| Title: | On the stability of canonical correlation analysis and partial least squares with application to brain-behavior associations |
| Journal: | Communications Biology |
| Published: | 21 Feb 2024 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/38383808/ |
| DOI: | https://doi.org/10.1038/s42003-024-05869-4 |
| URL: | https://www.nature.com/articles/s42003-024-05869-4.pdf |
Publication 8703
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
Associations between datasets can be discovered through multivariate methods like Canonical Correlation Analysis (CCA) or Partial Least Squares (PLS). A requisite property for interpretability and generalizability of CCA/PLS associations is stability of their feature patterns. However, stability of CCA/PLS in high-dimensional datasets is questionable, as found in empirical characterizations. To study these issues systematically, we developed a generative modeling framework to simulate synthetic datasets. We found that when sample size is relatively small, but comparable to typical studies, CCA/PLS associations are highly unstable and inaccurate; both in their magnitude and importantly in the feature pattern underlying the association. We confirmed these trends across two neuroimaging modalities and in independent datasets with n ≈ 1000 and n = 20,000, and found that only the latter comprised sufficient observations for stable mappings between imaging-derived and behavioral features. We further developed a power calculator to provide sample sizes required for stability and reliability of multivariate analyses. Collectively, we characterize how to limit detrimental effects of overfitting on CCA/PLS stability, and provide recommendations for future studies.</p>
9 Authors
- Markus Helmer
- Shaun Warrington
- Ali-Reza Mohammadi-Nejad
- Jie Lisa Ji
- Amber Howell
- Benjamin Rosand
- Alan Anticevic
- Stamatios N. Sotiropoulos
- John D. Murray
1 Application
| Application ID | Title |
|---|---|
| 43822 | Multi-Modal Analysis of the UK Biobank Neuroimaging Data |
Enabling scientific discoveries that improve human health