[BioC] [Rocky] - R code for PGSEA package to identify differentially expressed genes

[Valerie Obenchain]

Hi Rocky,

I'm pasting your message here so we can keep all comments on the mailing 
list.

## ---------------------------------------------------------------

Dear Valerie,

I would like to find down-regulated expression signatures for each 
sample from GSE11024 dataset containing 79 samples. How could I 
implement using PGSEA to get it. I would be glad and highly appreciate 
for your kindness.

http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE11024
GSM278765 - GSM278774 (10 samples) -> CC_KIDNEY
GSM278775 - GSM278780 (6 samples) -> CHR_KIDNEY
GSM278781 - GSM278792 (12 samples) -> NOR_KIDNEY
GSM278793 - GSM278799 (7 samples) -> ON_KIDNEY
GSM278800 - GSM278816 (17 samples) -> Pappilary_KIDNEY
GSM278817 - GSM278843 (27 samples) -> WM_KIDNEY

I also tried with affy and LIMMA package for normalizing the GSM CEL 
files using these code –

source("http://www.bioconductor.org/biocLite.R")
biocLite("limma")
biocLite("affy")
library(affy)
setwd("/home/haojamrocky/DATA/GSE11024")
rawdata<-ReadAffy()
eset <- expresso(rawdata, 
normalize.method="quantiles",bgcorrect.method="rma",pmcorrect.method="pmonly",summary.method="liwong")
write.exprs(eset, file="mydata.txt")

Warm regards,

Rocky

## ---------------------------------------------------------------


Have you tried following the PGSEA workflow vignette? If yes, at what 
point do you get an error?

The vignette says you need a gene expression dataset which I assume you 
downloaded from the ncbi site above. You also need a list of gene sets. 
Looking at the man page (?PGSEA) we see the PGSEA function requires the 
expression data to be a matrix, eset or ExpressionSet. Is your 
expression data in one of these forms? The gene list must be a 
GeneSetCollection or list of SMC objects. If you are having trouble 
creating your gene set see the vignettes in the GSEAbase, GO or GSEAlm 
packages.

Valerie




On 01/25/12 21:53, Valerie Obenchain wrote:
> On 01/25/12 21:12, Valerie Obenchain wrote:
>> Hi Rocky,
>>
>> gsub() "finds and replaces". In this example we replace the dot with 
>> an underscore. The double backslash in front of the dot is "escaping" 
>> the dot because it is a special character (metacharacter). See ?regex 
>> and ?gsub for details.
>>
>> gsub("\\.", "A", c("foo.bar", ".foo"))
> sorry, changed my example ... this should have been
>
> gsub("\\.", "_", c("foo.bar", ".foo"))
>>
>> In the line you have below there are two gsub() statements. We can 
>> rewrite the statement in pieces,
>>
>> p <- phenoData(gse[[1]])$"characteristics_ch1"
>> x <- gsub("subtype: ", "", p)
>> subtype <- gsub("\\.", "_", x)
>>
>> Take a look at the output of each of these steps to better understand 
>> what is going on.
>>
>> I don't understand the second part of your question, re: GSE11024. 
>> What have you tried? Are you getting an error?
>>
>>
>> Valerie
>>
>>
>>
>>
>>
>>
>> On 01/24/12 23:01, haojam wrote:
>>> Bioconductor Group,
>>>
>>>
>>>
>>> Hi,
>>>
>>> I have a special request regarding PGSEA Package to identify 
>>> downregulated
>>> genes. I am not clear about this line
>>> subtype<- gsub("\\.", "_",gsub("subtype: ", "",
>>> phenoData(gse[[1]])$"characteristics_ch1"))
>>>
>>> What does gsub ( ) mean for and \\ . I would like to look for GSE11024
>>> having 5 subtypes and 0ne normal group, how could I proceed.
>>>
>>>
>>>
>>> http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE11024
>>> GSM278765 - GSM278774 (10 samples) -> CC_KIDNEY
>>> GSM278775 - GSM278780 (6 samples) -> CHR_KIDNEY
>>> GSM278781 - GSM278792 (12 samples) -> NOR_KIDNEY
>>> GSM278793 - GSM278799 (7 samples) -> ON_KIDNEY
>>> GSM278800 - GSM278816 (17 samples) -> Pappilary_KIDNEY
>>> GSM278817 - GSM278843 (27 samples) -> WM_KIDNEY
>>>
>>>
>>>
>>> Warm regards,
>>>
>>> Rocky
>>>
>>> SNU College of Medicine
>>>
>>> Korea
>>>
>>>
>>> [[alternative HTML version deleted]]
>>>
>>> _______________________________________________
>>> 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
>>
>> _______________________________________________
>> 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
>
> _______________________________________________
> 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