def tajimas_d(char_matrix, ignore_uncertain=True):
deprecate.dendropy_deprecation_warning(
preamble="The 'dendropy.popgenstat' module has moved to 'dendropy.calculate.popgenstat'.",
old_construct="from dendropy import popgenstat\npopgenstat.tajimas_d(...)",
new_construct="from dendropy.calculate import popgenstat\npopgenstat.tajimas_d(...)",
)
return popgenstat.tajimas_d(char_matrix=char_matrix, ignore_uncertain=ignore_uncertain)
def tajimas_d(char_matrix, ignore_uncertain=True):
deprecate.dendropy_deprecation_warning(
preamble=
"The 'dendropy.popgenstat' module has moved to 'dendropy.calculate.popgenstat'.",
old_construct=
"from dendropy import popgenstat\npopgenstat.tajimas_d(...)",
new_construct=
"from dendropy.calculate import popgenstat\npopgenstat.tajimas_d(...)",
)
return popgenstat.tajimas_d(char_matrix=char_matrix,
ignore_uncertain=ignore_uncertain)
def test_tajimas_d(self):
self.assertAlmostEqual(popgenstat.tajimas_d(self.data, ignore_uncertain=True), 1.12467, 4)
def testTajimasD(self):
self.assertAlmostEqual(popgenstat.tajimas_d(self.matrix), -1.44617198561, 4)
def test_tajimas_d(self):
self.assertAlmostEqual(
popgenstat.tajimas_d(self.data, ignore_uncertain=True), 1.12467, 4)
def testTajimasD(self):
self.assertAlmostEqual(popgenstat.tajimas_d(self.matrix),
-1.44617198561, 4)
def testTajimasD_with_missing(self):
self.assertAlmostEqual(popgenstat.tajimas_d(self.matrix_with_missing), -1.44617198561, 4)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("source_trees",
metavar="SOURCE_TREEFILE [SOURCE_TREEFILE [SOURCE_TREEFILE]]",
nargs="+",
help="Path to containing tree files. Specify '-' to read from standard input.")
parser.add_argument("-f", "--input-format",
default="nexus",
dest="schema",
help="Input trees format (default: $(default)s).")
parser.add_argument("-z", "--random-seed",
type=int,
default=None,
help="Seed for random number generator engine.")
parser.add_argument("-t", "--title",
default="bpprun",
help="Run title (default: '%(default)s')")
data_options = parser.add_argument_group("Data Options")
data_options.add_argument("--population-size",
type=int,
default=1.0,
help="Population size (default: %(default)s).")
data_options.add_argument("--num-individuals-per-population",
type=int,
default=4,
help="Number of individuals sampled per incipient species lineage (default: %(default)s).")
data_options.add_argument("--num-loci-per-individual",
type=int,
default=10,
help="Number of loci sampled per individual (default: %(default)s).")
data_options.add_argument("--num-characters-per-locus",
type=int,
default=1000,
help="Number of characters sampled per locus (default: %(default)s).")
data_options.add_argument("--mutation-rate-per-site",
type=float,
# default=0.00001,
default=1e-8,
help="Per-site mutation rate (default: %(default)s).")
parser.add_argument("--no-scale-tree-by-mutation-rate",
action="store_true",
help="Do not scale tree by mutation rate.")
args = parser.parse_args()
if args.random_seed is None:
random_seed = random.randint(0, sys.maxsize-1)
else:
random_seed = args.random_seed
rng = random.Random(random_seed)
_log("Random seed: {}".format(random_seed))
sg = seqgen.SeqGen()
sg.seq_len = args.num_characters_per_locus
sg.scale_branch_lens = args.mutation_rate_per_site
if "-" in args.source_trees:
filepaths = sys.stdin.read().split("\n")
args.source_trees.remove("-")
else:
filepaths = []
manifest_entries = []
filepaths.extend(args.source_trees)
for idx, filepath in enumerate(filepaths):
job_title = "{}_{:05d}".format(args.title, idx+1)
manifest_entry = collections.OrderedDict()
_log("{} of {}: {}: {}".format(idx+1, len(filepaths), job_title, filepath))
source_tree = dendropy.Tree.get(
path=filepath,
schema=args.schema,
extract_comment_metadata=True,
preserve_underscores=True,
)
manifest_entry["speciation_initiation_from_orthospecies_rate"] = try_to_coerce_to_float(source_tree.annotations["speciation_initiation_from_orthospecies_rate"].value)
manifest_entry["speciation_initiation_from_incipient_species_rate"] = try_to_coerce_to_float(source_tree.annotations["speciation_initiation_from_incipient_species_rate"].value)
manifest_entry["speciation_completion_rate"] = try_to_coerce_to_float(source_tree.annotations["speciation_completion_rate"].value)
manifest_entry["orthospecies_extinction_rate"] = try_to_coerce_to_float(source_tree.annotations["orthospecies_extinction_rate"].value)
manifest_entry["incipient_species_extinction_rate"] = try_to_coerce_to_float(source_tree.annotations["incipient_species_extinction_rate"].value)
manifest_entry["max_time"] = try_to_coerce_to_float(source_tree.annotations["max_time"].value)
manifest_entry["max_extant_orthospecies"] = try_to_coerce_to_float(source_tree.annotations["max_extant_orthospecies"].value)
manifest_entry["num_extant_lineages"] = try_to_coerce_to_float(source_tree.annotations["num_extant_lineages"].value)
