[BioC] limma and print tip groups
Gordon Smyth
smyth at wehi.EDU.AU
Fri Sep 15 12:45:25 CEST 2006
Dear Michael,
limma expects spots to be in "standard order", in which the fastest
to slowest moving indices are spot columns, spot rows, grid columns,
grid rows. Image analysis programs pretty much always output data in
this order. If your data is not in this order, you will need to
re-order your RGList object before print-tip normalization.
Best wishes
Gordon
>Date: Fri, 15 Sep 2006 10:11:53 +0100
>From: "michael baron \(IAH-P\)" <michael.baron at bbsrc.ac.uk>
>Subject: [BioC] limma and print tip groups
>To: <bioconductor at stat.math.ethz.ch>
>Message-ID:
>
><88F7F2EC6769F94487FAEA19357EA51101663CDC at iahpe2ksrv1.iah.bbsrc.ac.uk>
>Content-Type: text/plain; charset="iso-8859-1"
>
>Hi all,
>
>I trawled a bit back through the archives and couldn't see an answer
>to this, so perhaps someone can help.
>
>Limma uses a Printer Class to identify which print group a spot
>belongs to. This data object includes the number of grids, their
>arrangement and how many rows and columns of spots their are in each
>grid. This is easy to add to an RGList manually. The package does
>not seem to need a specific columns of info relating each spot to a
>particular grid, so I am guessing that there is an assumption that
>the functions are expecting the data to be arranged in a particular
>way. Is this so? I am analysing some data that is arranged in what I
>suspect is a non-standard way (all the spots in the first row across
>the ARRAY first, then the second row of the array, and so on). Is
>limma expecting the data to be 1st row of 1st grid, then 2nd row of
>1st grid, and so on? If not, could someone tell me what it is
>expecting, or perhaps just that I am totally wrong and limma is
>working out the grid-spot linkage in some other way?
>
>Thanks in advance,
>
>Michael
>
>*******************************************************
>
>Michael D. Baron Ph.D.
>Group Leader, Morbillivirus-Host Interactions
>Institute for Animal Health
>Ash Road,
>Pirbright,
>Surrey GU24 0NF
>U.K.
More information about the Bioconductor
mailing list