[BioC] limma2annaffy error
James W. MacDonald
jmacdon at med.umich.edu
Thu Mar 17 17:26:39 CET 2011
Hi Mali,
Thanks for the bug report. Fixed now in svn, which will be propagated to
the devel version in 24 hours or so.
Best,
Jim
On 3/17/2011 5:11 AM, mali salmon wrote:
> Thanks a lot Jim, it was really helpful!!!
> I debugged the function and found the problem.
> My lmFit object, and hence the "tables" object contain gene symbols which I
> added to the lmFit object in some point. In the code when it reaches the
> line "aafTable(`Fold Change` = round(fld, 2))" it actually reads the symbols
> instead of the fold change.
> May I suggest to replace the line "fld<-tables[[i]][,2]" with
> "fld<-tables[[i]][,"logFC"], this can prevent the error in the future.
>
> I solved this problem and encountered another one, and after debugging I
> think I found a small bug in the code.
> The problem appears when I set in the command line "tstat = FALSE". The
> first hrml file for the first contrast is generated successfully, but when
> it comes to generate the "testtable" for the second contrast I get the error
>
>
> "Error in merge.aafTable(testtable, aafTable(`p-value` =
> round(tables[[i]][, :
> The tables must have the same number of rows"
>
> This is because when tstat=FALSE, the testtable is not reset and updated to
> the new comparison.
>
> >if (tstat) testtable<- aafTable(`t-statistic` = round(tables[[i]][, "t"],
> 2))
> >if (pval) {
> > if (!exists("testtable")) {
> > testtable<- aafTable(`p-value` = round(tables[[i]][,
> > "adj.P.Val"], 3))
> > }
> > else {
> > testtable<- merge(testtable, aafTable(`p-value` =
> round(tables[[i]][,
> > "adj.P.Val"], 3)))
> > }
>> }
>
> When I set "tstat=TRUE" everything is OK
>
> All the best
> Mali
>
>
>
>
>
>
>
> This
> All the best
> Mali
>
> On Wed, Mar 16, 2011 at 3:32 PM, James W. MacDonald
> <jmacdon at med.umich.edu>wrote:
>
>> Hi Mali,
>>
>>
>> On 3/16/2011 3:23 AM, mali salmon wrote:
>>
>>> Hi All
>>> I am analysing affymetrix data, and I am trying to use "limma2annaffy"
>>> function from "affycoretools" package to create html tables.
>>> When I run the following command:
>>>
>>> limma2annaffy(eset, fit2, design, cont.matrix,lib="hgu133a2" , adjust =
>>>>
>>> "fdr",
>>> anncols = aaf.handler()[c(1:3, 4:5, 11:12)], pfilt =
>>> 0.05,fldfilt=2,tstat = FALSE, pval = TRUE, FC = TRUE,
>>> expression = TRUE, html = TRUE, text = FALSE, save = FALSE)
>>> I get the error:
>>> "Error in round(fld, 2) : Non-numeric argument to mathematical function"
>>>
>>> In order to understand why I get this error, I have tried to generate the
>>> "tables" object from the limma2annaffy code.
>>> However, when I get to "tableFilt" function, R can't find it although the
>>> package is loaded.
>>>
>>
>> The easiest way to debug a function is to do
>>
>> debug(limma2annaffy)
>>
>> then run the function call. You can then step through the execution of the
>> function line by line, and look at the intermediate objects being made, such
>> as the tables list. Since you will be debugging from within the package
>> namespace, all functions will be available to you.
>>
>> Note however that the tableFilt() function simply takes the output from
>> topTable() and then filters based on the p-value and fold change you
>> supplied. The error indicates that you are getting non-numeric values in the
>> second column of your topTable() output, which is unexpected.
>>
>>
>>
>>> When I type ?tableFilt I get the documentation, but when I try to view the
>>> code I get
>>>
>>>> tableFilt
>>>>
>>> Error: object 'tableFilt' not found
>>>
>>
>> That's because this package has a namespace, and the tableFilt() function
>> isn't exported. You can access unexported functions using :::, e.g.,
>> affycoretools:::tableFilt.
>>
>> Best,
>>
>> Jim
>>
>>
>>
>>> Any ideas?
>>> Thanks
>>> Mali
>>>
>>>
>>> sessionInfo()
>>>>
>>> R version 2.12.1 (2010-12-16)
>>> Platform: i386-redhat-linux-gnu (32-bit)
>>>
>>> locale:
>>> [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
>>> [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
>>> [5] LC_MONETARY=C LC_MESSAGES=en_US.UTF-8
>>> [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
>>> [9] LC_ADDRESS=C LC_TELEPHONE=C
>>> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
>>>
>>> attached base packages:
>>> [1] stats graphics grDevices utils datasets methods base
>>>
>>> other attached packages:
>>> [1] hgu133a2.db_2.4.5 org.Hs.eg.db_2.4.6 annaffy_1.22.0
>>> [4] annotate_1.28.0 affycoretools_1.22.0 KEGG.db_2.4.5
>>> [7] GO.db_2.4.5 RSQLite_0.9-4 DBI_0.2-5
>>> [10] AnnotationDbi_1.12.0 simpleaffy_2.26.1 gcrma_2.22.0
>>> [13] genefilter_1.32.0 hgu133a2cdf_2.7.0 affy_1.28.0
>>> [16] Biobase_2.10.0 limma_3.6.9
>>>
>>> loaded via a namespace (and not attached):
>>> [1] affyio_1.18.0 biomaRt_2.6.0 Biostrings_2.18.2
>>> [4] Category_2.16.1 GOstats_2.16.0 graph_1.28.0
>>> [7] GSEABase_1.12.2 IRanges_1.8.9 preprocessCore_1.12.0
>>> [10] RBGL_1.26.0 RCurl_1.5-0 splines_2.12.1
>>> [13] survival_2.36-2 tcltk_2.12.1 tools_2.12.1
>>> [16] XML_3.2-0 xtable_1.5-6
>>>
>>> [[alternative HTML version deleted]]
>>>
>>> _______________________________________________
>>> 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
>>>
>>
>> --
>> James W. MacDonald, M.S.
>> Biostatistician
>> Douglas Lab
>> University of Michigan
>> Department of Human Genetics
>> 5912 Buhl
>> 1241 E. Catherine St.
>> Ann Arbor MI 48109-5618
>> 734-615-7826
>> **********************************************************
>> Electronic Mail is not secure, may not be read every day, and should not be
>> used for urgent or sensitive issues
>>
>
--
James W. MacDonald, M.S.
Biostatistician
Douglas Lab
University of Michigan
Department of Human Genetics
5912 Buhl
1241 E. Catherine St.
Ann Arbor MI 48109-5618
734-615-7826
**********************************************************
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues
More information about the Bioconductor
mailing list