stepwise error with 4way cross in small population

[HECTOR LOPEZ MORENO]

Hi Karl,

I am writing to you about a problem I am having with stepwise and that has already been addressed in this group but I don't know if it was really solved (see the conversation you had with a member of the group in 2011: https://groups.google. com/g/rqtl-disc/c/dTwC_KxEuiY/m/BfqPk8J9vIAJ ).

This is a summary of the dataset they had problems with in 2011:

> halibut= read.cross(format='csvr', file='halibut_famc_male_qtl.csv', genotypes=1:14, estimate.map=F)

 --Read the following data:

 24  individuals

 2186  markers

 1  phenotypes

 --Cross type: 4way 

And this is a summary of my dataset:

--Read the following data:

                   67  individuals

                   3082  markers

                   6  phenotypes

 --Cross type: 4way

Both datasets have few individuals and the type of cross is 4way. I think the main problem is the small size of the population since I used the same code for another population but with ~170 individuals and I did not have any problems (both populations are of the same species and very similar). Similar to the person who faced this issue in 2011, I temporarily solved the problem by considering only additive QTLs ( additive.only =TRUE), however, I was wondering if there was any other solution available since according to previous results there may be some significant interactions for the traits I'm working with. Thank you so much in advance for you help.

Best,

Hector