def single_chain_run():
basic_chain(start_x=starting_trees[0],
summaries=summaries,
posterior_function=posterior,
proposal=mp[0],
post=None,
N=sum(sim_lengths),
sample_verbose_scheme=summary_verbose_scheme[0],
overall_thinning=int(options.thinning_coef),
i_start_from=0,
temperature=1.0,
proposal_update=None,
multiplier=multiplier,
appending_result_file=options.result_file,
appending_result_frequency=sim_lengths[0])
def test_prior_model_no_admixes(start_tree,
sim_length=100000,
summaries=None,
thinning_coef=1):
posterior = initialize_prior_as_posterior()
if summaries is None:
summaries = [
s_variable('posterior'),
s_variable('mhr'),
s_no_admixes()
]
proposal = adaptive_proposal_no_admix()
#proposal.props=proposal.props[2:] #a little hack under the hood
#proposal.params=proposal.params[2:] #a little hack under the hood.
sample_verbose_scheme = {summary.name: (1, 0) for summary in summaries}
final_tree, final_posterior, results, _ = basic_chain(
start_tree,
summaries,
posterior,
proposal,
post=None,
N=sim_length,
sample_verbose_scheme=sample_verbose_scheme,
overall_thinning=int(thinning_coef + sim_length / 60000),
i_start_from=0,
temperature=1.0,
proposal_update=None,
check_trees=False)
save_to_csv(results, summaries)
return results
def test_prior_model(start_tree,
sim_length=100000,
summaries=None,
thinning_coef=1):
posterior = initialize_prior_as_posterior()
if summaries is None:
summaries = [
s_variable('posterior'),
s_variable('mhr'),
s_no_admixes()
]
proposal = adaptive_proposal() #basic_meta_proposal()
sample_verbose_scheme = {summary.name: (1, 0) for summary in summaries}
sample_verbose_scheme['posterior'] = (1, 1000)
final_tree, final_posterior, results, _ = basic_chain(
start_tree,
summaries,
posterior,
proposal,
post=None,
N=sim_length,
sample_verbose_scheme=sample_verbose_scheme,
overall_thinning=int(thinning_coef + sim_length / 60000),
i_start_from=0,
temperature=1.0,
proposal_update=None,
check_trees=True)
print results
save_to_csv(results, summaries)
return results
def run_chain(self, p):
start_tree, post, N, sample_verbose_scheme, overall_thinning, i_start_from, temperature, proposal_update, multiplier = p
#print 'run_chain proposal.node_naming.n', self.proposal.node_naming.n
return basic_chain(start_tree, self.summaries, self.posterior_function,
self.proposal, post, N, sample_verbose_scheme,
overall_thinning, i_start_from, temperature,
proposal_update, multiplier)
def mcmc_search(likelihood,
pops,
generations,
short_to_full,
init=None,
N=1000):
'''
This function wraps the MCMC call, and takes the MCMC output and finds the maximum posterior value.
'''
if init is None:
init = choice(pops, 2**generations)
init = ''.join(map(str, init))
summaries = [basic_summary('posterior'), basic_summary('x')]
summary_verbose_scheme = {summary.name: (1, 1) for summary in summaries}
proposal_function = MetaProposal(pops=pops)
_, _, s = basic_chain(init,
summaries,
likelihood,
proposal_function,
post=None,
N=N,
sample_verbose_scheme=summary_verbose_scheme,
overall_thinning=1,
i_start_from=0,
temperature=1.0)
res = list(s)
#print(res)
#we know that the first column is iteration number, the second posterior, and the last is the configuration x.
