Best package or code to filter Affymetrix probes by present calls??
1
0
Entering edit mode
@garcia-orellanamiriam-5283
Last seen 7.1 years ago
Normalization gcrma Normalization gcrma • 1.9k views
0
Entering edit mode
@iain-gallagher-2532
Last seen 6.2 years ago
United Kingdom
0
Entering edit mode
0
Entering edit mode
Okko: Thanks for the code. I tried to get the INI genes with GCRMA normalized data as the paper and also the user guide indicate as an appropriate one to use (however I was not able to find a model code provided to use with gcrma). I set the codes as next and was able to make it work with qfarms but not with GCRMA, which is the normalization method my lab group have decided to use. Any idea what I am doing wrong ? Thanks > esetfarms <- qFarms(rawData) background correction: none normalization: quantiles PM/MM correction : pmonly expression values: farms background correcting...done. normalizing...done. 24128 ids to be processed | | |####################| > INIs <- INIcalls(esetfarms) # apply I/NI calls > summary(INIs) Summary Informative probe sets : 33.48% Non-Informative probe sets : 66.52% esetgcrma <- gcrma(rawData) Adjusting for optical effect..................Done. Computing affinities.Done. Adjusting for non-specific binding..................Done. Normalizing Calculating Expression > INIs <- INIcalls(esetgcrma) # apply I/NI calls Error in apply(object@assayData$se.exprs, 1, min) : dim(X) must have a positive length ******************************** Miriam Garcia, MS PhD candidate Department of Animal Sciences University of Florida ________________________________ From: Djork-Arné Clevert [okko@clevert.de] Sent: Tuesday, June 19, 2012 5:17 AM To: Iain Gallagher Cc: Garcia Orellana,Miriam; bioconductor@r-project.org Subject: Re: [BioC] Best package or code to filter Affymetrix probes by present calls?? Hi Miriam, I suggest you to filter the probe-sets according to their internal consistency -> only rely on probe-sets, where each probe of the set tells you the same story. For further details please read our informative/ non-informative call (I/NI call) paper in Bioinformatics (http://bioinformatics.oxfordjournals.org/content/23/21/2897.full) The summarization using the farms-algorithm and filtering with I/NI calls are just four lines of code: library(farms) data <- ReadAffy() eset <- qFarms(data) inis <- INIcalls(eset) Cheers, Okko Am 18.06.2012 um 23:47 schrieb Iain Gallagher: library(affy) dataIn <- ReadAffy() # read in the data rmaData <- rma(dataIn) # normalise and summarise, could also use gcrma etc etc calls <- mas5calls(data) # get PMA calls calls <- exprs(calls) absent <- rowSums(calls == 'A') # how may samples are each gene 'absent' in absent <- which (absent == ncol(calls)) # which genes are 'absent' in all samples rmaFiltered <- rmaData[-absent,] # filters out the genes 'absent' in all samples Best Iain ________________________________ From: "Garcia Orellana,Miriam" <mgarciao@ufl.edu<mailto:mgarciao@ufl.edu>> To: "bioconductor@r-project.org<mailto:bioconductor@r-project.org>" <bioconductor@r-project.org<mailto:bioconductor@r-project.org>> Sent: Monday, 18 June 2012, 17:46 Subject: [BioC] Best package or code to filter Affymetrix probes by present calls?? Dear R users: First thank to all users for their direct or indirect support with previous question. Now. I am rephrasing this question since I did not get any help the last 3 days. I am having hard time to analyze my microarray data, since the use of R environment is a new world for me. I have 18 affymetrix bovine arrays from liver samples of 30d old calves that born from cows fed 3 types of prepartum dam diets (factor DD, 6 arrays per DD) and were fed just milk replacer the first 30d of life ( factor MR, 9 array per MR). Biologically I will expect that the main factor driving any difference will be the MR rather than the DD (unless some imprinting genes are expressed). So I have the idea to filter non expressed genes using the simpleaffy package using the manual but I don't know what is wrong when I try to load the covdesc file I got error. I have a folder in my directory that contains all 18 CEL files and also the covdesc (extension .prn - is this the right one?). Since I was able to run the gcrma normalization so the working directory maybe well set, what I got is the next when using the option read.affy to read the covdesc file. raw.data <- ReadAffy() gcrma.eset <- call.exprs(raw.data, "gcrma") Loading required package: AnnotationDbi Adjusting for optical effect..................Done. Computing affinities.Done. Adjusting for non-specific binding..................Done. Normalizing Calculating Expression raw.data <- read.affy() ##read data in working directory Error in file(file, "rt") : cannot open the connection In addition: Warning message: In file(file, "rt") : cannot open file './covdesc': No such file or directory raw.data<- read.affy("covdesc") Error in file(file, "rt") : cannot open the connection In addition: Warning message: In file(file, "rt") : cannot open file './covdesc': No such file or directory I would really appreciate if you can suggest me any simple method to filter my genes ( I want to keep probes that are present in at least 4 of the 9 arrays in at least one of the MR groups, or do you think I should consider the interaction prepartum diet * milk replacer (then 3 arrays per interaction group and try to have at least 2 present genes in at least 1 of the 6 interactions) Thanks in advance for any help. Miriam [[alternative HTML version deleted]] _______________________________________________ Bioconductor mailing list Bioconductor@r-project.org<mailto:bioconductor@r-project.org> https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor [[alternative HTML version deleted]] _______________________________________________ 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 [[alternative HTML version deleted]] ADD REPLY 0 Entering edit mode Hi