def test_masked_windowed_divergence(self):
h = HaplotypeArray([[0, 0, 0, 0], [0, 0, 0, 1], [0, 0, 1, 1],
[0, 1, 1, 1], [1, 1, 1, 1], [0, 0, 1, 2],
[0, 1, 1, 2], [0, 1, -1, -1], [-1, -1, -1, -1]])
h1 = h.take([0, 1], axis=1)
h2 = h.take([2, 3], axis=1)
ac1 = h1.count_alleles()
ac2 = h2.count_alleles()
pos = SortedIndex([2, 4, 7, 14, 15, 18, 19, 25, 27])
mask = np.tile(np.repeat(np.array([True, False]), 5), 3)
expect, _, _, _ = allel.windowed_divergence(pos,
ac1,
ac2,
size=5,
start=1,
stop=31)
expect = expect[::2]
actual, _, _, _ = allel.windowed_divergence(pos,
ac1,
ac2,
size=10,
start=1,
stop=31,
is_accessible=mask)
assert_array_almost_equal(expect, actual)
def pairDxy(c, chrsize, ac_subpops, pos, pop2color, plot=False, blenw=10000, nwindow=100):
"""Calculates DXY
"""
dxydict = {}
windlen = int(chrsize / nwindow)
for x, y in combinations(ac_subpops.keys(), 2):
# segregating only ?
acu = ac_subpops[x] + ac_subpops[y]
flt = acu.is_segregating() & (acu.max_allele() == 1)
print("{} retaining {} SNPs".format("{}-{}".format(x, y),
np.count_nonzero(flt)))
posflt = pos[flt]
ac1 = allel.AlleleCountsArray(ac_subpops[x].compress(flt,
axis=0)[:, :2])
ac2 = allel.AlleleCountsArray(ac_subpops[y].compress(flt,
axis=0)[:, :2])
# all sites
# ac1 = ac_subpops[x]
# ac2 = ac_subpops[y]
# posflt = pos
# whole chrom
dxy = allel.windowed_divergence(posflt, ac1, ac2, size=blenw,
start=1, stop=chrsize)
dxy_m, dxy_se, *f = jackknife(dxy[0])
dxy_windowed = allel.windowed_divergence(posflt, ac1, ac2,
size=windlen, start=1,
stop=chrsize)
dxy4plot = (dxy_windowed[0], dxy_windowed[1])
dxydict["{}-{}".format(x, y)] = (dxy_m, dxy_se, dxy4plot)
if plot:
plot_dxy(dxydict, pop2color, list(ac_subpops.keys()), c, chrsize)
return(dxydict)
def dxy(p1, pos, gt, win_size, length_bp):
"""Calculate pairwise divergence between two populations.
Parameters
----------
p1 : int
size of subpop1.
pos : TYPE
DESCRIPTION.
gt : TYPE
DESCRIPTION.
win_size : TYPE
DESCRIPTION.
length_bp : TYPE
DESCRIPTION.
Returns
-------
dxy_win : TYPE
DESCRIPTION.
"""
ac1, ac2, pos_s = get_ac_seg(p1, pos, gt)
dxy_win = allel.windowed_divergence(pos_s,
ac1,
ac2,
size=win_size,
start=1,
stop=length_bp)
return dxy_win[0]
def test_windowed_divergence(self):
# simplest case, two haplotypes in each population
h = HaplotypeArray([[0, 0, 0, 0], [0, 0, 0, 1], [0, 0, 1, 1],
[0, 1, 1, 1], [1, 1, 1, 1], [0, 0, 1, 2],
[0, 1, 1, 2], [0, 1, -1, -1], [-1, -1, -1, -1]])
h1 = h.take([0, 1], axis=1)
h2 = h.take([2, 3], axis=1)
ac1 = h1.count_alleles()
ac2 = h2.count_alleles()
# mean pairwise divergence
# expect = [0/4, 2/4, 4/4, 2/4, 0/4, 4/4, 3/4, -1, -1]
pos = SortedIndex([2, 4, 7, 14, 15, 18, 19, 25, 27])
expect = [(6 / 4) / 10, (9 / 4) / 10, 0 / 11]
actual, _, _, _ = allel.windowed_divergence(pos,
ac1,
ac2,
size=10,
start=1,
stop=31)
assert_array_almost_equal(expect, actual)
size=winsize,
start=start,
stop=stop,
step=int(winsize / 2))
new_dat = format_results(stat=tajD,
stat_name="tajD",
chrom=chrom,
windows=windows,
nvar=counts,
pop=pop)
df_list.append(new_dat)
if 'dxy' in args.s and args.p2 != "None":
dxy, windows, n_bases, counts = allel.windowed_divergence(
pos,
ac,
ac2,
size=winsize,
start=start,
stop=stop,
step=int(winsize / 2))
new_dat = format_results(stat=dxy,
stat_name="dxy",
chrom=chrom,
windows=windows,
nvar=counts,
pop=pop)
df_list.append(new_dat)
if 'FD' in args.s and args.p2 != "None":
