def test_post(self):
k = 6
n = 1e4
rho = 1.5e-8 * 10
mu = 2.5e-8 * 10
length = 10000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
print "muts", len(muts)
print "recombs", len(arglib.get_recomb_pos(arg))
times = arghmm.get_time_points(ntimes=10)
arghmm.discretize_arg(arg, times)
tree = arg.get_marginal_tree(0)
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
# remove chrom
new_name = "n%d" % (k - 1)
keep = set(arg.leaf_names()) - set([new_name])
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
model = arghmm.ArgHmm(arg, seqs, new_name=new_name,
times=times, rho=rho, mu=mu)
print "states", len(model.states[0])
probs = arghmm.get_posterior_probs(model, length, verbose=True)
for pcol in probs:
p = sum(map(exp, pcol))
print p, " ".join("%.3f" % f for f in map(exp, pcol))
fequal(p, 1.0, rel=1e-2)
def test_norecomb_plot(self):
k = 50
n = 1e4
rho = 1.5e-8 * .0001
rho2 = 1.5e-8 * 10
mu = 2.5e-8 * 100
length = 1000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
times = arghmm.get_time_points(ntimes=20)
arghmm.discretize_arg(arg, times)
tree = arg.get_marginal_tree(0)
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
# get thread
new_name = "n%d" % (k - 1)
keep = ["n%d" % i for i in range(k - 1)]
arglib.subarg_by_leaf_names(arg, keep)
arg.set_ancestral()
arg.prune()
model = arghmm.ArgHmm(arg,
seqs,
new_name=new_name,
times=times,
rho=rho2,
mu=mu)
print "states", len(model.states[0])
print "muts", len(muts)
# simulate a new thread
states = list(islice(hmm.sample_hmm_states(model), 0, arg.end))
data = list(hmm.sample_hmm_data(model, states))
seqs[new_name] = "".join(data)
#alignlib.print_align(seqs)
thread = [
model.times[model.states[i][s][1]] for i, s in enumerate(states)
]
p = plot(thread, style="lines")
probs = arghmm.get_posterior_probs(model, length, verbose=True)
print "done"
high = list(arghmm.iter_posterior_times(model, probs, .75))
low = list(arghmm.iter_posterior_times(model, probs, .25))
p.plot(high, style="lines")
p.plot(low, style="lines")
pause()
def test_post_real(self):
k = 3
n = 1e4
rho = 1.5e-8
mu = 2.5e-8
length = 100000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
#arg = arglib.read_arg("test/data/real.arg")
#seqs = fasta.read_fasta("test/data/real.fa")
#arglib.write_arg("test/data/real.arg", arg)
#fasta.write_fasta("test/data/real.fa", seqs)
times = arghmm.get_time_points(maxtime=50000, ntimes=20)
arghmm.discretize_arg(arg, times)
new_name = "n%d" % (k - 1)
thread = list(
arghmm.iter_chrom_thread(arg, arg[new_name], by_block=False))
tree = arg.get_marginal_tree(0)
print tree.root.age
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
p = plot(cget(thread, 1), style="lines", ymin=10, ylog=10)
#alignlib.print_align(seqs)
# remove chrom
keep = ["n%d" % i for i in range(k - 1)]
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
model = arghmm.ArgHmm(arg,
seqs,
new_name=new_name,
times=times,
rho=rho,
mu=mu)
print "states", len(model.states[0])
#print "muts", len(muts)
print "recomb", len(model.recomb_pos) - 2, model.recomb_pos[1:-1]
probs = arghmm.get_posterior_probs(model, length, verbose=True)
high = list(arghmm.iter_posterior_times(model, probs, .95))
low = list(arghmm.iter_posterior_times(model, probs, .05))
p.plot(high, style="lines")
p.plot(low, style="lines")
pause()
def test_post_c(self):
k = 3
n = 1e4
rho = 1.5e-8 * 30
mu = 2.5e-8 * 100
length = 100
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
arg.prune()
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
print arglib.get_recomb_pos(arg)
print "muts", len(muts)
print "recomb", len(arglib.get_recomb_pos(arg))
times = arghmm.get_time_points(ntimes=10)
arghmm.discretize_arg(arg, times)
