Software

All of the methods and analyses were implemented in the Python programming language. We used the standard scientific Python stack for analysis: Numpy [74] as the basis for all numerical computing routines, Scipy [75] for various scientific utilities, and pandas for structured data manipulation [76]. For neuroimaging data analysis, we used the nibabel library for I/O operations and basic image manipulation (http://nipy.org/nibabel) and the nilearn library (http://nilearn.github.io) to plot brain slices. Word clouds were generated using the word_cloud package (https://github.com/amueller/word_cloud). An open-source implementation of our GC-LDA model, including documentation and examples, is publicly available on GitHub (https://github.com/timothyrubin/python_gclda). The topic maps reported here are available as a collection of interactive, downloadable whole-brain maps from the NeuroVault website (http://neurovault.org).