gmane.science.biology.informatics.conductor

EdgeR: problem with large logFC value

2012-05-23T22:12:48

Dear all,

I am doing DGE multifactorial analysis with EdgeR but I have a problem
with somes locus I obtained a large logFC value , perhaps because any
comparision there is some zero.
Is this possible? or maybe I have some error

Thank for your help, waiting your response.
Regards,

This is the scrip that I am using:

d.all<- readDGE(counts.all, skip = 5, comment.char = "!")
colnames(d.all$counts)<-counts.all[,3]
cpmd.all<-cpm(d.all)
d.all <- d.all[rowSums(cpmd.all > 1) >= 2, ]
d.all<- calcNormFactors(d.all)
designall<-model.matrix(~time+time:d.all$samples$group)
d.all <- estimateGLMCommonDisp(d.all, designall)
glmfit.all <- glmFit(d.all, designall,dispersion = d.all$common.dispersion)
lrt.all <- glmLRT(d.all, glmfit.all, coef=4)
topTags(lrt.all,sort.by = "logFC")
logConc      logFC       LR      P.Value          FDR
AT1G22220 -12.30768 -144269488 62.70746 2.398056e-15 2.995171e-12
AT1G08430 -13.45102 -144269487 37.11160 1.115581e-09 2.960891e-07
AT4G11180 -13.84678  144269487 38.22337 6.309211e-10 1.83511

problem with large logFC value

2012-05-23T22:07:07

Dear all,

I am doing DGE multifactorial analysis with EdgeR but I have a problem
with somes locus I obtained a large logFC value , perhaps because any
comparision there is some zero.
Is this possible? or maybe I have some error

Thank for your help, waiting your response.
Regards,

This is the scrip that I am using:

d.all<- readDGE(counts.all, skip = 5, comment.char = "!")
colnames(d.all$counts)<-counts.all[,3]
cpmd.all<-cpm(d.all)
d.all <- d.all[rowSums(cpmd.all > 1) >= 2, ]
d.all<- calcNormFactors(d.all)
designall<-model.matrix(~time+time:d.all$samples$group)
d.all <- estimateGLMCommonDisp(d.all, designall)
glmfit.all <- glmFit(d.all, designall,dispersion = d.all$common.dispersion)
lrt.all <- glmLRT(d.all, glmfit.all, coef=4)
topTags(lrt.all,sort.by = "logFC")
logConc      logFC       LR      P.Value          FDR
AT1G22220 -12.30768 -144269488 62.70746 2.398056e-15 2.995171e-12
AT1G08430 -13.45102 -144269487 37.11160 1.115581e-09 2.960891e-07
AT4G11180 -13.84678  144269487 38.22337 6.309211e-10 1.835116

SNP detection of multiple samples

Wang, Li — 2012-05-23T21:29:54

Hi, all

I have transcriptomes of 31 samples. I have detected SNPs for each sample compared to Reference genome. 

My goal is to detect common SNPs among all the 31 samples based on 31 tables of SNPs (each sample vs reference genome).

I am writing to ask if there is a bioconductor package which can achieve my goal. 

Thanks!

Li
_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

methylumi on Infiniium27k idat files

chris briggs — 2012-05-23T20:20:59

I've been using methylumi to extract beta-vals, m-vals, etc. from my 24 idat
files and find many instances of beta values = "NA" even though the detection p
values are far less than 0.05. What causes a beta value of "NA"?

K_samps <-
read.table("KyleA_met27k_Samples_051812.txt",row.names=2,sep='\t',header=T)

K_idat<-methylumIDAT(barcodes = K_samps$barcode, pdat = K_samps)

K_bgcorr <- methylumi.bgcorr(K_idat) 

K_betaVals <-betas(K_bgcorr )

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

Genes to SNPs Tool

Ovokeraye Achinike-Oduaran — 2012-05-23T20:25:13

Hi all, 

Is there a tool in R that can give me a list of associated SNPs when I my input is a list of genes. I had a look at biomaRt manual...but I wasn't quite sure of a way forward.