tree = arg.get_marginal_tree(0)
print tree.root.age
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
# remove chrom
keep = ["n%d" % i for i in range(k - 1)]
arglib.subarg_by_leaf_names(arg, keep)
model = arghmm.ArgHmm(arg,
seqs,
new_name="n%d" % (k - 1),
times=times,
rho=rho,
mu=mu)
print "states", len(model.states[0])
util.tic("C")
probs1 = list(arghmm.get_posterior_probs(model, length, verbose=True))
util.toc()
util.tic("python")
probs2 = list(hmm.get_posterior_probs(model, length, verbose=True))
util.toc()
print "probs1"
pc(probs1)
print "probs2"
pc(probs2)
for col1, col2 in izip(probs1, probs2):
for a, b in izip(col1, col2):
fequal(a, b)
def test_norecomb_plot(self):
k = 50
n = 1e4
rho = 1.5e-8 * .0001
rho2 = 1.5e-8 * 10
mu = 2.5e-8 * 100
length = 1000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
times = arghmm.get_time_points(ntimes=20)
arghmm.discretize_arg(arg, times)
tree = arg.get_marginal_tree(0)
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
# get thread
new_name = "n%d" % (k-1)
keep = ["n%d" % i for i in range(k-1)]
arglib.subarg_by_leaf_names(arg, keep)
arg.set_ancestral()
arg.prune()
model = arghmm.ArgHmm(arg, seqs, new_name=new_name, times=times,
rho=rho2, mu=mu)
print "states", len(model.states[0])
print "muts", len(muts)
# simulate a new thread
states = list(islice(hmm.sample_hmm_states(model), 0, arg.end))
data = list(hmm.sample_hmm_data(model, states))
seqs[new_name] = "".join(data)
#alignlib.print_align(seqs)
thread = [model.times[model.states[i][s][1]]
for i, s in enumerate(states)]
p = plot(thread, style="lines")
probs = arghmm.get_posterior_probs(model, length, verbose=True)
print "done"
high = list(arghmm.iter_posterior_times(model, probs, .75))
low = list(arghmm.iter_posterior_times(model, probs, .25))
p.plot(high, style="lines")
p.plot(low, style="lines")
pause()
def test_smcify_arg_remove_thread(self):
rho = 1.5e-8 # recomb/site/gen
l = 100000 # length of locus
k = 6 # number of lineages
n = 2*10000 # effective popsize
arg = arglib.sample_arg(k, n, rho, 0, l)
remove_chroms = set("n%d" % (k-1))
keep = [x for x in arg.leaf_names() if x not in remove_chroms]
arg = arg.copy()
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
def test_post_real(self):
k = 3
n = 1e4
rho = 1.5e-8
mu = 2.5e-8
length = 100000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
#arg = arglib.read_arg("test/data/real.arg")
#seqs = fasta.read_fasta("test/data/real.fa")
#arglib.write_arg("test/data/real.arg", arg)
#fasta.write_fasta("test/data/real.fa", seqs)
times = arghmm.get_time_points(maxtime=50000, ntimes=20)
arghmm.discretize_arg(arg, times)
new_name = "n%d" % (k - 1)
thread = list(arghmm.iter_chrom_thread(arg, arg[new_name],
by_block=False))
tree = arg.get_marginal_tree(0)
print tree.root.age
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
p = plot(cget(thread, 1), style="lines", ymin=10, ylog=10)
#alignlib.print_align(seqs)
# remove chrom
keep = ["n%d" % i for i in range(k-1)]
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
model = arghmm.ArgHmm(arg, seqs, new_name=new_name, times=times,
rho=rho, mu=mu)
print "states", len(model.states[0])
#print "muts", len(muts)
print "recomb", len(model.recomb_pos) - 2, model.recomb_pos[1:-1]
probs = arghmm.get_posterior_probs(model, length, verbose=True)
high = list(arghmm.iter_posterior_times(model, probs, .95))
low = list(arghmm.iter_posterior_times(model, probs, .05))
p.plot(high, style="lines")
p.plot(low, style="lines")
pause()
def test_post_c(self):
k = 3
n = 1e4
rho = 1.5e-8 * 30
mu = 2.5e-8 * 100
length = 100
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
arg.prune()