Miriam, the following code will extract the informative probe sets from the GCRMA ExpressionSet: esetfarms <- qFarms(rawData) INIs <- INIcalls(esetfarms) I.Probes <- getI_ProbeSets(INIS) esetgcrma <- gcrma(rawData) I.esetgcrma <- esetgcrma[I.Probes] Cheers, Okko Am 19.06.2012 um 23:55 schrieb Garcia Orellana,Miriam: > Okko: > Thanks for the code. I tried to get the INI genes with GCRMA normalized data as the paper and also the user guide indicate as an appropriate one to use (however I was not able to find a model code provided to use with gcrma). I set the codes as next and was able to make it work with qfarms but not with GCRMA, which is the normalization method my lab group have decided to use. Any idea what I am doing wrong ? > Thanks > > > > esetfarms <- qFarms(rawData) > background correction: none > normalization: quantiles > PM/MM correction : pmonly > expression values: farms > background correcting...done. > normalizing...done. > 24128 ids to be processed > | | > |####################| > > INIs <- INIcalls(esetfarms) # apply I/NI calls > > summary(INIs) > Summary > Informative probe sets : 33.48% > Non-Informative probe sets : 66.52% > > > > esetgcrma <- gcrma(rawData) > Adjusting for optical effect..................Done. > Computing affinities.Done. > Adjusting for non-specific binding..................Done. > Normalizing > Calculating Expression > > INIs <- INIcalls(esetgcrma) # apply I/NI calls > Error in apply(object@assayData$se.exprs, 1, min) : > dim(X) must have a positive length > > > ******************************** > > Miriam Garcia, MS > PhD candidate > Department of Animal Sciences > University of Florida > From: Djork-Arné Clevert [okko@clevert.de] > Sent: Tuesday, June 19, 2012 5:17 AM > To: Iain Gallagher > Cc: Garcia Orellana,Miriam; bioconductor@r-project.org > Subject: Re: [BioC] Best package or code to filter Affymetrix probes by present calls?? > > Hi Miriam, > > I suggest you to filter the probe-sets according to their internal consistency -> only rely on probe-sets, where each probe of the set tells you the same story. > For further details please read our informative/ non-informative call (I/NI call) paper in Bioinformatics (http://bioinformatics.oxfordjournals.org/content/23/21/2897.full) > > The summarization using the farms-algorithm and filtering with I/NI calls are just four lines of code: > > library(farms) > data <- ReadAffy() > eset <- qFarms(data) > inis <- INIcalls(eset) > > > Cheers, > Okko > > > > Am 18.06.2012 um 23:47 schrieb Iain Gallagher: > >> library(affy) >> >> dataIn <- ReadAffy() # read in the data >> rmaData <- rma(dataIn) # normalise and summarise, could also use gcrma etc etc >> calls <- mas5calls(data) # get PMA calls >> calls <- exprs(calls) >> absent <- rowSums(calls == 'A') # how may samples are each gene 'absent' in >> absent <- which (absent == ncol(calls)) # which genes are 'absent' in all samples >> rmaFiltered <- rmaData[-absent,] # filters out the genes 'absent' in all samples >> >> Best >> >> Iain >> >> >> >> >> ________________________________ >> From: "Garcia Orellana,Miriam" <mgarciao@ufl.edu> >> To: "bioconductor@r-project.org" <bioconductor@r-project.org> >> Sent: Monday, 18 June 2012, 17:46 >> Subject: [BioC] Best package or code to filter Affymetrix probes by present calls?? >> >> Dear R users: >> First thank to all users for their direct or indirect support with previous question. Now. I am rephrasing this question since I did not get any help the last 3 days. I am having hard time to analyze my microarray data, since the use of R environment is a new world for me. >> I have 18 affymetrix bovine arrays from liver samples of 30d old calves that born from cows fed 3 types of prepartum dam diets (factor DD, 6 arrays per DD) and were fed just milk replacer the first 30d of life ( factor MR, 9 array per MR). Biologically I will expect that the main factor driving any difference will be the MR rather than the DD (unless some imprinting genes are expressed). >> So I have the idea to filter non expressed genes using the simpleaffy package using the manual but I don't know what is wrong when I try to load the covdesc file I got error. >> I have a folder in my directory that contains all 18 CEL files and also the covdesc (extension .prn - is this the right one?). Since I was able to run the gcrma normalization so the working directory maybe well set, what I got is the next when using the option read.affy to read the covdesc file. >> >>> raw.data <- ReadAffy() >>> gcrma.eset <- call.exprs(raw.data, "gcrma") >> Loading required package: AnnotationDbi >> >> Adjusting for optical effect..................Done. >> Computing affinities.Done. >> Adjusting for non-specific binding..................Done. >> Normalizing >> Calculating Expression >>> raw.data <- read.affy() ##read data in working directory >> Error in file(file, "rt") : cannot open the connection >> In addition: Warning message: >> In file(file, "rt") : >> cannot open file './covdesc': No such file or directory >>> raw.data<- read.affy("covdesc") >> Error in file(file, "rt") : cannot open the connection >> In addition: Warning message: >> In file(file, "rt") : >> cannot open file './covdesc': No such file or directory >> >> I would really appreciate if you can suggest me any simple method to filter my genes ( I want to keep probes that are present in at least 4 of the 9 arrays in at least one of the MR groups, or do you think I should consider the interaction prepartum diet * milk replacer (then 3 arrays per interaction group and try to have at least 2 present genes in at least 1 of the 6 interactions) >> Thanks in advance for any help. >> Miriam >> >> >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> 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 >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> 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 > > [[alternative HTML version deleted]]
0
Entering edit mode