F2 cross and pgm

26 views
Skip to first unread message

Adrianna Jurek

unread,
Mar 18, 2025, 8:30:41 AMMar 18
to R/qtl discussion
Hello, 

I am continuing my qtl analysis, and I was looking for some clarification on how to assign pgm for my population. I have samples from Parentals (for ex. B6 and DBA), F1s, and F2s. The Parentals were used to transform the data to the format AA:AB:BB with B6 as the female and DBA as the male. I also have 3-4 more of each parental genotype/phenotype in the spreadsheet. In this case, I have been using pgm==0 for the B6 as well as the F1 and F2 populations, and pgm==1 for the DBA. Is this the correct way to denote pgm? 

Any insight is greatly appreciated.
Thank you!
Adrianna 

Karl Broman

unread,
Mar 18, 2025, 10:04:16 AMMar 18
to R/qtl discussion
The pgm variable is used to deal with the X chromosome, as in a cross (BxD)x(BxD) with females listed first, the female F2s will have genotypes BB or BD on the X, while in the cross (DxB)x(DxB) the female F2s will have genotype BD or DD.

pgm stands for "paternal grandmother". Use 0 for the F2s whose paternal grandmother is B and 1 for the F2s whose paternal grandmother is D.

The parentals and F1s shouldn't be included in the data used for QTL analysis, just the F2s.

karl

Adrianna Jurek

unread,
Mar 18, 2025, 10:24:46 AMMar 18
to R/qtl discussion
Hi Karl, 

Thanks so much for the help. I have now excluded the parents and the F1s and I am receiving basically no warnings about the Chi-square approximation now. As for transforming the data, is there a specific way you would recommend, especially for the X chromosome? I apologize for all the questions, I'm still trying to figure out r and how to manage large amounts of data appropriately, efficiently, and as the program intends. 

Thank you for your time, 
Adrianna



Karl Broman

unread,
Mar 18, 2025, 10:28:44 AMMar 18
to R/qtl discussion
I'm not sure what you mean by transforming the data.
What is it that you are trying to do?

karl

Adrianna Jurek

unread,
Mar 18, 2025, 10:37:02 AMMar 18
to R/qtl discussion
I am trying to get the data from ATCG format to AA:AB:BB format. I originally used the male and female grandparents as the template for the genotypes. In my csv file I would filter for TT in the female, and any F2 that had that genotype at those markers would be AA, same thing for GG,CC, and AA. I then filtered the male for TT, AA, GG, and CC and those would be changed to BB at the corresponding markers. Any heterozygote was denoted as AB. This was done for the X chromosome too though I now realize that was not the proper way to handle it, as all our F2s have the same female grandparent. 

Thank you!
Adrianna

Karl Broman

unread,
Mar 18, 2025, 10:44:47 AMMar 18
to R/qtl discussion
You should be able to transform the X chromosome genotypes in the same way that you transformed the autosomal genotypes.

karl

Adrianna Jurek

unread,
Mar 18, 2025, 11:00:22 AMMar 18
to R/qtl discussion
Great! Thank you so much for your quick and helpful responses, I very much appreciate it.

Adrianna
Reply all
Reply to author
Forward
0 new messages