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
print arglib.get_recomb_pos(arg)
print "muts", len(muts)
print "recomb", len(arglib.get_recomb_pos(arg))
times = arghmm.get_time_points(ntimes=10)
arghmm.discretize_arg(arg, times)
tree = arg.get_marginal_tree(0)
print tree.root.age
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
# remove chrom
keep = ["n%d" % i for i in range(k-1)]
arglib.subarg_by_leaf_names(arg, keep)
model = arghmm.ArgHmm(arg, seqs, new_name="n%d" % (k-1), times=times,
rho=rho, mu=mu)
print "states", len(model.states[0])
util.tic("C")
probs1 = list(arghmm.get_posterior_probs(model, length, verbose=True))
util.toc()
util.tic("python")
probs2 = list(hmm.get_posterior_probs(model, length, verbose=True))
util.toc()
print "probs1"
pc(probs1)
print "probs2"
pc(probs2)
for col1, col2 in izip(probs1, probs2):
for a, b in izip(col1, col2):
fequal(a, b)
def test_post3(self):
k = 3
n = 1e4
rho = 1.5e-8 * 3
mu = 2.5e-8 * 100
length = 10000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
arg.prune()
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
times = arghmm.get_time_points(ntimes=10)
arghmm.discretize_arg(arg, times)
tree = arg.get_marginal_tree(0)
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
thread = list(arghmm.iter_chrom_thread(arg, arg["n2"], by_block=False))
p = plot(cget(thread, 1), style="lines", ymin=0)
# remove chrom
keep = ["n0", "n1"]
arglib.subarg_by_leaf_names(arg, keep)
arg.set_ancestral()
arg.prune()
model = arghmm.ArgHmm(arg,
seqs,
new_name="n2",
times=times,
rho=rho,
mu=mu)
print "states", len(model.states[0])
print "muts", len(muts)
print "recomb", len(model.recomb_pos) - 2, model.recomb_pos[1:-1]
p.plot(model.recomb_pos, [1000] * len(model.recomb_pos),
style="points")
probs = arghmm.get_posterior_probs(model, length, verbose=True)
high = list(arghmm.iter_posterior_times(model, probs, .95))
low = list(arghmm.iter_posterior_times(model, probs, .05))
p.plot(high, style="lines")
p.plot(low, style="lines")
pause()
def test_iter_sprs_remove_thread(self):
rho = 1.5e-8 # recomb/site/gen
l = 100000 # length of locus
k = 6 # number of lineages
n = 2*10000 # effective popsize
arg = arglib.sample_arg(k, n, rho, 0, l)
remove_chroms = set("n%d" % (k-1))
keep = [x for x in arg.leaf_names() if x not in remove_chroms]
arg = arg.copy()
arglib.subarg_by_leaf_names(arg, keep)
for a, b in izip(arglib.iter_arg_sprs(arg),
arglib.iter_arg_sprs_simple(arg)):
self.assertEqual(a, b)
def test_post3(self):
k = 3
n = 1e4
rho = 1.5e-8 * 3
mu = 2.5e-8 * 100
length = 10000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
arg.prune()
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
times = arghmm.get_time_points(ntimes=10)
arghmm.discretize_arg(arg, times)
tree = arg.get_marginal_tree(0)
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
thread = list(arghmm.iter_chrom_thread(arg, arg["n2"], by_block=False))
p = plot(cget(thread, 1), style="lines", ymin=0)
# remove chrom
keep = ["n0", "n1"]
arglib.subarg_by_leaf_names(arg, keep)
arg.set_ancestral()
arg.prune()
model = arghmm.ArgHmm(arg, seqs, new_name="n2", times=times,
rho=rho, mu=mu)
print "states", len(model.states[0])
print "muts", len(muts)
print "recomb", len(model.recomb_pos) - 2, model.recomb_pos[1:-1]
p.plot(model.recomb_pos, [1000] * len(model.recomb_pos),
style="points")
probs = arghmm.get_posterior_probs(model, length, verbose=True)
high = list(arghmm.iter_posterior_times(model, probs, .95))
low = list(arghmm.iter_posterior_times(model, probs, .05))
p.plot(high, style="lines")
p.plot(low, style="lines")
pause()
def test_post2(self):
k = 2
n = 1e4
rho = 1.5e-8 * 10
mu = 2.5e-8 * 10
