[BioC] limma: cannot repoduce older analysis

Philipp Pagel p.pagel at wzw.tum.de
Thu Jul 24 15:00:56 CEST 2008

	Hi Wolfgang,

> the quickest way to figure out what's going on may be to download and
> install an older version of R (as good as you can remember what it was),
> run
>  biocLite("limma")
> and then your script and see exactly at what point your results start to
> differ.

Good advice - thanks! I tried R 2.6.2 / limma_2.12.0 and was able to
nail down the difference to lmFit() which looks like it has undergone a
major re-write somewhere between 2.12.0 and 2.14.2.

So now I'll spend some time studying the source code - maybe I can
figure out what exactly causes the change.

If anyone else has experienced a similar effect I'd be interested to
hear about it...

> Is "targets" the same? 

Yes, it is.


Dr. Philipp Pagel
Lehrstuhl für Genomorientierte Bioinformatik
Technische Universität München
Wissenschaftszentrum Weihenstephan
85350 Freising, Germany

More information about the Bioconductor mailing list