[question] integration of microarray data from different sources in limma package
1
0
Entering edit mode
이선재 ▴ 30
@-4889
Last seen 10.2 years ago
Hi Johannes, I would like to explain my data in detail for your deep understanding. There are three replicate data from two-channel microarray Two of them were scanned by genepix scanner and the other was scanned by agilent scanner. All of them are originated from same microarray platform( agilent chip ). Just different scanner makes them provide different outputs, but its original information is same. However, when I want to use three microarray data for the analysis using limma package, it is difficult to use built-in functions in the package to normalize and analyze them for extracting DEGs(differentially expressed genes). For example, library(limma) setwd("~~") targets_agilent = readTargets("targets_agilent.txt") targets_genepix = readTargets("targets_genepix.txt") rg_agilent = read.maimages(targets_agilent$FileName, source="agilent") rg_genepix = read.maimages(targets_genepix$FileName, source="genepix") rg_all = cbind( rg_agilent, rg_genepix ) When I tried to use rg_all object for further analysis, it provokes some problems because cbind function just concetanates the data without consideration of interal information. Please let me know your precious opinion for this beginner in this field. Thanks in advance. Sincerely, Sunjae LEE 2011-10-01 ?? 2:29, Freudenberg, Johannes (NIH/NIEHS) [E] ? ?: > Hi Sunjae, > >> Is there any possibility to combine, normalize and analyze them without problems? > I think the order should be "normalize, combine, and analyze (hopefully) without problems." I don't really see a way (nor a good reason) for combining the different platforms first and then doing the pre-processing. > > Best regards, > --Johannes > > > > -----Original Message----- > From: ??? [mailto:sjlee at biosoft.kaist.ac.kr] > Sent: Tuesday, September 27, 2011 11:05 PM > To: bioconductor at stat.math.ethz.ch > Subject: [BioC] [question] integration of microarray data from different sources in limma package > > Dear developers, > > I want to ask some questions about microarray data pre-processing > > In the limma package, it supports reading microarray data from different source such as agilent(agilent feature extraction), genepix(genepix pro) format by fuction read.maimages(). > > And cbind fuction enables us to combine RGList obtained from micorarray data from different source > > However, when we try to normalize the combined RGList data through normalizeBetweenArrays function and analyze differentially expressed gene through lmFit, eBayes function, it produces some problems of not- compatible integrations due to diffrenence of its sources. ( I just want to integrate microarray data from agilent and genepix format as soon as possible ) > > Is there any possibility to combine, normalize and analyze them without problems? > > please let me know how to solve them. > > > > Sincerely, > > Sunjae LEE > > Ph.D. candidate, BISL, Dept. of Bio and Brain engineering, KAIST 335-Gwahangno, Yuseong-gu, Daejeon 305-701, Republic of Korea > > _______________________________________________ > 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 -- Sunjae LEE Ph.D. candidate, BISL, Dept. of Bio and Brain engineering, KAIST 335-Gwahangno, Yuseong-gu, Daejeon 305-701, Republic of Korea TEL +82-42-350-4353 (cell.) +82-10-3090-9050
Microarray limma BRAIN Microarray limma BRAIN • 1.3k views
ADD COMMENT
0
Entering edit mode
@andreas-heider-4538
Last seen 9.8 years ago
Dear Sunjae Lee, I think what you are after is some kind of batch effect removal between samples run on the same platform. I can confirm you, that I am about to submit a package to Bioconductor that can be used to solve this problem. The package makes use of empirical Bayes methods to minimize batch effects. This is documented in detail in he following publication in Biostatistics: "Adjusting batch effects in microarray expression data using empirical Bayes methods" As my package is not available yet, you might be able to use the scripts and functions provided in this publication. However, you can also wait for the submission the package. Hope I could help, Andreas Sunjae LEE: > Hi Johannes, > > I would like to explain my data in detail for your deep understanding. > There are three replicate data from two-channel microarray > > Two of them were scanned by genepix scanner and the other was scanned by > agilent scanner. > > All of them are originated from same microarray platform( agilent chip > ). Just different scanner makes them provide different outputs, but its > original information is same. > > However, when I want to use three microarray data for the analysis using > limma package, it is difficult to use built-in functions in the package > to normalize and analyze them for extracting DEGs(differentially > expressed genes). > > For example, > > library(limma) > setwd("~~") > targets_agilent = readTargets("targets_agilent.txt") > targets_genepix = readTargets("targets_genepix.txt") > rg_agilent = read.maimages(targets_agilent$FileName, source="agilent") > rg_genepix = read.maimages(targets_genepix$FileName, source="genepix") > rg_all = cbind( rg_agilent, rg_genepix ) > > When I tried to use rg_all object for further analysis, it provokes some > problems because cbind function just concetanates the data without > consideration of interal information. > > Please let me know your precious opinion for this beginner in this field. > > Thanks in advance. > > Sincerely, > > Sunjae LEE 2011-10-01 ?? 2:29, Freudenberg, Johannes (NIH/NIEHS) [E] ? ?: > Hi Sunjae, > >> Is there any possibility to combine, normalize and analyze them without problems? > I think the order should be "normalize, combine, and analyze (hopefully) without problems." I don't really see a way (nor a good reason) for combining the different platforms first and then doing the pre- processing. > > Best regards, > --Johannes > > > > -----Original Message----- > From: ??? [mailto:sjlee@biosoft.kaist.ac.kr] > Sent: Tuesday, September 27, 2011 11:05 PM > To: bioconductor@stat.math.ethz.ch > Subject: [BioC] [question] integration of microarray data from different sources in limma package > > Dear developers, > > I want to ask some questions about microarray data pre-processing > > In the limma package, it supports reading microarray data from different source such as agilent(agilent feature extraction), genepix(genepix pro) format by fuction read.maimages(). > > And cbind fuction enables us to combine RGList obtained from micorarray data from different source > > However, when we try to normalize the combined RGList data through normalizeBetweenArrays function and analyze differentially expressed gene through lmFit, eBayes function, it produces some problems of not- compatible integrations due to diffrenence of its sources. ( I just want to integrate microarray data from agilent and genepix format as soon as possible ) > > Is there any possibility to combine, normalize and analyze them without problems? > > please let me know how to solve them. > > > > Sincerely, > > Sunjae LEE > > Ph.D. candidate, BISL, Dept. of Bio and Brain engineering, KAIST 335-Gwahangno, Yuseong-gu, Daejeon 305-701, Republic of Korea > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- Sunjae LEE Ph.D. candidate, BISL, Dept. of Bio and Brain engineering, KAIST 335-Gwahangno, Yuseong-gu, Daejeon 305-701, Republic of Korea TEL +82-42-350-4353 (cell.) +82-10-3090-9050 ------------------------------ [[alternative HTML version deleted]]
ADD COMMENT

Login before adding your answer.

Traffic: 991 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6