res_dic = {
find_smallest_equivalence_class(x): p
for x, p in zip(res[2], res[1])
}
xs, posteriors = map(list, list(zip(*list(res_dic.items()))))
list_sorted = sorted(enumerate(posteriors), key=lambda elem: elem[1])
i = list_sorted[-1][0]
#print(xs[i], posteriors[i])
x_to_return = []
p_to_return = []
for i_tup in list_sorted[-10:][::-1]:
i = i_tup[0]
x_to_return.append(xs[i])
p_to_return.append(posteriors[i])
return list(zip(x_to_return, p_to_return))
def mcmc(observed_covariance, df, outgroup=False):
nodes=['s'+str(i+1) for i in range(observed_covariance.shape[0])]
start_x=identifier_to_tree_clean(simulate_tree(4,0)),0
summaries=[summary.s_posterior(),
summary.s_basic_tree_statistics(tree_statistics.unique_identifier_and_branch_lengths, 'tree', output='string'),
summary.s_variable('add', output='double'),
summary.s_no_admixes(),]
options=options_object(outgroup)
proposal=make_proposal(options)[0]
posterior_function=posterior_class(observed_covariance, M=df, nodes=nodes)
sample_verbose_scheme={'posterior':(1,200), 'tree':(1,0),'add':(1,200),'no_admixes':(1,200)}
a=basic_chain(start_x, summaries, posterior_function, proposal, post=None, N=5000,
sample_verbose_scheme=sample_verbose_scheme, overall_thinning=100, i_start_from=0,
temperature=1.0, proposal_update=None, multiplier=None, check_trees=False,
appending_result_file=None, appending_result_frequency=10)
return a[2]
def func(nstart_tree):
n, start_tree = nstart_tree
final_tree, final_posterior, results, _ = basic_chain(
start_tree,
summaries,
posterior,
proposal,
post=None,
N=sim_length,
sample_verbose_scheme=sample_verbose_scheme,
overall_thinning=int(thinning_coef + sim_length / 60000),
i_start_from=0,
temperature=1.0,
proposal_update=None,
check_trees=True)
save_to_csv(results,
summaries,
filename='results_' + str(n + 1) + 'csv',
origin_layer=(n + 1, 1))
def trivial_simulation(start_val, reps, thinning_coef=1):
posterior = initialize_trivial_posterior()
summaries = [Trivial_Summary()]
proposal = trivial_proposal()
sample_verbose_scheme = {summary.name: (1, 0) for summary in summaries}
final_tree, final_posterior, results, _ = basic_chain(
start_val,
summaries,
posterior,
proposal,
post=None,
N=reps,
sample_verbose_scheme=sample_verbose_scheme,
overall_thinning=int(thinning_coef + reps / 60000),
i_start_from=0,
temperature=1.0,
proposal_update=None,
check_trees=False)
print results
save_to_csv(results, summaries)
return results
def test_posterior_model(true_tree=None,
start_tree=None,
sim_length=100000,
summaries=None,
thinning_coef=19,
admixtures_of_true_tree=None,
no_leaves_true_tree=4,
filename='results.csv',
sim_from_wishart=False,
wishart_df=None,
sap_sim=False,
sap_ana=False,
resimulate_regrafted_branch_length=False,
emp_cov=None,
big_posterior=False,
rescale_empirical_cov=False):
if true_tree is None:
if admixtures_of_true_tree is None:
admixtures_of_true_tree = geom.rvs(p=0.5) - 1
true_tree = generate_phylogeny(no_leaves_true_tree,
admixtures_of_true_tree,
skewed_admixture_prior=sap_sim)
else:
no_leaves_true_tree = get_no_leaves(true_tree)
admixtures_of_true_tree = get_number_of_admixes(true_tree)
true_x = (true_tree, 0)
m = make_covariance(true_tree, get_trivial_nodes(no_leaves_true_tree))
if start_tree is None:
start_tree = true_tree
start_x = (start_tree, 0)
if wishart_df is None:
wishart_df = n_mark(m)
if sim_from_wishart:
r = m.shape[0]
print m
m = wishart.rvs(df=r * wishart_df - 1, scale=m / (r * wishart_df))
print m
if emp_cov is not None:
m = emp_cov
if big_posterior:
posterior = initialize_big_posterior(m,
wishart_df,
use_skewed_distr=sap_ana)
else:
posterior = initialize_posterior(m,
wishart_df,
use_skewed_distr=sap_ana,
rescale=rescale_empirical_cov)
print 'true_tree=', unique_identifier_and_branch_lengths(true_tree)
post_ = posterior(true_x)
print 'likelihood(true_tree)', post_[0]
print 'prior(true_tree)', post_[1]
print 'posterior(true_tree)', sum(post_[:2])
if summaries is None:
summaries = [s_posterior(), s_variable('mhr'), s_no_admixes()]
proposal = adaptive_proposal(
resimulate_regrafted_branch_length=resimulate_regrafted_branch_length)
#proposal.props=proposal.props[2:] #a little hack under the hood
#proposal.params=proposal.params[2:] #a little hack under the hood.
sample_verbose_scheme = {summary.name: (1, 0) for summary in summaries}
sample_verbose_scheme['posterior'] = (1, 1)
sample_verbose_scheme['no_admixes'] = (1, 1)
final_tree, final_posterior, results, _ = basic_chain(
start_x,
summaries,
posterior,
proposal,
post=None,
N=sim_length,
sample_verbose_scheme=sample_verbose_scheme,
overall_thinning=int(max(thinning_coef, sim_length / 60000)),
i_start_from=0,
temperature=1.0,
proposal_update=None,
check_trees=False)
save_to_csv(results, summaries, filename=filename)
return true_tree
def _basic_chain_unpacker(args):
return basic_chain(*args)