Any suggestions will be greatly appreciated.

Thanks.

Avoks 

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

Re: Problem running HumMeth27QCReport after upgrading R to2.15

Djie Tjwan Thung — 2012-05-23T18:31:55

Hi Niles,

I stumbled upon somewhat the same problem as you, although I do not use
HumMeth27QCReport. Since Bioconductor 2.10, methylumi reads methylation
files differently:

For a methylation file it tries to guess what seperator delimits your file
by calling the function inside methylumiR:

.getFileSeparator(filename)

the function .getFileSeperator then tries to automatically detect which
seperator delimits the file by assuming your file has at least one column
with ProbeID in its name:
## Use "ProbeID" as an indicator of Where the metaData stops
##   intelligently find nMetaDataLines
nMetaDataLines <- grep('ProbeID', info, ignore.case=TRUE) - 1

The methylation demo data of HumMeth27QCReport lacks a ProbeID column and I
assume you also miss such a column in your methylation data. As several
functions of HumMeth27QCReport, act as wrapper functions around lumi and
methylumi functions, I guess this is why the problem occurs.

If you can export your methylation files again with ProbeID columns (I
believe the

Re: question about how analysis affymetrix huex st array

cstrato — 2012-05-23T17:45:05

Dear Jianhong,

You do not describe what you are comparing:

You say that you tried oligo, xps, aroma.affymetrix, and also 
APT(Affymetrix Power Tools). Then you say that only the results of MiDAS 
from APT and results of FIRMA from aroma.affymetrix have about 40% 
overlap. All the others have less than 10% overlap.
What do you mean with less than 10% overlap???
What did you compare???

Please note that MIDAS was developed by Affymetrix as alternative to 
ANOVA to measure differences between exon level signal...
The Affymetrix "exon_alt_transcript_analysis_whitepaper.pdf" says:
The basic idea is the following:
• We use PLIER to generate a robust estimate...

It seems that you are comparing apples and oranges, since you seem to 
compare MIDAS (based on PLIER) to RMA!?

If you want to compare the different packages you need to compare the 
results obtained from e.g. RMA and not from MIDAS, which most packages 
did not implement (although xps has implemented FIRMA).

With respect to APT vs xps please read the

Problem running HumMeth27QCReport after upgrading R to 2.15

Niles Oien [guest] — 2012-05-23T16:46:50


Hi,

I just upgraded R from 2.14 to 2.15 and I am finding that the HumMeth27QCReport package no longer runs, even on the demo data that come with the package.

What I do is :

library(HumMeth27QCReport)
Dir <- "/my/dir/Methylation/demo27"
Plat <- "Hum27"
ImportDataR <- ImportData(Dir)
ControlResults <- getAssayControls(ImportDataR,
                                   platform=Plat)

QCResults <- QCCheck(ImportDataR, pval=0.05)

normMvalues <- HumMeth27QCReport(ImportDataR,
         platform=Plat, pval=0.05, ChrX=FALSE, "euclidian")

but for the calls to QCCheck() and HumMeth27QCReport() I get the message :

Error in gregexpr(sep.i, dataLine1)[[1]] : subscript out of bounds

Any suggestions would be appreciated, I am appending my sessionInfo() from R.

Thanks,

Niles Oien, University of Colorado, Boulder, USA.



R version 2.15.0 (2012-03-30)
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] tcltk     gr

Re: File format for single channel analysis ofAgilentmicroarray data with Limma?

