I work with diploid and sexual species. Currently, I have AFLP data from populations of a self-sterile species and a self-fertile species. I was asked to estimate observed (Ho) and expected heterozygosity (He) levels and to compare selfing rates (s) in order to get a notion of how mating systems could influence population genetic patterns in these species... but I didn't find any software/test/calculation/published work/etc, which I could use dominant markers to estimate Ho, He and s. If someone could help me I will appreciate it!
Verena
Hickory (https://kholsinger.github.io/Hickory/) can be used to generate Fst and inbreeding coefficient estimates from dominant markers such as AFLP, although they have to be interpreted with caution. I think the data it uses must be variation between samples in the overall frequency of the band-absent genotypes. The more inbred an individual, the more likely they would be homozygous for the band-absent allele and thus the higher genome-wide frequency of band absence.
The problem is that both the Bayesian sampling algorithm and technical issues can generate biases. In one of our studies from many years ago (Molecular Ecology (2007) 16:4014–4027) we obtained a lot of inbreeding coefficient estimates of 0.4-0.7 in a self-incompatible plant species, which couldn't have been correct. AFLP tends to produce band dropout at larger fragment size in a fraction of samples, perhaps if there was more shearing in the DNA isolation or sample preparation steps, and that will also generate higher band-absent frequencies. However, it's certainly worth a try if you have the data already, and you can see if the results make sense.
I will have a look at Hickory and see what results I obtain
Thanks a lot David L Remington !
- Badges
- Science method