length = 10000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
print "muts", len(muts)
times = arghmm.get_time_points()
arghmm.discretize_arg(arg, times)
thread = list(arghmm.iter_chrom_thread(arg, arg["n1"], by_block=False))
tree = arg.get_marginal_tree(0)
print tree.root.age
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
p = plot(cget(thread, 1), style="lines", ymin=0)
#alignlib.print_align(seqs)
# remove chrom
keep = ["n0"]
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
model = arghmm.ArgHmm(arg,
seqs,
new_name="n1",
times=times,
rho=rho,
mu=mu)
print "states", len(model.states[0])
probs = arghmm.get_posterior_probs(model, length, verbose=True)
high = list(arghmm.iter_posterior_times(model, probs, .95))
low = list(arghmm.iter_posterior_times(model, probs, .05))
p.plot(high, style="lines")
p.plot(low, style="lines")
pause()
def test_post2(self):
k = 2
n = 1e4
rho = 1.5e-8 * 10
mu = 2.5e-8 * 10
length = 10000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
print "muts", len(muts)
times = arghmm.get_time_points()
arghmm.discretize_arg(arg, times)
thread = list(arghmm.iter_chrom_thread(arg, arg["n1"], by_block=False))
tree = arg.get_marginal_tree(0)
print tree.root.age
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
p = plot(cget(thread, 1), style="lines", ymin=0)
#alignlib.print_align(seqs)
# remove chrom
keep = ["n0"]
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
model = arghmm.ArgHmm(arg, seqs, new_name="n1", times=times,
rho=rho, mu=mu)
print "states", len(model.states[0])
probs = arghmm.get_posterior_probs(model, length, verbose=True)
high = list(arghmm.iter_posterior_times(model, probs, .95))
low = list(arghmm.iter_posterior_times(model, probs, .05))
p.plot(high, style="lines")
p.plot(low, style="lines")
pause()
def test_post(self):
k = 6
n = 1e4
rho = 1.5e-8 * 10
mu = 2.5e-8 * 10
length = 10000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
print "muts", len(muts)
print "recombs", len(arglib.get_recomb_pos(arg))
times = arghmm.get_time_points(ntimes=10)
arghmm.discretize_arg(arg, times)
tree = arg.get_marginal_tree(0)
treelib.draw_tree_names(tree.get_tree(), minlen=5, scale=4e-4)
# remove chrom
new_name = "n%d" % (k - 1)
keep = set(arg.leaf_names()) - set([new_name])
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
model = arghmm.ArgHmm(arg,
seqs,
new_name=new_name,
times=times,
rho=rho,
mu=mu)
print "states", len(model.states[0])
probs = arghmm.get_posterior_probs(model, length, verbose=True)
for pcol in probs:
p = sum(map(exp, pcol))
print p, " ".join("%.3f" % f for f in map(exp, pcol))
fequal(p, 1.0, rel=1e-2)
def test_determ(self):
k = 8
n = 1e4
rho = 1.5e-8
mu = 2.5e-8
length = 100000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
times = arghmm.get_time_points(maxtime=50000, ntimes=20)
arghmm.discretize_arg(arg, times)
new_name = "n%d" % (k - 1)
thread = list(
arghmm.iter_chrom_thread(arg, arg[new_name], by_block=False))
thread_clades = list(
arghmm.iter_chrom_thread(arg,
arg[new_name],
by_block=True,
use_clades=True))
# remove chrom
keep = ["n%d" % i for i in range(k - 1)]
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
model = arghmm.ArgHmm(arg,
seqs,
new_name=new_name,
times=times,
rho=rho,
mu=mu)
for i, rpos in enumerate(model.recomb_pos[1:-1]):
pos = rpos + 1
model.check_local_tree(pos, force=True)
#recomb = arghmm.find_tree_next_recomb(arg, pos - 1)
tree = arg.get_marginal_tree(pos - .5)
last_tree = arg.get_marginal_tree(pos - 1 - .5)
states1 = model.states[pos - 1]
states2 = model.states[pos]
(recomb_branch, recomb_time), (coal_branch, coal_time) = \
arghmm.find_recomb_coal(tree, last_tree, pos=rpos)
