[BioC] Limma: reading in additional spot data (flags)

[Gordon K Smyth]

> Dear all,
>
> I would like to use Limma to read in not only the signal and background
> value, but also to read in additional spot data (spot quality flags, in
> this case). Our lab uses Imagene for the image processing of slides. Is
> there any way I can do this without completely hacking the source code
> and without reading the same files twice?

Use the wt.fun argument.

> If not, what would be the best
> way to do this? Secondly, is there a particular reason why the mean
> signal is read in into Limma, but the median background is read?

Mean is used for foreground because it is directly proportional to total
material hybridized to the spot.  Median is used for background because it
is lower and more robust than the mean.  (Lower backgrounds are desirable
from a stability point of view.)

> I would
> like to be able to read in the median spot intensities. Again, is this
> possible without too much hacking?

Use columns=list(f="Signal Median",b="Background Median")

Gordon

> Regards,
>
> Edo Plantinga