Fst from VariantMatrix

[agladstein]

From the documentation it looks to me like libsequence.fst.Fst takes the object polytable.SimData as input. Is that correct?
I'm reading in msprime simulations as numpy matrix

tree_np_matrix_vmat = variant_matrix.VariantMatrix(tree_np_matrix, pos)

So, can libsequence.fst.Fst take variant_matrix.VariantMatrix or variant_matrix.VariantMatrix.count_alleles as input?
Or else, how do I calculate Fst from the VariantMatrix?

[molpopgen]

You convert your numpy data to the correct format to make a SimData. See molpopgen/libsequence#35

Alternately, since the memory layout is the same as allel, use that for Fst in the meanwhile.

This is a consequence of no time to work on libsequence.

[agladstein]

Ah...

I haven't been able to find in the docs how to convert from numpy to create the SimData.

I'm not sure what you mean by "use that or Fst".

[molpopgen]

I haven't been able to find in the docs how to convert from numpy to create the SimData.

"Just do it"--all you need is Python. The input for a VariantMatrix is a matrix of integers with rows as sites. The input for SimData is a list of strings, where each string is a haplotype. (List of positions is needed for both, of course.)

I'm not sure what you mean by "use that or Fst".

s/or/for/

[agladstein]

hahaha ok :D