Confused about data types

[N I]

Dear all / Karl,

I am struggling with one of my analyses in my undergrad dissertation.

I have nominal scores of plant habit (1-9). I have two replicates of these scores for each of my plant lines.
I know it's not a good idea to use a mean of this data, because it makes for a wacky distribution.

I have used means of my reps for all my other data, which is in continuous measurements (eg. plant height). 

What can I do about my nominal data, please? Can R/QTL handle it? 
Was I right in using a mean of my reps in my continuous data types?

Best wishes,
Nicola Ibbotson

[Karl Broman]

I think it is likely fine to treat it as a continuous measurement. 

R/qtl doesn't have any special facility for handling nominal phenotypes.

But as long as you calculate significance thresholds using a permutation test, you should be okay.

karl