FD, windows, n_bases, counts = allel.windowed_df(
pos,
ac,
ac2,
#
# Maybe if we look at Dxy we see something clearer?
# In[28]:
clu_varbool = np.logical_and(oc_genvars_seg["POS"] > loc_start - 1e5,
oc_genvars_seg["POS"] <= loc_end + 1e5)
clu_ehh_pos = oc_genvars_seg["POS"].subset(sel0=clu_varbool)
size = 5000
step = 1000
# divergence between col-296G and gam-296G on 0 background
dxy_div_col02_gam02 = allel.windowed_divergence(
ac1=oc_genalco_sps_seg_inv_gty["col_0_2"].subset(sel0=clu_varbool),
ac2=oc_genalco_sps_seg_inv_gty["gam_0_2"].subset(sel0=clu_varbool),
pos=oc_genvars_seg["POS"].subset(sel0=clu_varbool),
size=size,
step=step)
# divergence between col-296G and gam-wt on 0 background
dxy_div_col02_gam00 = allel.windowed_divergence(
ac1=oc_genalco_sps_seg_inv_gty["col_0_2"].subset(sel0=clu_varbool),
ac2=oc_genalco_sps_seg_inv_gty["gam_0_0"].subset(sel0=clu_varbool),
pos=oc_genvars_seg["POS"].subset(sel0=clu_varbool),
size=size,
step=step)
# divergence between col-wt and gam-296G on 0 background
dxy_div_col00_gam02 = allel.windowed_divergence(
ac1=oc_genalco_sps_seg_inv_gty["col_0_0"].subset(sel0=clu_varbool),
ac2=oc_genalco_sps_seg_inv_gty["gam_0_2"].subset(sel0=clu_varbool),
def win_pi_sims(path, neut_mut, n_pops, n_sims, T, win_size, L, N):
foname = os.path.basename(path[:-1])
print(("Base filename:" + foname), flush=True)
x = np.arange(n_pops)
combs = list(itertools.combinations(x, 2))
pis = np.zeros((len(T), n_sims, n_pops, int(L / win_size)))
div = np.zeros((len(T), n_sims, len(combs), int(L / win_size)))
fst = np.zeros((len(T), n_sims, len(combs), int(L / win_size)))
tajd = np.zeros((len(T), n_sims, n_pops, int(L / win_size)))
for t in range(len(T)):
for i in range(n_sims):
files = glob(path + str(T[t]) + "N_sim_" + str(i) +
"_RAND_*[0-9]_overlaid.trees")
print(files)
assert (len(files) == 1), str(
len(files)) + " file(s) found with glob T: " + str(
T[t]) + " sim:" + str(i)
filename = files[0]
print(filename)
ts = pyslim.load(filename).simplify()
#print(("Pi0: ", ts.pairwise_diversity(samples=ts.samples(population=0)),"Pi1: ", ts.pairwise_diversity(samples=ts.samples(population=1))), flush=True)
s1 = timer()
acs, pos = ac_from_ts(ts, n_pops, N)
for j in range(n_pops):
pi, windows, n_bases, counts = allel.windowed_diversity(
pos, acs[j], size=win_size, start=1, stop=L)
pis[t, i, j, :] = pi
D, windows, counts = allel.windowed_tajima_d(pos,
acs[j],
size=win_size,
start=1,
stop=L)
tajd[t, i, j, :] = D
s2 = timer()
print(("Calculating windowed Pi/TajD... Time elapsed (min):" +
str(round((s2 - s1) / 60, 3))),
flush=True)
s1 = timer()
for k in range(len(combs)):
dxy, windows, n_bases, counts = allel.windowed_divergence(
pos,
acs[combs[k][0]],
acs[combs[k][1]],
size=win_size,
start=1,
stop=L)
div[t, i, k, :] = dxy
fstat, windows, counts = allel.windowed_hudson_fst(
pos,
acs[combs[k][0]],
acs[combs[k][1]],
size=win_size,
start=1,
stop=L)
fst[t, i, k, :] = fstat
s2 = timer()
print(("Calculating windowed Dxy and Fst... Time elapsed (min):" +
str(round((s2 - s1) / 60, 3))),
flush=True)
s1 = timer()
print((pis.shape), flush=True)
print((tajd.shape), flush=True)
print((div.shape), flush=True)
output = open(path + foname + '_pis.pkl', 'wb')
pickle.dump(pis, output)
output.close()
output = open(path + foname + '_tajd.pkl', 'wb')
pickle.dump(tajd, output)
output.close()
output = open(path + foname + '_div.pkl', 'wb')
pickle.dump(div, output)
output.close()
output = open(path + foname + '_fst.pkl', 'wb')
pickle.dump(fst, output)
output.close()
if (0):
plt.subplot(2, 1, 1)
plt.plot(np.transpose(pis[0, 0, :]), "-")
plt.title('0N after split')
plt.ylabel('Pi')
plt.subplot(2, 1, 2)
plt.plot(np.transpose(pis[9, 0, :]), "-")
plt.title('10N after split')
plt.xlabel('Window')
plt.ylabel('Pi')
plt.tight_layout()
plt.savefig(path + foname + '_landscape.pdf')
plt.close()
s2 = timer()
print(("Saving stats and plots to file... Time elapsed (min):" +
str(round((s2 - s1) / 60, 3))),
flush=True)