[BioC] Heatdiagram

James Wettenhall wettenhall at wehi.edu.au
Mon Jun 14 18:03:36 CEST 2004

Hi Binita,

On Mon, 14 Jun 2004, Binita Dutta wrote:
> I have done analysis by limma package but finally when i want to plot 
> heatdiagram, all clones which are differenetially expressed are labeled on 
> Y axis (which makes it impossible to read the label), where as i am 

I'm not so familiar with the new heatDiagram function in limma, 
but in the old heatdiagram function (lower case 'd') (still 
available in limma), you can specify critical.primary and 
critical.other.  Choosing higher values for these will limit the 
number of genes displayed in your heatdiagram.  Have a look at 
the help for heatdiagram:

The first argument to heatdiagram could be your 
matrix of (moderated) t statistics (i.e. fit$t, after 
running eBayes on fit).  Then critical.primary and 
critical.other would be (moderated) t statistic values.  
Remember that the heatdiagram is plotted relative to one 
"primary" contrast.  For this contrast the colors should appear 
in order from green to red (or the other way around, depending 
on how you look at the diagram).  This should help you to 
understand the difference between the two t statistic 
threshold values: critical.primary and critical.other.  

So for example, you could try increasing critical.primary from 
the default value of 4 to 10 and increase critical.other from 3 
to 8.  To get an idea of how much to increase them in your case, 
have a look at the "t" column in your toptable.

Hope this helps,

More information about the Bioconductor mailing list