Hi, Is there a work around for ComBat() where I can run the function if I have only one sample in one batch? I'm working with affymetrix, human chip U133A and trying to remove batch effect on scan date. I have some samples that were scanned individually on separate dates. So there are some dates where only one sample was run. Any insight regarding this will be appreciated. Thanks! -- output of sessionInfo(): > sessionInfo() R version 3.1.0 (2014-04-10) Platform: x86_64-apple-darwin13.1.0 (64-bit) locale: [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8 attached base packages: [1] parallel stats graphics grDevices utils datasets methods base other attached packages: [1] bladderbatch_1.2.0 sva_3.10.0 mgcv_1.7-29 nlme_3.1-117 corpcor_1.6.6 affy_1.42.2 [7] Biobase_2.24.0 BiocGenerics_0.10.0 BiocInstaller_1.14.2 loaded via a namespace (and not attached): [1] affyio_1.32.0 grid_3.1.0 lattice_0.20-29 Matrix_1.1-3 preprocessCore_1.26.1 [6] tools_3.1.0 zlibbioc_1.10.0 -- Sent via the guest posting facility at bioconductor.org.

On Fri, Jun 6, 2014 at 11:43 AM, Guest [guest] <guest at="" bioconductor.org=""> wrote: > Hi, > > Is there a work around for ComBat() where I can run the function if I have only one sample in one batch? No. ComBat adjusts location (mean) and scale (variance) for each batch. There's no way to estimate/adjust scale with just one sample in a batch. > > I'm working with affymetrix, human chip U133A and trying to remove batch effect on scan date. I have some samples that were scanned individually on separate dates. So there are some dates where only one sample was run. > > Any insight regarding this will be appreciated. Produce a sample clustering tree to see how different the samples scanned on close dates are - often they are not very different and can be combined into a single batch (not ideal but better than nothing). Make sure the resulting batches are not completely confounded with your variable of interest (treatment or condition). Peter