Hi Henrik, Thanks a lot for your reply.
I am aware that probes/cells are not annotated just like probesets or transcript clusters, so I just didn't explain my problem properly. I have already done the analysis at the exon and transcript level using your ExonRmaPlm() with mergeGroups=TRUE or FALSE for each case so that part of the analysis, the one most people are interested in, is fine. I am interested in mapping intensities to sequences of nucleotides: I have downloaded Affymetrix's library files and in one of them one can find the target sequence corresponding to each "probe identity/index" - from 1 to 6.5 million - that can be uniquely obtained using the (x,y) coordinates of the probe on the cdf file. The reason is that I would like to compare the performance of Exon arrays to NGS and that's why the individual probes are so important for me. So my question is whether it is possible to perform an analysis at the "sequence level". Yet another question, maybe too general, I have also noticed that in some cases the 4 probes interrogating the same exon (Probe Selection Region) have extremely different intensities. Do you think it is a consequence of the GC content of the individual probes or would you blame it on cross-hybridazation? Would you somehow correct the probe intensities by their GC content? Again, thank you very much for your answer and for providing this discussion group that gets direct support. I find it extremely useful. Best regards, María On Jul 10, 1:40 am, Henrik Bengtsson <henrik.bengts...@aroma- project.org> wrote: > Hi Maria. > > On Wed, Jul 6, 2011 at 5:59 AM, Maria Rodrigo-Domingo <mrod...@gmail.com> > wrote: > > Hi all, > > > I am using aroma.affymetrix for an analysis on Human Exon Array data > > and would need to access the individual probe intensities mappeda to > > probe_id after background correction and normalization. Is there an > > easy way to do this? Below you can see what I have tried so far. > > > With the notation onhttp://www.aroma-project.org/node/37I am looking > > at object csN - data at probe-level after BC and QN > > >> affyBatch <- extractAffyBatch(csN) > > # I get an error message about package "huex10stv2cdf" not being > > available from Bioconductor. > > Yes, that approach, which tries to extract a "Bioconductor" AffyBatch > object, is basically just a convenient wrapper around ReadAffy() of > the affy package, and that is what needs a 'huex10stv2cdf' package. > > >> probes_matrix <- extractMatrix(csN, cells=NULL, field=c("intensities", > >> "stdvs", "pixels"), drop=FALSE, verbose=verbose) # I do not get any > >> information about which probe the intensity belongs to > > It might be that you've missed that *probes* (aka *cells* as we prefer > to call them) do *not* have annotations in Affymetrix arrays. It is > only *probesets* (aka *unit groups*) that have annotations. > Probesets, which are defined by the CDF you choose, specifies which > sets of probes should be grouped together. > > For example, if you know you are interested in probes 1040-1054, you > can extra their probe intensities across all arrays in the > AffymetrixCelSet ('csN') as you do: > > > cells <- 1040:1054; > > Y <- extractMatrix(csN, cells=cells, field=c("intensities", "stdvs", > > "pixels"), drop=FALSE, verbose=verbose); > > str(Y) > > num [1:15, 1:6] 37 32 96 214 706 29 39 218 39 792 ... > - attr(*, "dimnames")=List of 2 > ..$ : NULL > ..$ : chr [1:6] "huex_wta_cerebellum_A" "huex_wta_cerebellum_B" ... > > Note how there are column names, which corresponds to the array/sample > names, but there are no row names simply because probes don't have > names. Probes are specified by the (x,y) coordinates on the array and > there is a one-to-one mapping (x,y) <-> probe index. You can read > more about this in the help of the affxparser package: > > > help("2. Cell coordinates and cell indices", package="affxparser") > >> intensities <- getIntensities(csN) > > # no probe or array information > > Same reason as above. (Also, I recommend to use extractMatrix() > instead of getIntensities()). > > >> unitIntensities <- getUnitIntensities(csN, verbose=verbose) > > # I get transcript and probeset information and the individual > > intensities of the 4 probes in each probeset, but numbered 1 to 4, so > > No we're getting closer. The getUnitIntensities() returns a list > structure containing probe intensities. The list structure reflects > the structure of the CDF, i.e. which cells belongs to which unit > groups. Note how the items in the list have names. These are the so > called "unit names". Each item in turn consists of a sublist, which > corresponds to the units groups. They also have names which are the > so called "group names". > > > I still cannot map them to their target sequence. > > Q. So, does this mean that you are interested in finding the genomic > location of each probe? Is that the kind of annotation you are > looking for when you ask for "probe_id"? Note that when they designed > the custom CDFs for FIRMA, they did map the 25-mer probes to the > genome by their sequences, cf. subpages > viahttp://aroma-project.org/chipTypes/HuEx-1_0-st-v2 > > Q. Could it be that you are interested in transcript or exon > summaries? See how-to page 'Extract probeset summaries (chip > effects) as a data frame' > [http://www.aroma-project.org/howtos/extractDataFrame] and the example > for 'HuEx-1_0-st-v2' - is that what you're looking for? > > The answer will depend on what you are really after. > > /Henrik > > > > > > > > > > > I have now run out of ideas so I would really appreciate any > > suggestions. Thank you very much in advance for your help. > > > Best regards, > > Maria Rodrigo > > > PhD student in Biostatistics > > Department of Haematology, Medical Center Aalborg Hospital Science and > > Innovation Center AHSIC Denmark > > > -- > > When reporting problems on aroma.affymetrix, make sure 1) to run the latest > > version of the package, 2) to report the output of sessionInfo() and > > traceback(), and 3) to post a complete code example. > > > You received this message because you are subscribed to the Google Groups > > "aroma.affymetrix" group with websitehttp://www.aroma-project.org/. > > To post to this group, send email to aroma-affymetrix@googlegroups.com > > To unsubscribe and other options, go tohttp://www.aroma-project.org/forum/ -- When reporting problems on aroma.affymetrix, make sure 1) to run the latest version of the package, 2) to report the output of sessionInfo() and traceback(), and 3) to post a complete code example. You received this message because you are subscribed to the Google Groups "aroma.affymetrix" group with website http://www.aroma-project.org/. To post to this group, send email to aroma-affymetrix@googlegroups.com To unsubscribe and other options, go to http://www.aroma-project.org/forum/
