[BioC] Doubt and Problem(?) with analysis with marray tools

Gordon Smyth smyth at wehi.edu.au
Tue Nov 4 23:10:56 MET 2003

>I followed all of the steps to make a normalization using the tool
>I normalized the data and I saw the plotgraphs. All is good.
>However, I have a doubt and a problem.
>My genes are spoteds two times in the membrane (in duplicate). I generated
>the file of data with all of the duplicated genes, 7680 spots.
>When I executed the command maM(ka.susc.norm), I could verify that the genes
>continued duplicated. I ask: Should I must get the average among the two
>genes (duplicates) before normalizing them?


>  Or does the tool marray detect
>the duplicity of the genes and does it already execute the average between

No in both cases. There is no reason for "duplicate" spots to have any 
effect on the normalization process. It is correct to treat the spots 
individually and this is what has been done.

>  and, in this case, did I execute some wrong step during the process?
>In my membrane exist 530 empty spots (without DNA). When I accomplish the
>normalization, these spots receive the value "NA". That is a problem?
>Today, in the morning, I tried to transform my object marrayNorm
>(ka.susc.norm) in an object limma using the function "as.MAList."
>The error
> > MA <- as.MAList(ka.susc.norm)
>Error in if (length(x) && x != "") return(x) else return(NULL) :
>         missing value where TRUE/FALSE needed

You've found a bug in as.MAList(). (I was trying to be clever by treating 
"" as effectively as empty string even though it has positive length. This 
strategy fails when it receives an entirely missing slot.)

I have fixed the bug in limma 1.3.1.


>occurred which I believe to be due to these "NA" in all of the slides. I see
>the help for as.MAList (?as.MAList) previous, but I do not solve the
>How could I solve this problem and to cure this doubt?
>Thanks for any comments
>Marcelo Luiz de Laia, M.Sc.
>Dep. de Tecnologia, Lab. Bioquímica e de Biologia Molecular
>Universidade Estadual Paulista - UNESP
>Via de Acesso Prof. Paulo Donato Castelane, Km 05
>14.884-900 - Jaboticabal, SP, Brazil
>PhoneFax: 16 3209-2675/2676/2677 R. 202/208/203 (trab.)
>Phone res: 16 3203 2328 - www.lbm.fcav.unesp.br - 
><https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor>mlaia at yahoo.com

More information about the Bioconductor mailing list