[R] (no subject)

[Rasmus Liland]

On 2020-12-11 20:14 +0500, Anas Jamshed wrote:
> On Fri, Dec 11, 2020 at 7:49 PM Rasmus Liland <jral using posteo.no> wrote:
> 
> > On 2020-12-11 19:16 +0500, Anas Jamshed wrote:
> > > On Fri, Dec 11, 2020 at 6:37 PM Rasmus Liland wrote:
> > > > On 2020-12-11 18:08 +0500, Anas Jamshed wrote:
> > > > >
> > > >
> > > > Anas Jamshed,
> > > >
> > > > I found this
> > > >
> > > > https://support.bioconductor.org/p/130817/
> > > >
> > > > maybe it helps ...
> > >
> > > still have the problem in 2nd error
> > >
> > > > > Also when i tried to:
> > > > >
> > > > > #Load the target files which the information about the sample and
> > > > > their corresponding group by
> > > > >
> > > > > targets<-read.delim(file="targets.txt", header=T)and create design
> > and
> > > > > fit the design by
> > > > > design <- model.matrix(~0+ conditions)
> > > > >
> > > > > It gives me the error :
> > > > >
> > > > > Error in model.frame.default(object, data, xlev = xlev) :
> > > > >   invalid type (closure) for variable 'conditions'
> >
> > Glad my suggestion helped.
> >
> > Do state how you solved that for someone
> > else to find it another time (maybe
> > yourself even ... ).
> >
> > One problem at a time ... pocito pocito
> > ...
> >
> > Read here or something
> >
> > https://stackoverflow.com/questions/33023508/why-am-i-getting-the-error-invalid-type-closure
> > ...
> >
> > > https://postimg.cc/1fKPj1xg
> >
> > Right, it says the object is not a
> > matrix ... there is a flag there called
> > «data,» perhaps look into specifying you
> > matrix there ...
> >
> > It would be more helpful for me as a
> > helper if you stated your problem in a
> > small example code snippet, instead of
> > just the error.  I might lack the
> > sufficient amount of teaching emphathy
> > there to se clearly through images and
> > error messages from a distance.  E.g.
> > use dput to paste some small dataset
> > here ...
> >
> > R
> 
> E-MTAB is an original sample data file and another one is normalized data
> file  but I don't know why I get just one gene(up reg) when I apply top
> table and decide test function
> 
> My R history file is :
> library(oligo)
> if (!requireNamespace("BiocManager", quietly = TRUE))
> install.packages("BiocManager")
> BiocManager::install("pd.hg.u133.plus.2")
> list.celfiles()
> setwd("C:/Users/USER/Desktop/RNA_Seq")
> list.celfiles()
> names = list.celfiles()
> array = read.celfiles(names)
> array
> eset = rma(array)
> write.exprs(eset, file = "data_normalized.txt") #this will be your
> normalized data by rma
> eset
> targets<-read.delim(file="targets.txt", header=T)
> targets<-read.delim(file="E-MTAB-5716.sdrf.txt", header=T)
> targets
> design <- model.matrix(~0+ conditions)
> fit <- lmFit(eset, design)
> fit <- lmFit(eset, targets)
> design <- model.matrix(~ description + 0, gset)
> design
> fit <- lmFit(eset, design)
> targets$Source.Name <-fl
> targets$Source.Name <-fl
> targets$Source.Name <-f1
> sml <- paste("G", sml, sep="")
> targets$Source.Name
> design <- model.matrix(~ description + 0, eset)
> design <- model.matrix(~ targets + 0, eset)
> design <- model.matrix(~ targets + 0, conditions())
> design <- model.matrix(~ targets + 0, conditions)
> design <- model.matrix(~0+ conditions)
> design <- model.matrix(~ description + 0 + conditions)
> design <- model.matrix(~ description + 0 , conditions)
> design <- model.matrix(~ description + 0, gset)
> design <- model.matrix(~ description + 0, eset)
> design <- model.matrix(~ targets + 0, eset)
> targets$Source.Name
> design <- model.matrix(~ Source.Name + 0, eset)
> design <- model.matrix(~ Source + 0, eset)
> gset
> gset$description
> eset <- eset[[idx]]
> eset
> design <- model.matrix(~ description + 0, eset)
> fvarLabels(eset) <- make.names(fvarLabels(eset))
> gsms <- paste0("000000000000000000000000000000XXXXXXXXXXXXXXX11111",
> "1111111111XXXXXXXXXXXXXXXXXXX")
> sml <- c()
> for (i in 1:nchar(gsms)) { sml[i] <- substr(gsms,i,i) }
> make.names()
> fvarLabels(eset) <- make.names(fvarLabels(eset))
> sel <- which(sml != "X")
> sml <- sml[sel]
> gset <- eset[ ,sel]
> eset
> design <- model.matrix(~0+ conditions)
> design <- model.matrix(~0+ eset)
> design
> fit <- lmFit(eset, design)
> fit
> contrast.matrix <- makeContrasts(group1=condition1-control,
> group2=condition2-control, levels = design)
> fit
> cont.matrix <- makeContrasts(G1-G0, levels=design)
> sml <- paste("G", sml, sep="")    # set group names
> fl <- as.factor(sml)
> sml
> cont.matrix <- makeContrasts(G1-G0, levels=design)
> design
> gset
> design
> cont.matrix <- makeContrasts(eset, levels=design)
> fit2 <- contrasts.fit(fit, cont.matrix)
> fit2
> fit3 <- eBayes(fit2, 0.01)
> fit3
> tT <- topTable(fit3, adjust="fdr", sort.by="B", number=1250)
> tT
> tT <- topTable(fit3, adjust="fdr", sort.by="B", number=500000)
> tT
> tT <- topTable(fit3, adjust="fdr", sort.by="B", number=500000,p=0.05)
> tT
> fit.cont <- contrasts.fit(fit, contrast.matrix)
> fit.cont <- contrasts.fit(fit, contrast.matrix)
> fit.cont <- contrasts.fit(fit2, contrast.matrix)
> fit.cont <- contrasts.fit(fit2, contrasts.fit())
> results<-decideTests(fit3,adjust.method="fdr",p=0.05)
> results
> summary(results)
> cont.matrix <- makeContrasts(eset, levels=design)
> fit.cont <- contrasts.fit(fit, cont.matrix)
> fit.cont
> fit.cont<- eBayes(fit.cont)
> fit.cont
> results<-decideTests(fit.cont,adjust.method="fdr",p=0.001)
> results
> summary(results)
> heatmap(results)
> heatmap(results[:,])
> heatmap(results[,])
> heatmap(results[,0])
> heatmap(results[1,4])
> heatmap(results[1,1])
> heatmap(results[2,2])
> heatmap(results[3,2])
> heatmap(results[,:])
> heatmap(results[:,])
> heatmap(results[1,])
> heatmap(results[1,:])

I think that's too many unspecific lines 
and too large files directly here on 
email (24MiB!).

Would you please narrow down your 
question.