Dear answers,
I used Fu's Fs to test the neutrality and demographic history in Arlequin. But I get a value like 340282346638528859811704183484516925440.00000 (P=1). I think this value certainly can't use. Can any friend explain why this happened? How to solve this problem without abandon Arllequin?
Thanks a lot.
Hello Cheng,
Thank you for your question and for providing your input files. I've run into a similar issue myself, specifically when comparing singleton sequences to larger groups. In your case, I see your Hap 7 and Hap 11 are represented by just a single sequence a piece. Are these the haplotypes that are giving you issues?
If so, I would simply drop them from your neutrality tests. I ran into this when assessing sequences from different populations, and had a couple pops represented by just one or two individuals. I wound up resolving the issue by re-running the analysis after including these "smaller" populations in a larger, regional group.
I realize this post is 2 years old, but I hope this helps you and others nonetheless!
Cheers,
Wyatt
Hello Wyatt,
Thank you for your answer. Now I understand the reason why I got a large Fu's Fs value. This is not late in my future study.
Sincerely,
Cheng
Dears Cheng Jilong and Wyatt Shell
I also got the sample problem. Could you tell me, kow did you solve this problem?
Did you publish these values or inform the exclusion of haplotypes with just one specimen/sequence?
Thanks all,
Renato
Hey Renato,
Hoping you're well and thank you for your question. As I offer in my suggestions for Cheng Jilong, I've run into similar issues when comparing populations of many individuals to those represented by one or just a few individuals.
In your case, do you have pops/groups represented by just one or two sequences? If so, if you try re-running the analysis without these sequences do your FS values improve (i.e. fall back into a realistic range)? If so, is there some way you can bin these sequences to include them in a larger group or perhaps drop them entirely?
In my opinion, I would do what you can to fix the wacky FS value before submitting for publication as this will absolutely stand out as a red flag for any savvy pop gen reviewer.
Hoping this is of help and please don't hesitate to reach back out here or through DM.
Cheers,
Wyatt
- Badges
- Science topic
- Similar topics
- Social Psychology
- Values