I have a data file (created by SDS), with 6 samples, like this: # Status Sample Detector Task Avg Ct 1 Analyzed 1 MammU6-4395470 Endogenous Control 22.733 2 Analyzed 2 MammU6-4395470 Endogenous Control 22.582 3 Analyzed 3 MammU6-4395470 Endogenous Control 22.656 4 Analyzed 4 MammU6-4395470 Endogenous Control 22.785 5 Analyzed 5 MammU6-4395470 Endogenous Control 22.684 6 Analyzed 7 MammU6-4395470 Endogenous Control 22.411 I tried running: readCtData("Data.txt", SDS=TRUE, n.features=381, type=5, Ct=6, header=TRUE, feature=4) the HTqPCR package reads the data file as if there were only 1 sample: An object of class "qPCRset" Size: 381 features, 1 samples Feature types: Endogenous Control, Target Feature names: MammU6-4395470 MammU6-4395470 MammU6-4395470 ... Feature classes: Feature categories: OK Sample names: Data NA NA ... How do I read the data into HTqPCR so that it recognizes that I have 6 samples? ./adam

Hello Adam, originally HTqPCR only read in 1 sample per file. I've recently added the parameter n.data to readCtData() to indicate the number of sample in each file being processed (along with fixing a few bugs). I still need to submit this to BioC though, so you can either wait a few days and get this from the the devel repository, or if it's urgent I can send you the corrected readCtData() function off-list. Cheers \Heidi > I have a data file (created by SDS), with 6 samples, like this: > # Status Sample Detector Task Avg Ct > 1 Analyzed 1 MammU6-4395470 Endogenous Control 22.733 > 2 Analyzed 2 MammU6-4395470 Endogenous Control 22.582 > 3 Analyzed 3 MammU6-4395470 Endogenous Control 22.656 > 4 Analyzed 4 MammU6-4395470 Endogenous Control 22.785 > 5 Analyzed 5 MammU6-4395470 Endogenous Control 22.684 > 6 Analyzed 7 MammU6-4395470 Endogenous Control 22.411 > > I tried running: > > readCtData("Data.txt", SDS=TRUE, n.features=381, type=5, Ct=6, > header=TRUE, feature=4) > > the HTqPCR package reads the data file as if there were only 1 sample: > > An object of class "qPCRset" > Size: 381 features, 1 samples > Feature types: Endogenous Control, Target > Feature names: MammU6-4395470 MammU6-4395470 MammU6-4395470 ... > Feature classes: > Feature categories: OK > Sample names: Data NA NA ... > > How do I read the data into HTqPCR so that it recognizes that I have 6 > samples? > > ./adam > > _______________________________________________ > 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 >

Hello Adam, just a quick email to let you know that I've implemented your suggested changes in HTqPCR. This includes e.g. a getCt() and history () funciton, along with adjusted p-values for the t-test. These functions are available in v. 1.1.2 that should be available from the BioC devel repository within the next few days. Cheers \Heidi On 12 Jan 2010, at 17:30, Adam Kiezun wrote: > Great, > Thanks Heidi. This worked nicely. > >> Can you give me a bit more details about how these files were >> generated? >> Is it a standard TLDA card from ABI (with 381 features or have 3 >> genes >> automatically been excluded)? Are the cards analysed in batch >> somehow on >> the SDS software, or are they processed individually and then just >> exported to a text file together? > > I actually don't know. I'm analysing the data for someone else who did > the experiment. The cards contain microRNAs, not protein-coding genes. > There are 2 plates actually, one with 381 and the other with 216 > miRNAs. We got the files from AB in the format I showed you - I don't > know how they exported it. > > One useful feature for the HTqPCR package would be to include adjusted > p-values (method could be passed as a parameter eg "FDR"), and fold > changes. It's something everyone needs anyway. Right now, I called > p.adjust and 2^(-ddCt) myself and added those as a column in the > result table. But it would be nicer to have this work automagically. > > Another convenience would be to have the qPCR object remember what > normalizations and filterings were done on it (like the objects in the > lumi package). Right now, I store this in the name of the variable > (eg raw.cat.filt.qnorm) but that's suboptimal. > > It should be more intuitive to access Ct values from the qPCR object. > Right now, it's something like exprs(object). How about simply > ctValues(object) or something? > > A nitpick: it would be nice to have the documentation contain the code > to create all figures presented (missing for Figures 6, 7) > > All in all, it's a great package. Thanks for your help and I look > forward to future improvements. > ./adam