 3.6.1.
A walk through of R installation can be found here: https://vimeo.com/201690416
Participants will also need to install ggplot2 and lme4. A walk through of package installation: https://vimeo.com/201696152
Github code and data is now updated on this Github repository. Follow us on Twitter for updates: @NPNLatUSC, @JamesMFinley, @Keith_Lohse!
Optional: ATLAS Stroke MRI and Lesion Mask Dataset: http://fcon_1000.projects.nitrc.org/indi/retro/atlas.html (Please allow at least 1 day for processing).