Mutation frequencies under selection (interactive Python)¶
This notebook requires Binder for the interactive features to function. See here for more information.
This applet uses moments [Jouganous et al., 2017] to calculate the site-frequency spectrum in a sample of size 20 as a function of \(\gamma = 2N_es\), the scaled strength of selection.
import holoviews as hv
import moments
hv.extension('bokeh')
def fs_negative_selection(gamma):
fs_neg = moments.Spectrum(
moments.LinearSystem_1D.steady_state_1D(20, gamma=gamma)
)[1:-1]
b = hv.Histogram(([int(i)+1 for i in range(20)], fs_neg))
b = b.redim(y=hv.Dimension('y',range=(0, 1.1*fs_neg.max())))
b = b.opts(xlabel="Sample frequency", ylabel="Density")
return b
kdims = [hv.Dimension('gamma',range=(-10, 10), step=1, default=0)]
fsplot = hv.DynamicMap(fs_negative_selection,
kdims=kdims).opts(framewise=True, width=500)
fsplot
