Hi and the "combineAffyBatch" function in the matchprobes package might be somewhat helpful (see package vignette), but as Kasper says, this is not something you should do before getting an understanding of how AffyBatch represents Affymetrix data, and of how it uses cdf and probe packages to that end (see vignette "Primer" of the affy package and in particular Section 7 for a start). Best wishes Wolfgang ---------------------------------------------------- Wolfgang Huber, EMBL-EBI, http://www.ebi.ac.uk/huber Kasper Daniel Hansen ha scritto: > Please re-read Laurents reply. > > The problem is that even thought he same probe(set) may be present on > the two arrays, the (x,y) location (spatial) of the probe is most likely > not the same. So you will need to remap everything in a spatial way. > > Yes, that can be done using R and following some of the things in the > post Jenny linked to + some custom code to take care of the spatial > re-organization. But it will require that you familiarize yourself with > the underlying data structures and it won't be a 1-2 hour exercise > (unless you really know your way around these data structures). > > Kasper > > On Dec 11, 2008, at 14:01 , shirley zhang wrote: > >> Hi Larurent, >> >> Thanks for your quick reply. You are right, CDF file also includes >> the location of probes on the chip. In that case, can I modify the >> HuEx .cdf file to mask those probes not found in Hugene array as >> NA/missing, and then use this modified .cdf file to analyze exon >> array? If can, do you know how to do it in R or other tools? >> >> Thanks a lot, >> Xiaoling >> >> >> >> >> On Thu, Dec 11, 2008 at 10:40 PM, Laurent Gautier <laurent at="" cbs.dtu.dk=""> >> wrote: >>> shirley zhang wrote: >>>> >>>> Dear List, >>>> >>>> I have a quick question. Since the HuGene probes are largely a subset >>>> of the HuEx probes, we would like to use that subset of probes for the >>>> HuEx array, then compare the results with that of using HuEx's own >>>> core probes which we already have. For this purpose, can I use >>>> HuGen's .cdf file to analyze HuEx arrays? >>> >>> I'd say that you mostly likely can't. >>> >>> The CDF definition contains information relative to the relative spatial >>> location of the features (probes) on the chip, and unless the HuGene >>> is a >>> subset also in the spatial sense of it (that is "like HuEx but with >>> probes >>> missing from their location on the array"). >>> >>>> If not, how to >>>> generate/modify a .cdf file based on HuGen probes to analyze HuEx >>>> array data? >>> >>> HuGen(e{0,1}), HuEx, or else, you should have the corresponding .cdf >>> distributed with the arrays (and you'll contact Affymetrix if this is >>> not >>> the case). >>> >>> >>> L. >>> >>>> Thanks, >>>> Shirley >>>>

Hi Mark, Thanks for you suggestion. Actually I just got the mapping file from Affy. If anybody in this list needs this file, you can contact Affy or me. I appreciate all of your help, Shirey On Wed, Dec 17, 2008 at 6:08 PM, Mark Robinson <mrobinson at="" wehi.edu.au=""> wrote: > Shirley. > > I don't think you'll find a probe id to probe id mapping. I think the way > to go is take the probe *sequences* from HuEx and match them to the > sequences on HuGene. > > At that link you sent below, there is a section called 'Sequence Files' and > there is a 'probe.tab' file that you can download (you can get a > corresponding file for HuGene). Use this to create the set of HuEx probes > that are represented on HuGene ... and then do your thing with removing > stuff from the CDF. > > Hope that helps. > > Cheers, > Mark > > > On 18/12/2008, at 8:24 AM, shirley zhang wrote: > >> Hi All, >> >> Thanks for all of your help. >> >> Since we already have 80 HuEx chips data, we want to >> reanalyze these HuEx data by only looking at those probes on HuGen >> array, then to see whether we get similar results as before. If the >> result is consistent, we are happy to switch the platform in our other >> studies. In order to find the matched probes between HuEx and HuGene, >> I looked at the data that has already been generated at Affymetrix >> that compares the probes from the Exon array with the Gene array at >> >> >> http://www.affymetrix.com/support/technical/byproduct.affx?product =huexon-st. >> >> This kind of data is exactly what I want. However, those comparison >> files, for example "Human Exon 1.0 ST to Human Gene 1.0 ST, Best Match >> (zip, 485 KB)" only includes the best matched probeset Ids between >> HuEx and HuGene. Where can I get the matched probe Id between these >> two platforms? If I can get the matched probe Ids, I can modify HuEx >> CDF file in the way that masking those probes not found in Hugene >> array as NA/missing, and then use this modified .cdf file to analyze >> exon array. >> >> Thanks again, >> Xiaoling >> >> On Fri, Dec 12, 2008 at 5:33 AM, Henrik Bengtsson <hb at="" stat.berkeley.edu=""> >> wrote: >>> >>> Hi, >>> >>> to second what already have been said and explained; you cannot reuse >>> that CDF for a different chip type. The chip types doesn't even have >>> the same number of probes. >>> >>> Note that: >>> >>> Robinson, M. D. & Speed, T. P. A comparison of Affymetrix gene >>> expression arrays. BMC Bioinformatics, 2007, 8, 449. URL: >>> http://www.biomedcentral.com/1471-2105/8/449/abstract/ >>> >>> did this kind of comparison across these chip types. Mark Robinson >>> used aroma.affymetrix for this. I'm sure he reads this thread so I >>> can correct me if I'm wrong. He and Elizabeth Purdom (I think) >>> created custom CDFs for this, cf. >>> >>> http://groups.google.com/group/aroma-affymetrix/web/huex-1-0-st-v2 >>> >>> Hope this helps >>> >>> Henrik >>> >>> >>> >>> On Thu, Dec 11, 2008 at 6:02 AM, shirley zhang <shirley0818 at="" gmail.com=""> >>> wrote: >>>> >>>> Dear List, >>>> >>>> I have a quick question. Since the HuGene probes are largely a subset >>>> of the HuEx probes, we would like to use that subset of probes for the >>>> HuEx array, then compare the results with that of using HuEx's own >>>> core probes which we already have. For this purpose, can I use >>>> HuGen's .cdf file to analyze HuEx arrays? If not, how to >>>> generate/modify a .cdf file based on HuGen probes to analyze HuEx >>>> array data? >>>> >>>> Thanks, >>>> Shirley >>>> >>>> _______________________________________________ >>>> Bioconductor mailing list >>>> Bioconductor at stat.math.ethz.ch >>>> https://stat.ethz.ch/mailman/listinfo/bioconductor >>>> Search the archives: >>>> http://news.gmane.org/gmane.science.biology.informatics.conductor >>>> >>> >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor at stat.math.ethz.ch >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: >> http://news.gmane.org/gmane.science.biology.informatics.conductor > > ------------------------------ > Mark Robinson > Epigenetics Laboratory, Garvan > Bioinformatics Division, WEHI > e: m.robinson at garvan.org.au > e: mrobinson at wehi.edu.au > p: +61 (0)3 9345 2628 > f: +61 (0)3 9347 0852 > ------------------------------ > > > > >