Pyrcca: Regularized Kernel Canonical Correlation Analysis in Python and Its Applications to Neuroimaging

• Published in: Frontiers in neuroinformatics Vol. 10; p. 49
• Main Authors: Bilenko, Natalia Y., Gallant, Jack L.
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 22.11.2016; Frontiers Media S.A
• Subjects: Analysis of covariance; Bioinformatics; Brain architecture; Brain mapping; canonical correlation analysis; Correlation analysis; covariance analysis; cross-subject alignment; Datasets; Eigenvalues; fMRI; Functional magnetic resonance imaging; Medical imaging; Methods; Multivariate analysis; Neuroimaging; Neuroscience; partial least squares regression; Principal components analysis; python; Registration; Retina; United States > US; California; fMRI; covariance analysis; cross-subject alignment; partial least squares regression; canonical correlation analysis; Python
• ISSN: 1662-5196; 1662-5196
• DOI: 10.3389/fninf.2016.00049

In this article we introduce Pyrcca, an open-source Python package for performing canonical correlation analysis (CCA). CCA is a multivariate analysis method for identifying relationships between sets of variables. Pyrcca supports CCA with or without regularization, and with or without linear, polynomial, or Gaussian kernelization. We first use an abstract example to describe Pyrcca functionality. We then demonstrate how Pyrcca can be used to analyze neuroimaging data. Specifically, we use Pyrcca to implement cross-subject comparison in a natural movie functional magnetic resonance imaging (fMRI) experiment by finding a data-driven set of functional response patterns that are similar across individuals. We validate this cross-subject comparison method in Pyrcca by predicting responses to novel natural movies across subjects. Finally, we show how Pyrcca can reveal retinotopic organization in brain responses to natural movies without the need for an explicit model.