Skip to main content
. 2011 Sep 16;12:369. doi: 10.1186/1471-2105-12-369

Figure 3.

Figure 3

fRMA can be used with incrementally growing datasets. Distribution of difference in expression estimates for 22283 probesets across 200 breast tumor arrays (GSE11121) when preprocessing with fRMA trained using the first N batches and fRMA trained using all batches.