READ ME file 
Monday 15th May 2006
********************************************************************************

CDF files:
- pfsangerb520296.cdf contains all the probes present on the PFSANGER array
- pfsanger.Thu_May_4_2006.allCdsProbes.cdf is a custom cdf file and contains all the unique probes mapping a given gene (exons + introns + probes overlapping annotated boundaries)
- pfsanger.Thu_May_4_2006.exonsets.cdf is a custom cdf file and contains all the unique probes mapping exon(s) only for a given gene
- pfsanger.Thu_May_4_2006.intronsets.cdf is a custom cdf file and contains all the unique probes mapping intron(s) only for a given gene


The custom CDF files have been generated using the exonerate software that perfect matches the 25bp-probes on the Plasmodium falciparum genome version 2.1 (http://www.genedb.org/genedb/malaria/).
During the process, all individual probes are annotated according to the latest available annotations and non-unique probes were removed from the set.
Probes mapping to exons/introns/gene were subsequently grouped into a probe set.
I suggest you open-up one of those CDF custom file using a text editor (but not on windows) to see the structure and probes organisation. 

To see how many probes are within a probe set, you need to open-up a terminal window and do:

$ grep -c 'probeset' pfsanger.Thu_May_4_2006.xxxxxx.cdf 

(where probeset = MALxxP1.xxx / PFxxxxxc/w / PFxx_xxxx)

Example:
$ grep -c 'MAL13P1.113' pfsanger.Thu_May_4_2006.exonsets.cdf
$35


This command line will return the number of times the 'probeset' is found in the CDF file. Because of the structure of this CDF file, this probe number will be -1 the number returned.
In the above example, there are 34 probes constituting the exon(s) probe set of MAL13P1.113.

Using R/Bioconductor version 2.3 you need to generate a custom cdf package:
#from the directory where your downloaded CDF will be, open up an R session

R> library(makecdfenv) 					#load the package 'makecdfenv'
R> make.cdf.package("pfsanger.ThuMay42006.xxxxxx.cdf", species="Plasmodium_falciparum")
								#this will generate a subdirectory pfsanger.ThuMay42006.xxxxxxcdf in your working directory, which contains the package. Please refer to the help page for more information.

Now open a different terminal and install the cdf package using:

$ R CMD INSTALL pfsanger.ThuMay42006.xxxxxxcdf 	#that will go to the default library of Bioconductor


When starting an analysis please do:
R> library(affy)
R> my.data <- ReadAffy()
R> my.data@cdfName <- "pfsanger.ThuMay42006.xxxxxxcdf"

Then proceed to further pre-processing and/or analysis as required.



The pfsangerb520296.cdf needs to be installed in R the same way. However you may have some difficulties installing it as it requires a lot of memory (a Linux 64bit machine is necessary).
As an alternative, you can use RMAExpress (http://rmaexpress.bmbolstad.com/) to pre-process and calculate the signal intensities, then either transfer back into R/Bioconductor or work on the table of matrices returned.




For further enquiries please refer to:
Celine Carret: ckc@sanger.ac.uk
Matloob Qureshi: mq2@sanger.ac.uk
Al Ivens: alicat@sanger.ac.uk