Hooiveld, Guido — 2012-05-23T15:52:47

Hi Parisa,

I also once struggled with reading in some Agilent singe channel arrays (that I downloaded from GEO; GSE27784), but for me these line of codes worked (in particularly note that the 2nd line is different than the one that is given on the website you linked to; specifically the statement source="agilent.median"):

HTH,
Guido

Read GSM686624_251486829200_S01_GE1_105_Jan09_1_1.txt 
Read GSM686625_251486829201_S01_GE1_105_Jan09_1_2.txt 
Read GSM686626_251486829328_S01_GE1_105_Jan09_1_3.txt 
Read GSM686627_251486829200_S01_GE1_105_Jan09_1_2.txt 
Read GSM686628_251486829200_S01_GE1_105_Jan09_1_4.txt 
Read GSM686629_251486829201_S01_GE1_105_Jan09_1_4.txt 
Read GSM686630_251486829328_S01_GE1_105_Jan09_1_4.txt 
Read GSM686631_251486829328_S01_GE1_105_Jan09_1_1.txt 
Read GSM686632_251486829328_S01_GE1_105_Jan09_1_2.txt 
Read GSM686633_251486829200_S01_GE1_105_Jan09_1_3.txt 
Read GSM686634_251486829201_S01_GE1_105_Jan09_1_3.txt 
Read GSM686635_251486829201_S01_GE1_105_Jan09_1_1.txt 
Array 1 corrected
Array 2

Re: limma all adj.P.Val the same

Sam McInturf — 2012-05-23T14:01:18

File format for single channel analysis of Agilentmicroarray data with Limma?

Parisa [guest] — 2012-05-23T13:50:40

Hi,

I am following the protocol outlined here for analysis of single channel Agilent microarray data: 

http://matticklab.com/index.php?title=Single_channel_analysis_of_Agilent_microarray_data_with_Limma

I keep getting the following error message when using Limma's read.maimages function to load my data into an RGList object: 

Error in RG[[a]][, i] <- obj[, columns[[a]]] :
  number of items to replace is not a multiple of replacement length 

I think this may be due to my Agilent raw data txt files being in the wrong format. I am having difficulty finding an example Agilent feature extraction raw data txt file online to compare it to. A link to a screen shot of one of the files I am using is below. I would appreciate if someone could let me know if it is in the correct format, and if not then what format it should be in to prevent the above error message from coming up. 

Thank you,

Parisa 

http://www4.picturepush.com/photo/a/8322602/img/8322602.png

 -- output of sessionInfo(): 

Platform: x86_64-appl

Re: design and contrast matrix for limma time series withoutreplicates

shao chunxuan — 2012-05-23T09:15:41

Thanks very much for the thorough explanation, I will add tm:trt1  to my
model to see expression changes in two condition.
Best wishes,

On Tue, Mar 20, 2012 at 7:51 PM, James W. MacDonald <jmacdon-lfcS8c3Mqgg< at >public.gmane.org> wrote:

Create A Stranded probeAnno Object

Dario Strbenac — 2012-05-23T09:00:10

Hello,

I've read the help page and the vignette and I want to have the strand information of the probes used, but I can't see what the required format of the names of the vectors is. Can anyone provide a small example of how it's done ? It would be nice if there was a simple way to do this based on a data.frame or GRanges object of probe information, rather than having having to create lots of specially named vectors and mess around with environments. Also, the chr argument to segChrom has to be an integer vector, which means X and Y for human have to be artificially renamed as 23 and 24. The design could be more streamlined.

- Dario.

--------------------------------------
Dario Strbenac
Research Assistant
Cancer Epigenetics
Garvan Institute of Medical Research
Darlinghurst NSW 2010
Australia

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gman

cell cycle modeling

Anja Mirenska — 2012-05-23T08:20:33

Dear BioC users

I wonder if there is a way in R to analyse flow cytometry cell cycle data
using the Dean/Jet/Fox or the Watson Pragmatic or another similar
algorithm, similar to what FlowJo does. Are there any packages available
for this purpose?

Kind regards

Anja

[[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

Re: non-integer counts for edgeR

Gordon K Smyth — 2012-05-23T07:33:08

Hi Tim,

We have a few RNA-seq related papers in the pipeline, of which the voom 
paper is one.  We're investing a lot of time into trying to push edgeR and 
voom both along a similar pathway, to get the best use out of both.

Everyone in my lab would be happier if I didn't answer Bioconductor 
questions, because I'd get to their draft papers more quickly, and the 
voom one in particular ...

