singR: An R package for Simultaneous non-Gaussian Component Analysis for data integration

TL;DR

singR is an R package that performs simultaneous non-Gaussian component analysis to effectively integrate and extract shared features from multiple datasets, especially useful for neuroimaging data with non-Gaussian characteristics.

Contribution

This paper introduces the singR package, implementing a novel non-Gaussian component analysis method for data integration across multiple datasets.

Findings

Successfully extracts shared features in neuroimaging data

Accurately reflects information shared across datasets

Demonstrates effectiveness with toy and simulated examples

Abstract

This paper introduces an R package that implements Simultaneous non-Gaussian Component Analysis for data integration. SING uses a non-Gaussian measure of information to extract feature loadings and scores (latent variables) that are shared across multiple datasets. We describe and implement functions through two examples. The first example is a toy example working with images. The second example is a simulated study integrating functional connectivity estimates from a restingstate functional magnetic resonance imaging dataset and task activation maps from a working memory functional magnetic resonance imaging dataset. The SING model can produce joint components that accurately reflect information shared by multiple datasets, particularly for datasets with non-Gaussian features such as neuroimaging.