Hi biostars,
I am trying to perform a differential expression analysis with RNA-seq data. My dependent variable Y are tumor growth estimates. Each estimate is associated with a varying degree of error. Now here is my question, is there a way to model Y incorporating the uncertainty of each data point using the limma-voom package?
In their paper from this year Ritchie et al I found this section:
Quantitative weights allow for unequal quality
Another unique feature of limma is the ability to incorporate quantitative weights into all levels of the statistical analysis, from normalization to linear modelling and gene set testing. Weights can be applied to genes or to RNA samples or to individual expression values. Weights can be used to give more emphasis to control probes during normalization, or can be used to down-weight measurements or samples that are less reliable in a gene expression analysis. The weights can be preset based on external quality information, or may be estimated from the expression data itself. The use of weights increases power to detect differentially expressed genes, and having a model based approach avoids the need for ad hoc decisions about which observations or samples to filter out (11).
I am not a statistician but it seems to me this only discusses the use of weights in the independent variables.
Can I add weights to my dependent variable?
Thanks
