Will, great package! I have a conceptual question about GLM-PCA with covariates: when I have sample (or gene) covariates, is GLM-PCA estimating a factor model on the variance that is not explained by the covariates? How is GLM-PCA with covariates similar or different from SVA or RUV methods (beyond different link functions)?
Will, great package! I have a conceptual question about GLM-PCA with covariates: when I have sample (or gene) covariates, is GLM-PCA estimating a factor model on the variance that is not explained by the covariates? How is GLM-PCA with covariates similar or different from SVA or RUV methods (beyond different link functions)?