Regards
Gordon

---------------------------------------------
Professor Gordon K Smyth,
Bioinformatics Division,
Walter and Eliza Hall Institute of Medical Research,
1G Royal Parade, Parkville, Vic 3052, Australia.
smyth-QpO9qclcJEu6c6uEtOJ/EA< at >public.gmane.org
http://www.wehi.edu.au
http://www.statsci.org/smyth

On Tue, 22 May 2012, Tim Triche, Jr. wrote:


______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane

Re: PANTHER in R.... Gene lists in sigPathway?

Ovokeraye Achinike-Oduaran — 2012-05-23T07:09:35

Thanks a bunch, Felipe.

Regards,

Avoks

On Tue, May 22, 2012 at 6:45 PM, Felipe Riveroll Aguirre
<friveroll-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

Re: non-integer counts for edgeR

Tim Triche, Jr. — 2012-05-23T06:28:19

Hi Dr. Smyth,

Will you be writing / have you already written a paper summarizing what
voom() does and how?

My understanding of voom() comes from
http://statsandgenomes.wordpress.com/2011/11/23/bioinformatics-seminar-professor-gordon-smyth-variance-models-for-rna-seq/wherein
it appears to be the current best general-purpose tool for treating
RNAseq data as if it were microarray data.  This would seem to bring up
issues with rare transcripts and the estimation of their abundance, which
is why I ask about a writeup.

Thanks as always for your patient explanations and high standards of work.

--t



On Tue, May 22, 2012 at 6:42 PM, Gordon K Smyth <smyth-QpO9qclcJEu6c6uEtOJ/EA< at >public.gmane.org> wrote:


_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

Re: edgeR, comparing models

Gordon K Smyth — 2012-05-23T05:23:49

Dear Alpesh,

I've commited a change to edgeR today, so that you can now produce the 
qq-plot of the goodness of fit statistics by

   gof(fit, plot=TRUE)

Best wishes
Gordon

---------------------------------------------
Professor Gordon K Smyth,
Bioinformatics Division,
Walter and Eliza Hall Institute of Medical Research,
1G Royal Parade, Parkville, Vic 3052, Australia.
smyth-QpO9qclcJEu6c6uEtOJ/EA< at >public.gmane.org
http://www.wehi.edu.au
http://www.statsci.org/smyth

On Wed, 23 May 2012, Gordon K Smyth wrote:


______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

How to make Illumina cluster files from HapMap

JiangZhengyu — 2012-05-23T03:45:17




Hi, Does anyone know how to make the illumina cluster files (AA LogR and AA theta, BB logR and BB theta,...) from HapMap samples? Is there any R package in the bioconductor to do this kind of job? Thanks,Best,Zhengyu       
[[alternative HTML version deleted]]

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

non-integer counts for edgeR

Gordon K Smyth — 2012-05-23T01:42:21

Dear Mete,

No, you cannot use non-integer counts with edgeR.

If you must use non-integer counts, please use the voom() function in the 
limma package instead.  This will do an analysis that is not too different 
from edgeR, just a little less powerful, and is not bothered by 
non-integer values.

Best wishes
Gordon


______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

edgeR, comparing models

Gordon K Smyth — 2012-05-23T01:35:45

Dear Alpesh,

What you're doing looks correct.

However, if you are doing this analysis (Figure 2 from 
http://nar.oxfordjournals.org/content/40/10/4288 ) to decide whether you 
need to use tagwise dispersion for your data, note that we have never 
observed a real data set for which tagwise estimation is not required. 
Hence it seems unnecessary to make these plots as a routine diagnostic.

Yes, dispersion=0 is Poisson.

To plot highlighted blue points, see ?points as well as

   xy <- qqnorm(pcom)

etc.

Best wishes
Gordon


______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}

_______________________________________________
Bioconductor mailing list
Bioconductor-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
https://stat.ethz.ch/mailman/listinfo/bioconductor
Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor