Hi,
I believe you cannot calculate pairwise Fst with allele frequency data since you don't have any information on heterozygosity.
That being said, you might be able to simulate populations using the rgenotypes function from the mmod package and calculate the statistics on the simulated populations and get confidence intervals. That being said, the function can only simulation one locus and one population at a time, so you will have to construct a couple of loops/lapply functions to get the job done for each of your populations and loci.
Hope that helps,
Zhian