Hi Mireia,

Not having access to the .locs files primarily affects the ability
to accurately perform quality assessment (and maybe control) on BeadArrays.
 Assuming your array was scanned on an iScan machine (and I don't think you
get the "_perBeadFile" extension unless it was), then the intensities will
have been extracted from two separate images, one from each half of the
array surface.  This intensities are recorded in the .txt file along with
the ProbeID and coordinates but unfortunately the image each was recorded
from is not indicated, which makes the coordinate information less useful
than it might be.  You essentially have the two images overlaid on top of
one another.

Quality assessment tools such as imageplot() rely on the coordinate
information to allow you observe spatial trends, so running this on an
iScan file where you haven't been able to separate the two sets of
coordinates will give a muddled picture.  You'll probably still be able to
see a significant artefact, but it may appear to cover a larger area than
in actuality and its impact may be diluted by combining beads in affected
and unaffected areas of the array surface.

Similarly, the tools BASH and HULK attempt to correct for spatial trends in
intensities by constructing the network of neighbouring beads.  This
network construction will certainly be compromised in this situation and so
I would not recommend running either tool if you only have the
perBeadFile.txt from an iScan device.

That said, performing summarisation using beadarray should be fine and I
would still recommend using the log2 scale that is the default in beadarray
over using Illumina's summarised values.  I suspect the lack of error
message is because the readIllumina() function actually looks for files
with "Swath" in them.  It's not often we get data that isn't a dump of
everything produced by the scanner.  This is probably an oversight though,
our intention is to at least warn a user of the issues described above.

I hope that helps,

Mike Smith


On 6 August 2013 11:13, Nogales Vilardell <cressi.nogales@gmail.com> wrote:

> Hi to everybody,
>
> I have started recently to analyse Illumina Bead Chip data. I started
> without any kind of problem and I was glad with the results until I saw
> that comment from the beadarray library authors:
>
> "iScan come in a different format,.... there are two images of each array
> section (along with two .locs files), which are labeled Swath1 and Swath2"
>
> I don't have those files, I have only the perBeadFile.txt and after the
> authors wrote: "Given this, simply reading the bead-level text file will
> result in any function that uses bead locations performing undesirably"
>
> And now I am afraid if I did something wrong.. I am not receving any
> message when I read my files with the function readIllumina and they have
> also wrote that I should receive a message from the function advising me to
> use processSwathData.
>
> Does anybody know in which cases using perBeadFile.txt is ok and in which
> cases no?
>
> thanks a lot for the help
>
> Best wishes
>
> Mireia
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor@r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>



-- 
Mike Smith
Research Associate
Statistics & Computational Biology Laboratory
Cambridge University

	[[alternative HTML version deleted]]