30 views

### nata...@gmail.com

Apr 18, 2021, 8:57:36 AMApr 18
to R/qtl discussion
hi
im trying to understand the fitqtl output of estimated additive effects
these are the results of the 1st phenotype:
Estimated effects:
-----------------
est         SE         t
Intercept   2.09856  0.11345 18.497
LG...@0.4a   0.47991  0.16158  2.970
LG...@0.4d  -0.52841  0.22494 -2.349
LG...@12.0a -0.59555  0.15523 -3.837
LG...@12.0d -0.32547  0.23249 -1.400
LG...@34.2a  0.42254  0.15301  2.762
LG...@34.2d  0.03115  0.23057  0.135
LG...@26.0a  0.22750  0.16192  1.405
LG...@26.0d  0.87725  0.24198  3.625
LG...@30.0a -0.00225  0.17215 -0.013
LG...@30.0d  0.74030  0.24974  2.964

what can I infer about the 1st loci?
what is the conclusion? Is the allele dominant?
do I have to consider the intercept?

thanks a lot
Nataly

### nata...@gmail.com

Apr 18, 2021, 9:01:37 AMApr 18
to R/qtl discussion
the cross is an F2 intercross

ב-יום ראשון, 18 באפריל 2021 בשעה 15:57:36 UTC+3, nata...@gmail.com כתב/ה:

### Karl Broman

Apr 18, 2021, 6:43:08 PMApr 18
to R/qtl discussion
The additive effect is half the difference between the two homozygotes' phenotype averages, and the dominance effect is the difference between the heterozygote and the midpoint between the two homozygotes. So if d == a or d == -a, one of the alleles is dominant. With a = 0.48 and d = 0.-52, it looks like dominance. But note that the standard errors are large, 0.2.

There's a bit of discussion of this in the R/qtl book in section 4.6; available online: https://rqtl.org/book/rqtlbook_ch04.pdf
Otherwise look for a book on quantitative genetics, like Falconer and Mackay or Lynch and Walsh.

karl

### Ramesh Bhat

Apr 19, 2021, 12:16:34 PMApr 19
Dear Dr. Karl,
I used two different methods of reading the hyper data as indicated below
data(hyper)
hyper=read.cross("csv", file="hyper.csv", genotypes=c("AA", "BB", "BA"), crosstype="f2", na.strings=c("-","NA"))

Found that effectplot() in the first case shows only two genotypes (BB and BA) while the effectplot with the read.cross shows three genotypes.

kind regards,

--
You received this message because you are subscribed to the Google Groups "R/qtl discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rqtl-disc+...@googlegroups.com.

--
Ramesh S. Bhat
Professor
Department of Biotechnology