CSC8309 -- Gene Expression and Proteomics

For this section you need two files, firstly SpeciesAffyID.txt and secondly SoybeanCutObjects.RData. Download these first and put then in your working directory.

Now, evalutate the following R.

##
## We are trying to do some subsetting because not all of the probes on the
## chip are from soy


## read in another data frame called Species.Affy.ID.
## this links species names to affy ids.
Species.Affy.ID <- read.table('SpeciesAffyID.txt', header = T, sep = "")
dim(Species.Affy.ID)



load( 'SoybeanCutObjects.RData' )

tv.for.glycine.max <- Species.Affy.ID$species == 'Glycine max'
table( tv.for.glycine.max )
listOutProbeSets <- Species.Affy.ID$affyID[ tv.for.glycine.max==FALSE ]

length( listOutProbeSets )
is.factor( listOutProbeSets )

## Create a character vector for listOutProbeSets
## One way: rename listOutProbeSets as a character vector
listOutProbeSets <- as.character(listOutProbeSets)

## Confirm that listOutProbeSets is a character vector
is.character(listOutProbeSets)

## check object
soy.ab


## this is the bit which actually removes the stuff we are not intereste
RemoveProbes(listOutProbes=NULL, listOutProbeSets, cdfpackagename, probepackagename)

## Check that the object has less IDs now. There should be 37444.
soy.ab

Evaluate the following R.

# Start preparation for phenoData slot in AffyBatch object
pd <- data.frame(population = c(1,1,1,2,2,2), replicate = c(1,2,3,1,2,3))

# Display contents of pd
pd

# Assign the sampleNames(soy.ab) to the rownames of pd
rownames(pd) <- sampleNames(soy.ab)

# Display contents of pd again, notice change in rownames
pd

## Continue preparation for phenoData slot
metaData <- data.frame(labelDescription = c( 'population', 'replicate' ))

## Establish new phenoData slot
phenoData(soy.ab) <- new( 'AnnotatedDataFrame', data = pd, varMetadata = metaData)

## Display pData(soy.ab)
pData(soy.ab)

## Display phenoData(soy.ab)
phenoData(soy.ab)

palette.gray <- c(rep(gray(0:10/10), times = seq(1,41, by = 4)))

library('RColorBrewer')
brewer.cols <- brewer.pal(6, 'Set1')