recomb_time = times.index(recomb_time)
coal_time = times.index(coal_time)
determ = arghmm.get_deterministic_transitions(
states1, states2, times, tree, last_tree, recomb_branch,
recomb_time, coal_branch, coal_time)
leaves1, time1, block1 = thread_clades[i]
leaves2, time2, block2 = thread_clades[i + 1]
if new_name in leaves1:
leaves1.remove(new_name)
if new_name in leaves2:
leaves2.remove(new_name)
node1 = arghmm.arg_lca(arg, leaves1, None, pos - 1).name
node2 = arghmm.arg_lca(arg, leaves2, None, pos).name
state1 = (node1, times.index(time1))
state2 = (node2, times.index(time2))
print pos, state1, state2
try:
statei1 = states1.index(state1)
statei2 = states2.index(state2)
except:
print "states1", states1
print "states2", states2
raise
statei3 = determ[statei1]
print " ", statei1, statei2, statei3, states2[statei3]
if statei2 != statei3 and statei3 != -1:
tree = tree.get_tree()
treelib.remove_single_children(tree)
last_tree = last_tree.get_tree()
treelib.remove_single_children(last_tree)
print "block1", block1
print "block2", block2
print "r=", (recomb_branch, recomb_time)
print "c=", (coal_branch, coal_time)
print "tree"
treelib.draw_tree_names(tree, minlen=8, maxlen=8)
print "last_tree"
treelib.draw_tree_names(last_tree, minlen=8, maxlen=8)
assert False
def test_determ(self):
k = 8
n = 1e4
rho = 1.5e-8
mu = 2.5e-8
length = 100000
arg = arglib.sample_arg(k, n, rho, start=0, end=length)
muts = arglib.sample_arg_mutations(arg, mu)
seqs = arglib.make_alignment(arg, muts)
times = arghmm.get_time_points(maxtime=50000, ntimes=20)
arghmm.discretize_arg(arg, times)
new_name = "n%d" % (k - 1)
thread = list(arghmm.iter_chrom_thread(arg, arg[new_name],
by_block=False))
thread_clades = list(arghmm.iter_chrom_thread(
arg, arg[new_name], by_block=True, use_clades=True))
# remove chrom
keep = ["n%d" % i for i in range(k-1)]
arglib.subarg_by_leaf_names(arg, keep)
arg = arglib.smcify_arg(arg)
model = arghmm.ArgHmm(arg, seqs, new_name=new_name, times=times,
rho=rho, mu=mu)
for i, rpos in enumerate(model.recomb_pos[1:-1]):
pos = rpos + 1
model.check_local_tree(pos, force=True)
#recomb = arghmm.find_tree_next_recomb(arg, pos - 1)
tree = arg.get_marginal_tree(pos-.5)
last_tree = arg.get_marginal_tree(pos-1-.5)
states1 = model.states[pos-1]
states2 = model.states[pos]
(recomb_branch, recomb_time), (coal_branch, coal_time) = \
arghmm.find_recomb_coal(tree, last_tree, pos=rpos)
recomb_time = times.index(recomb_time)
coal_time = times.index(coal_time)
determ = arghmm.get_deterministic_transitions(
states1, states2, times,
tree, last_tree,
recomb_branch, recomb_time,
coal_branch, coal_time)
leaves1, time1, block1 = thread_clades[i]
leaves2, time2, block2 = thread_clades[i+1]
if new_name in leaves1:
leaves1.remove(new_name)
if new_name in leaves2:
leaves2.remove(new_name)
node1 = arghmm.arg_lca(arg, leaves1, None, pos-1).name
node2 = arghmm.arg_lca(arg, leaves2, None, pos).name
state1 = (node1, times.index(time1))
state2 = (node2, times.index(time2))
print pos, state1, state2
try:
statei1 = states1.index(state1)
statei2 = states2.index(state2)
except:
print "states1", states1
print "states2", states2
raise
statei3 = determ[statei1]
print " ", statei1, statei2, statei3, states2[statei3]
if statei2 != statei3 and statei3 != -1:
tree = tree.get_tree()
treelib.remove_single_children(tree)
last_tree = last_tree.get_tree()
treelib.remove_single_children(last_tree)
print "block1", block1
print "block2", block2
print "r=", (recomb_branch, recomb_time)
print "c=", (coal_branch, coal_time)
print "tree"
treelib.draw_tree_names(tree, minlen=8, maxlen=8)
print "last_tree"
treelib.draw_tree_names(last_tree, minlen=8, maxlen=8)
assert False