Search
Question: Agilent data processing by Limma R package
0
2.5 years ago by
India
Agaz Hussain Wani260 wrote:

I have a few queries regarding Agilent data processing by Limma package.

1) I tried to test the use-case describe in Limma package vignette " Time Course E ects of Corn Oil on Rat Thymus with Agilent 4x44K Arrays". I downloaded the same data as named in the package. I tried the following code

SDRF <- read.delim("EGEOD-33005.sdrf.txt",check.names=FALSE,stringsAsFactors=FALSE)
y <- backgroundCorrect(x,method="normexp")
neg95 <- apply(y$E[y$genes$ControlType==-1,],2,function(x) quantile(x,p=0.95)) cutoff <- matrix(1.1*neg95,nrow(y),ncol(y),byrow=TRUE) isexpr <- rowSums(y$E > cutoff) >= 4
y0 <- y[y$genes$ControlType==0 & isexpr,]
Treatment <- SDRF[,"Characteristics[treatment]"]
levels <- c("10 ml/kg saline","2 ml/kg corn oil","5 ml/kg corn oil","10 ml/kg corn oil")
Treatment <- factor(Treatment,levels=levels)
design <- model.matrix(~Treatment)

and works fine.

But when i run the following line, gives an error.

fit <- lmFit(y0,design)

Error in array(x, c(length(x), 1L), if (!is.null(names(x))) list(names(x),  :
'data' must be of a vector type, was 'NULL'. 

And how to get the fold-change and p-value of the individual study in case of differential expression.

2) If i am not interested to perform the differential expression of the Agilent data, can i pass the individual text data file without targets(SDRF) file  and after preprocessing get the probe name, expression value and p-value of an individual study.

modified 22 months ago by Gordon Smyth33k • written 2.5 years ago by Agaz Hussain Wani260

apropos 1), you should do dim(y0) and dim(design), these may shed light.  you could also do debug(lmFit) prior to the lmFit call, and step through to see exactly how the error is triggered.

apropos 2), what is meant by "individual study"?

In the data said above, 19 samples are used. I mean to say that if i have one sample(GSM819076) and i want to get probe name , expression and p-value.

What p-value are you referring to? Obviously you can't get a p-value for differential expression if you only have one sample.

I am not talking about differential expression in case to get the p-value. If i have a Agilent raw data file and i want to pre-process the data to get probe name, expression value and p-value.

0
22 months ago by
Gordon Smyth33k
Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia
Gordon Smyth33k wrote:

The problem is that you haven't normalized the data, so that the data object y0 is not yet ready for lmFit().

Please follow the case study code in the limma User's Guide, which uses normalizeBetweenArrays().

Content
Help
Access

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