[BioC] cellHTS2 variance by plate calculation?

[Joseph Barry]

Dear Stephen,

In the median absolute deviation calculation, R by default adjusts by constant scale factor of 1.4826 for asymptotically normal consistency (see ?mad for further details). Could you check to see if this explains the discrepancy in your excel result?

If not, perhaps it would be helpful if you could repeat your calculations on a specific dataset that we both have access to and share your results. The KcViabSmall dataset that is distributed with cellHTS2 is one option. In this way it will be easier for others to pinpoint the source of this discrepancy.

Best wishes,
Joseph

On Dec 8, 2011, at 2:10 AM, Stephen Baird [guest] wrote:

> 
> To understand the workings of cellHTS2 I ran a couple of analyses of one plate with either negative or median scaling normalization with and without variance adjustment. Without variance adjustment my excel normalization calculations are the same as cellHTS2. With variance adjustment my numbers are always different (whether by plate or experiment adjustment).  As I understand calculating the variance adjustment, I should divide each normalized well plate value with the median absolute deviation value calculated on only the "sample" wells' normalized values.  Has anyone else verified the cellHTS2 output calculations using one of the variance options? or have a suggestions in how I can prove my calculations are wrong?
> 
> Thanks,
> Stephen
> 
> -- output of sessionInfo(): 
> 
>> sessionInfo()
> R version 2.14.0 (2011-10-31)
> Platform: x86_64-pc-linux-gnu (64-bit)
> 
> locale:
> [1] LC_CTYPE=en_CA.UTF-8       LC_NUMERIC=C
> [3] LC_TIME=en_CA.UTF-8        LC_COLLATE=en_CA.UTF-8
> [5] LC_MONETARY=en_CA.UTF-8    LC_MESSAGES=en_CA.UTF-8
> [7] LC_PAPER=C                 LC_NAME=C
> [9] LC_ADDRESS=C               LC_TELEPHONE=C
> [11] LC_MEASUREMENT=en_CA.UTF-8 LC_IDENTIFICATION=C
> 
> attached base packages:
> [1] grid      stats     graphics  grDevices utils     datasets  methods
> [8] base
> 
> other attached packages:
> [1] KEGG.db_2.5.0        HTSanalyzeR_2.5.1    RankProd_2.24.0
> [4] cellHTS2.alex_2.16.0 BioNet_1.10.1        RBGL_1.28.0
> [7] GSEABase_1.14.0      graph_1.30.0         annotate_1.30.0
> [10] igraph_0.5.5-2       GO.db_2.5.0          org.Hs.eg.db_2.5.0
> [13] RSQLite_0.9-4        DBI_0.2-5            AnnotationDbi_1.14.1
> [16] cellHTS2_2.16.0      locfit_1.5-6         lattice_0.20-0
> [19] akima_0.5-4          hwriter_1.3          vsn_3.20.0
> [22] splots_1.18.0        genefilter_1.34.0    Biobase_2.12.2
> [25] RColorBrewer_1.0-5
> 
> loaded via a namespace (and not attached):
> [1] affy_1.30.0           affyio_1.20.0         biomaRt_2.8.1
> [4] Category_2.18.0       limma_3.8.3           MASS_7.3-14
> [7] prada_1.28.0          preprocessCore_1.14.0 RCurl_1.6-9
> [10] rrcov_1.3-01          splines_2.14.0        stats4_2.14.0
> [13] survival_2.36-10      tools_2.14.0          XML_3.4-2
> [16] xtable_1.5-6
>> 
> 
> 
> --
> Sent via the guest posting facility at bioconductor.org.
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor