[BioC] AgiMicroRna - FilterMicroRna question
Martin Morgan
mtmorgan at fhcrc.org
Tue Jun 1 19:31:56 CEST 2010
On 06/01/2010 06:43 AM, Neel Aluru wrote:
> Hello,
>
> I have asked this question before and haven't heard from anyone. Sorry for reposting it as I spent lot of time on it and still cannot figure it out. I need to filter the data before statistical analysis so as to remove the genes that are not detected.
>
>> ddPROC = filterMicroRna(ddTGS.rma, dd.micro, control = TRUE,
> IsGeneDetected = TRUE, wellaboveNEG = FALSE, limIsGeneDetected = 50,
> limNEG = 25, makePLOT = FALSE, targets.micro, verbose = TRUE)
> FILTERING PROBES BY FLAGS
>
>
> FILTERING BY ControlType
> Error in matrix(ddFILT$other$gIsGeneDetected, nrow = dim(ddFILT)[1],
> ncol = dim(ddFILT)[2]) :
> attempt to set an attribute on NULL
>
>
> I checked my data files to see if the required column (IsGeneDetected) is present and it is there. But, for some reason it is not detecting and I do not understand the error message I am getting. If anyone can explain the error message to me that would be great. I have posted the session info below.
Hi Neel -- I can't help with specifics, but
> matrix(NULL)
Error in matrix(NULL) : attempt to set an attribute on NULL
so the proximate cause of the error message is likely that
ddFILT$other$gIsGeneDetected is equal to NULL, e.g., because it doesn't
exist. You can investigate this by inspecting the code, e.g.,
> options(error=browser())
and then re-running your code. See ?browser; when done use
options(error=NULL). Before that I'd revisit the help page for this
function and double-check that you are providing appropriate arguments.
I've added
> packageDescription('AgiMicroRna')$Maintainer
[1] "Pedro Lopez-Romero <plopez at cnic.es>"
to the email, as Pedro in the best position to help you.
Martin
> Thank you very much,
>
> Neel
>
>
>
>
> Session Info
>
>> library("AgiMicroRna")
>> targets.micro=readTargets(infile="targets.txt", verbose=TRUE)
>
> Target File
> FileName Treatment GErep Subject
> 36_DMSO_1 36_DMSO_1.txt 36DMSO 1 1
> 36_DMSO_2 36_DMSO_2.txt 36DMSO 1 2
> 36_DMSO_3 36_DMSO_3.txt 36DMSO 1 3
> 36_TCDD_1 36_TCDD_1.txt 36TCDD 2 1
> 36_TCDD_2 36_TCDD_2.txt 36TCDD 2 2
> 36_TCDD_3 36_TCDD_3.txt 36TCDD 2 3
> 60_DMSO_1 60_DMSO_1.txt 60DMSO 3 1
> 60_DMSO_2 60_DMSO_2.txt 60DMSO 3 2
> 60_DMSO_3 60_DMSO_3.txt 60DMSO 3 3
> 60_TCDD_1 60_TCDD_1.txt 60TCDD 4 1
> 60_TCDD_2 60_TCDD_2.txt 60TCDD 4 2
> 60_TCDD_3 60_TCDD_3.txt 60TCDD 4 3
>
>> dd.micro=read.maimages(targets.micro$FileName,
> columns=list(R="gTotalGeneSignal",G=
> "gTotalProbeSignal",Rb="gMeanSignal", Gb="gProcessedSignal"),
> annotation=c("ProbeUID","ControlType","ProbeName","GeneName","SystematicName",
> "sequence", "accessions","probe_mappings",
> "gIsGeneDetected","gIsSaturated","gIsFeatNonUnifOL",
> "gIsFeatPopnOL","chr_coord","gBGMedianSignal","gBGUsed"))
> Read 36_DMSO_1.txt
> Read 36_DMSO_2.txt
> Read 36_DMSO_3.txt
> Read 36_TCDD_1.txt
> Read 36_TCDD_2.txt
> Read 36_TCDD_3.txt
> Read 60_DMSO_1.txt
> Read 60_DMSO_2.txt
> Read 60_DMSO_3.txt
> Read 60_TCDD_1.txt
> Read 60_TCDD_2.txt
> Read 60_TCDD_3.txt
>> cvArray(dd.micro, "MeanSignal", targets.micro, verbose=TRUE)
> Foreground: MeanSignal
>
> FILTERING BY ControlType FLAG
>
> RAW DATA: 5335
> PROBES without CONTROLS: 4620
> ----------------------------------
> (Non-CTRL) Unique Probe: 490
> (Non-CTRL) Unique Genes: 231
> ----------------------------------
> DISTRIBUTION OF REPLICATED NonControl Probes
> reps
> 5 6 7 10
> 20 18 36 416
> ------------------------------------------------------
> Replication at Probe level- MEDIAN CV
> 36_DMSO_1 36_DMSO_2 36_DMSO_3 36_TCDD_1 36_TCDD_2 36_TCDD_3 60_DMSO_1
> 60_DMSO_2 60_DMSO_3
> 0.078 0.081 0.091 0.081 0.077 0.067
> 0.076 0.066 0.103
> 60_TCDD_1 60_TCDD_2 60_TCDD_3
> 0.073 0.086 0.069
> ------------------------------------------------------
> DISTRIBUTION OF REPLICATED Noncontrol Genes
> reps
> 20
> 231
> ------------------------------------------------------
>> ddTGS.rma = rmaMicroRna(dd.micro, normalize=TRUE, background=FALSE)
> Calculating Expression
>> ddPROC = filterMicroRna(ddTGS.rma, dd.micro, control = TRUE,
> IsGeneDetected = TRUE, wellaboveNEG = FALSE, limIsGeneDetected = 50,
> limNEG = 25, makePLOT = FALSE, targets.micro, verbose = TRUE)
> FILTERING PROBES BY FLAGS
>
>
> FILTERING BY ControlType
> Error in matrix(ddFILT$other$gIsGeneDetected, nrow = dim(ddFILT)[1],
> ncol = dim(ddFILT)[2]) :
> attempt to set an attribute on NULL
>
>> MMM = ddTGS.rma$Rb
>> colnames(MMM) = colnames(dd.micro$Rb)
>> maintitle='TGS.rma'
>> colorfill='blue'
>> ddaux=ddTGS.rma
>> ddaux$G=MMM
>> mvaMicroRna(ddaux, maintitle, verbose=TRUE)
>
> ------------------------------------------------------
> mvaMicroRna info:
> FEATURES : 231
> POSITIVE CTRL: 12
> NEGATIVE CTRL: 7
> STRUCTURAL: 3
>> rm(ddaux)
>> RleMicroRna(MMM,"RLE TGS.rma", colorfill)
>> boxplotMicroRna(MMM, maintitle, colorfill)
>> plotDensityMicroRna(MMM, maintitle)
>> spottypes = readSpotTypes()
>> ddTGS.rma$genes$Status = controlStatus(spottypes, ddTGS.rma)
> Matching patterns for: ProbeName GeneName
> Found 231 gene
> Found 1 BLANK
> Found 1 Blank
> Found 0 blank
> Found 6 positive
> Found 0 negative
> Found 0 flag1
> Found 0 flag2
> Found 6 flag3
> Found 5 flag4
> Found 1 flag5
> Setting attributes: values
>> i = ddTGS.rma$genes$Status == "gene"
>> esetPROC = esetMicroRna(ddTGS.rma[i,], targets.micro,
> makePLOT=TRUE, verbose = TRUE)
> outPUT DATA: esetPROC
> Features Samples
> 231 12
>> design=model.matrix(~-1+treatment)
>> print(design)
> treatment36DMSO treatment36TCDD treatment60DMSO treatment60TCDD
> 1 1 0 0 0
> 2 1 0 0 0
> 3 1 0 0 0
> 4 0 1 0 0
> 5 0 1 0 0
> 6 0 1 0 0
> 7 0 0 1 0
> 8 0 0 1 0
> 9 0 0 1 0
> 10 0 0 0 1
> 11 0 0 0 1
> 12 0 0 0 1
> attr(,"assign")
> [1] 1 1 1 1
> attr(,"contrasts")
> attr(,"contrasts")$treatment
> [1] "contr.treatment"
>
>> fit=lmFit(esetPROC, design)
>> cont.matrix = makeContrasts(treatment36TCDDvstreatment36DMSO =
> treatment36TCDD-treatment36DMSO, treatment60TCDDvstreatment60DMSO =
> treatment60TCDD-treatment60DMSO,treatment60TCDDvstreatment36TCDD =
> treatment60TCDD-treatment36TCDD, treatment60DMSOvstreatment36DMSO =
> treatment60DMSO-treatment36DMSO, levels=design)
>> print(cont.matrix)
> Contrasts
> Levels treatment36TCDDvstreatment36DMSO
> treatment60TCDDvstreatment60DMSO
> treatment36DMSO -1
> 0
> treatment36TCDD 1
> 0
> treatment60DMSO 0
> -1
> treatment60TCDD 0
> 1
> Contrasts
> Levels treatment60TCDDvstreatment36TCDD
> treatment60DMSOvstreatment36DMSO
> treatment36DMSO 0
> -1
> treatment36TCDD -1
> 0
> treatment60DMSO 0
> 1
> treatment60TCDD 1
> 0
>> fit2 = contrasts.fit(fit,cont.matrix)
>> print(head(fit2$coeff))
> Contrasts
> treatment36TCDDvstreatment36DMSO treatment60TCDDvstreatment60DMSO
> dre-let-7a 0.038640984 0.013333873
> dre-let-7b 0.074038749 -0.031608286
> dre-let-7c 0.026244357 -0.005682488
> dre-let-7d 0.067340768 0.055567054
> dre-let-7e 0.004569306 0.136348664
> dre-let-7f 0.042880109 0.085568058
> Contrasts
> treatment60TCDDvstreatment36TCDD treatment60DMSOvstreatment36DMSO
> dre-let-7a 1.7358343 1.76114142
> dre-let-7b 0.1366920 0.24233899
> dre-let-7c 0.9920976 1.02402449
> dre-let-7d 0.8098432 0.82161694
> dre-let-7e 0.1186829 -0.01309647
> dre-let-7f 1.1245878 1.08189990
>> fit2 = eBayes(fit2)
>> fit2 = basicLimma(esetPROC, design, cont.matrix, verbose = TRUE)
> DATA
> Features Samples
> 231 12
>
>> DE = getDecideTests(fit2, DEmethod = "separate", MTestmethod =
> "BH", PVcut = 0.1, verbose = TRUE)
>
> ------------------------------------------------------
> Method for Selecting DEGs: separate
> Multiple Testing method: BH - pval 0.1
>
> treatment36TCDDvstreatment36DMSO treatment60TCDDvstreatment60DMSO
> UP 0 5
> DOWN 0 1
> treatment60TCDDvstreatment36TCDD treatment60DMSOvstreatment36DMSO
> UP 56 51
> DOWN 80 91
> ------------------------------------------------------
>> pvalHistogram(fit2, DE, PVcut = 0.1, DEmethod ="separate",
> MTestmethod="BH",cont.matrix, verbose= TRUE)
>> significantMicroRna(esetPROC, ddTGS.rma, targets.micro, fit2,
> cont.matrix, DE, DEmethod = "separate", MTestmethod= "BH", PVcut =
> 0.1, Mcut=0, verbose=TRUE)
> ------------------------------------------------------
> CONTRAST: 1 - treatment36TCDDvstreatment36DMSO
>
> Error in data.frame(PROBE_ID, as.character(GENE_ID),
> as.character(chr_coord), :
> arguments imply differing number of rows: 231, 0
>
>
>
>
> Neel Aluru
> Postdoctoral Scholar
> Biology Department
> Woods Hole Oceanographic Institution
> Woods Hole, MA 02543
> USA
> 508-289-3607
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
--
Martin Morgan
Computational Biology / Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.
PO Box 19024 Seattle, WA 98109
Location: Arnold Building M1 B861
Phone: (206) 667-2793
More information about the Bioconductor
mailing list