[BioC] Using Sample IDs from Sample Name column

[Heidi Dvinge]

> Good morning,
> I have a large LDA data set of 480 cards. Each card has 8 samples with
> the same 48 assays in each lane. I do not have an array of sample IDs
> outwith the data (although I could make one if I had to = time
> consuming). readCtData seems to use the filename as the sample ID by
> default, or accept an array of filenames. My filenames are the barcode
> of the LDA, not the sample name.
>
> In readCtData is there a way to assign sample = from a column within the
> data files?
>
Hi Ross,

in principle, in readCtData you can change the sample name from the
default, i.e. the filename, saying e.g.

raw <- readCtData(..., samples=file$samplename).

However, if you have multiple samples on each TLDA card, then after some
initial QC and possibly after normalisation, you'd probably want to
reformat the qPCRset object anyway before testing for differential
expression. That way you'll have 1 sample per column with 48 rows
(assays), rather than 8 samples per column, thereby clearly separating the
samples before the statistical analysis. This can be achieved with
changeCtLayout, but you'd porbably need to assign the correct sample names
afterwards. This can be done at any stage with sampleNames(qPCRset) <-
"vector of whatever you want, from file or otherwhere".

HTH
\Heidi

> Regards,
>
> RH
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>