[BioC] Combining raw images and locs files from different scans

[Salvador]

Dear Mike,

Thanks for the quick response. It explained a lot. I'm still awaiting 
the data from the second scan. Hopefully the intensities won't be too 
different. I don't want to seem cheeky, but am very new to R and 
microarray analysis in particular. So any suggestions on how to use R to 
run a comparison between the scans would be greatly appreciated. All I 
can think of is building an MA plot for each sample's intensities from 
first and second scans. But even then, what would I consider to be "very 
different". I.e. is there any quantitative parameter I could use to 
decide, whether the data from the two arrays if comparable.

Many thanks,

Aliaksei.

На 19.12.2011 18:41, Mike Smith напісаў(ла):
> Hi Aliaksei,
>
> I'm afraid it isn't appropriate to use the jpeg images.  As far as I'm
> aware the jpegs produced by the Illumina scanner are always compressed,
> so it's impossible to reproduce the tiff images that the intensities are
> calculated from. Moreover, although the ordering of the beads on the
> array grid won't have changed, the positions of the beads within the
> image will almost certainly be different.  You can probably verify this
> by looking in the two perBeadFile.txt files that were created for an
> array from the two scans.  I'd expect you to see the same number of
> beads with each ID, and in the same order, but with slightly different
> coordinate pairs.  Given this, even if you could recreate the images
> from the first scan, it wouldn't be appropriate to use them in
> combination with the coordinates of the second.
>
> If you're worried about the effects of rescanning, as a first test I
> would read the two perBeadFile.txt files from the two scans and compare
> the intensity values.  Don't worry about trying to split them into
> swaths for this, as the ordering should be consistent between the two
> files.  If they look similar then I'd simply use the second scan.  If
> the impact of the rescanning looks large then you can use the
> processSwathData() function in beadarray to split the files from the
> second scan and then, based on the files that generates, go back and
> split the files from the first scan.  That'll be quite a lot more work
> though, so hopefully the second scan is fine.
>
> I hope that helps,
>
> Mike Smith
>
> On Mon, Dec 19, 2011 at 12:54 PM, Aliaksei Holik <salvador at bio.bsu.by
> <mailto:salvador at bio.bsu.by>> wrote:
>
>     Dear list members,
>
>     Hopefully, this will be a straightforward question. I had my samples
>     analysed on Illumina beadchip by university services using iScan.
>     When I received the data all images were in jpeg format and there
>     were no locs files, I assume due to compressed image format. After
>     some discussion with the services my chips were re-scanned producing
>     tiff images and appropriate locs files. However, there was almost 3
>     weeks interval between the first and the second scans. I am
>     concerned, therefore, that signal might have deteriorated and I'm
>     thus wary of using the intensities data and images from the second scan.
>
>     My question is therefore, can I convert the original jpeg files in
>     tiff format and combine them with locs files from the second scan to
>     import the intensities data generated during the first scan using
>     beadarray package. My reasoning being that no compression seemingly
>     was applied to jpeg files and the same probes should be in different
>     swaths during both first and the second scans. Please, correct me if
>     I'm wrong.
>
>     Many thanks!
>
>     Aliaksei.
>
>     _________________________________________________
>     Bioconductor mailing list
>     Bioconductor at r-project.org <mailto:Bioconductor at r-project.org>
>     https://stat.ethz.ch/mailman/__listinfo/bioconductor
>     <https://stat.ethz.ch/mailman/listinfo/bioconductor>
>     Search the archives:
>     http://news.gmane.org/gmane.__science.biology.informatics.__conductor <http://news.gmane.org/gmane.science.biology.informatics.conductor>
>
>
>
>
> --
> Mike Smith
> PhD Student
> Computational Biology Group
> Cambridge University
>