This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Limma - Special Case Of Contrasting

Dear all!

I'm trying to identify differentially expressed genes (DEGs) using limma with the following Targets file:

CEL_file    donor    treatmemt    cell_type 
file1    1    A    C2
file2    1    B    C2
file3    2    A    C2
file4    2    B    C2
file5    2    A    C1
file6    2    B    C1
file7    3    A    C2
file8    3    B    C2
file9    3    A    C1
file10    3    B    C1
file11    4    A    C2
file12    4    B    C2
file13    4    A    C1
file14    4    B    C1
file15    5    A    C2
file16    5    B    C2
file17    6    A    C2
file18    6    B    C2
file19    6    A    C1
file20    6    B    C1
file21    7    A    C2
file22    7    B    C2
file23    8    A    C2
file24    8    B    C2
file25    8    A    C1
file26    8    B    C1
file27    9    A    C2
file28    9    B    C2
file29    9    A    C1
file30    9    B    C1
file31    10    A    C2
file32    10    B    C2

I'd like to compare C1.A vs. (C1.B, C2.A, C2.B) while blocking the donors and get a list of DEGs. I'd greatly appreciate any ideas on how to approach this. Btw, I've already conducted C1.A vs. C1.B and C2.A vs. C2.B comparisons.

Thanks,

Mitja

sessionInfo()

R version 2.15.1 (2012-06-22)
Platform: x86_64-apple-darwin9.8.0/x86_64 (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] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] statmod_1.4.16             hgu133plus2cdf_2.10.0      AnnotationDbi_1.18.4       arrayQualityMetrics_3.12.0
[5] limma_3.12.3               affy_1.34.0                Biobase_2.16.0             BiocGenerics_0.2.0        

loaded via a namespace (and not attached):
 [1] affyio_1.24.0         affyPLM_1.32.0        annotate_1.34.1       beadarray_2.6.0       BeadDataPackR_1.8.0  
 [6] BiocInstaller_1.4.9   Biostrings_2.24.1     Cairo_1.5-2           cluster_1.14.3        colorspace_1.2-0     
[11] DBI_0.2-5             genefilter_1.38.0     grid_2.15.1           Hmisc_3.10-1          hwriter_1.3          
[16] IRanges_1.14.4        KernSmooth_2.23-8     lattice_0.20-10       latticeExtra_0.6-24   plyr_1.7.1           
[21] preprocessCore_1.18.0 RColorBrewer_1.0-5    reshape2_1.2.1        RSQLite_0.11.2        setRNG_2011.11-2     
[26] splines_2.15.1        stats4_2.15.1         stringr_0.6.1         survival_2.36-14      SVGAnnotation_0.93-1 
[31] tools_2.15.1          vsn_3.24.0            XML_3.95-0.1          xtable_1.7-0          zlibbioc_1.2.0
limma
• 2,818 views • link
updated by Ryan Thompson Ryan Thompson • written by mitmitrovic • mitmitrovic

I started editing this to make it more readable, but got bored. Please indent lines of code/data with 4 spaces and don't paste lines with tabs - they don't format properly - use spaces instead.

• 0 views • link
written by Neilfws Neilfws

1 answer

I believe the contrast would simply be C1.A - (C1.B + C2.A + C2.B)/3, assuming that your design matrix is model.matrix(~0+cell_type*treatment+donor, targets). If you are using a different design matrix the parametrization will be different. We can give you an answer if you show your design matrix.

• 1 views • link
written by Ryan Thompson Ryan Thompson

Log in to answer this question.

More posts like this