bioRxivJClub
Ridge Redundancy Analysis for High-Dimensional Omics Data
bioRxiv · · Preprint · Open access
Yoshioka, Aubert + 2 more
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
Ridge RDA for high-dimensional omics uses Singular Value Decomposition of the predictor matrix X to avoid covariance inversion and enable efficient joint tuning of r and {lambda}.
- Why it matters: Omics datasets often have strong multicollinearity and the “large p, small n” setting, making standard reduced rank regression difficult and requiring careful selection of both rank and regularization.
- What they did: The proposed computational framework leverages SVD to streamline computation and reduces coefficient storage from pq to (p+q)r, then performs cross-validated grid search over {lambda} and r.
- The result: Implemented in the R package rrda, the method provides a practical, scalable way to analyze high-dimensional omics data while jointly selecting rank and ridge penalty.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.
- Open access