I am currently working on my master thesis and I have calculated some individual inbreeding coefficients and Fis per population.

Can someone explain why Fis for a population is negative (e.g. -0.011) and individual inbreeding coefficients of that population are all positive / mean individual inbreeding coefficient of that population is positive (e.g. 0.242)?

Fis values were calculated in FSTAT program. And individual inbreeding coefficients were calculated in PLINK software.

(SNPs that had LD>0.1, MAF < 0.05 and more than 10% missed genotypes were excluded from the original dataset before analysis)

 

I hope someone has an answer to my question.

Thanks in advance!

 

 

The simple answer is you used two different methods to calculate inbreeding and got different results.

 

positive Fis values indicate that individuals in a population are more related than you would expect under a model of random mating.

 

negative Fis values indicate that individuals in a population are less related than you would expect under a model of random mating.

 

You don't indicate if your Fis values significantly deviate from zero, and there will be some stochastic variation even in populations which are in Hardy-Weinberg equilibirum.

 

Also from the Plink manual:

 

Note With whole genome data, it is probably best to apply this analysis to a subset that are pruned to be in approximate linkage equilibrium, say on the order of 50,000 autosomal SNPs. Use the --indep-pairwise and --indep commands to achieve this, described here.

 

Note The estimate of F can sometimes be negative. Often this will just reflect random sampling error, but a result that is strongly negative (i.e. an individual has fewer homozygotes than one would expect by chance at the genome-wide level) can reflect other factors, e.g. sample contamination events perhaps.

Hello!

I just saw this topic about Fis and I have another question linked to that.

Is there a significant level for Fis? I mean, when we have positive or negative values of Fis, is there a threshold to say individuals are more or less related than expected from random mating ? Or is it that the biggest values mean they are much more (or much less) related?

 

And also, in the output of Fstat, how do we interpret the both analyses of Fis? One is the global Fis, and the other table shows the smaller and greater values, but then, do we check only the two last tables?

 

Thanks a lot!

Emilie