manifest_entry["num_extant_orthospecies"] = try_to_coerce_to_float(source_tree.annotations["num_extant_orthospecies"].value)
manifest_entry["source_tree_type"] = source_tree.annotations["tree_type"].value
manifest_entry["population_size"] = args.population_size
manifest_entry["num_individuals_per_population"] = args.num_individuals_per_population
manifest_entry["num_loci_per_individual"] = args.num_loci_per_individual
manifest_entry["mutation_rate_per_site"] = args.mutation_rate_per_site
source_tree.calc_node_ages()
gene_trees = generate_contained_trees(
containing_tree=source_tree,
num_individuals_per_population=args.num_individuals_per_population,
num_gene_trees=args.num_loci_per_individual,
population_size=args.population_size,
rng=rng,
)
imap_filepath = "{}.input.imap.txt".format(job_title)
f = open(imap_filepath, "w")
for taxon in gene_trees.taxon_namespace:
f.write("{} {}\n".format(taxon.label.split("^")[1], taxon.population_label))
# f.write("{} {}\n".format(taxon.label.split("^")[0], taxon.population_label))
# f.write("{} {}\n".format(taxon.label, taxon.population_label))
f.write("\n")
d0 = sg.generate(gene_trees)
chars_filepath = "{}.input.chars.txt".format(job_title)
f = open(chars_filepath, "w")
for cm_idx, cm in enumerate(d0.char_matrices):
sys.stderr.write("Locus {}: pi = {}, Tajima's D = {}\n".format(
cm_idx+1,
popgenstat.nucleotide_diversity(cm),
popgenstat.tajimas_d(cm)))
cm.write(file=f, schema="phylip")
f.write("\n")
out_filepath = "{}.results.out.txt".format(job_title)
mcmc_filepath = "{}.results.mcmc.txt".format(job_title)
num_species = len(source_tree.taxon_namespace)
species_labels = " ".join(t.label for t in source_tree.taxon_namespace)
num_individuals_per_species = " ".join(str(args.num_individuals_per_population) for i in range(len(source_tree.taxon_namespace)))
# Inverse Gamma Prior
# IG(a,b), with mean given by b/(a-1)
# So,
# thetaprior 3 0.002
# has a mean of
# 0.002/(3-1) = 0.001
theta_prior_mean = args.population_size * 4 * args.mutation_rate_per_site
theta_prior_a = 3.0
theta_prior_b = theta_prior_mean * (theta_prior_a - 1)
if args.no_scale_tree_by_mutation_rate:
tau_prior_mean = source_tree.seed_node.age
else:
# tau_prior_mean = source_tree.seed_node.age * args.population_size * 4 * args.mutation_rate_per_site
tau_prior_mean = source_tree.seed_node.age * args.mutation_rate_per_site * (1.0 / (args.num_loci_per_individual * args.num_characters_per_locus))
# tau_prior_mean = source_tree.seed_node.age / 100000
tau_prior_a = 3.0
tau_prior_b = tau_prior_mean * (tau_prior_a - 1)
manifest_entry["num_input_lineages"] = len(species_labels)
manifest_entry["theta"] = theta_prior_mean
manifest_entry["theta_prior_a"] = theta_prior_a
manifest_entry["theta_prior_b"] = theta_prior_b
manifest_entry["root_age"] = source_tree.seed_node.age
manifest_entry["tau_prior_a"] = tau_prior_a
manifest_entry["tau_prior_b"] = tau_prior_b
species_tree = source_tree.as_string(
schema="newick",
suppress_leaf_taxon_labels=False,
suppress_leaf_node_labels=True,
suppress_internal_taxon_labels=True,
suppress_internal_node_labels=True,
suppress_rooting=True,
suppress_edge_lengths=True,
unquoted_underscores=True,
preserve_spaces=True,
store_tree_weights=False,
suppress_annotations=True,
suppress_item_comments=True,
)
bpp_config = BPP_TEMPLATE.format(
chars_filepath=chars_filepath,
imap_filepath=imap_filepath,
out_filepath=out_filepath,
mcmc_filepath=mcmc_filepath,
num_species=num_species,
species_labels=species_labels,
num_individuals_per_species=num_individuals_per_species,
species_tree=species_tree,
theta_prior_mean=theta_prior_mean,
theta_prior_a=theta_prior_a,
theta_prior_b=theta_prior_b,
tau_prior_mean=tau_prior_mean,
tau_prior_a=tau_prior_a,
tau_prior_b=tau_prior_b,
num_loci=args.num_loci_per_individual,
root_age=source_tree.seed_node.age
)
bpp_ctl_filepath = "{}.input.bpp.ctl".format(job_title)
f = open(bpp_ctl_filepath, "w")
f.write(bpp_config)
f.write("\n")
jobf = open("{}.job.sge".format(job_title), "w")
jobf.write("#! /bin/bash\n")
jobf.write("#$ -cwd\n")
jobf.write("#$ -V\n")
jobf.write("#$ -S /bin/bash\n")
jobf.write("#$ -l h_vmem=12G\n")
jobf.write("#$ -l virtual_free=12G\n")
jobf.write("bpp --cfile {}\n".format(bpp_ctl_filepath))
manifest_entry["source_tree_path"] = filepath
manifest_entry["results_filepath"] = out_filepath
manifest_entry["mcmc_filepath"] = mcmc_filepath
manifest_entries.append(manifest_entry)
out = _open_output_file_for_csv_writer(
filepath="{}_manifest.csv".format(args.title),
append=False)
with out:
writer = csv.DictWriter(
out,
fieldnames=manifest_entries[0].keys(),
restval="NA",
delimiter=",",
lineterminator=os.linesep,
)
writer.writeheader()
writer.writerows(manifest_entries)
def testTajimasD_with_missing(self):
self.assertAlmostEqual(popgenstat.tajimas_d(self.matrix_with_missing),
-1.44617198561, 4)
# -*- coding: utf-8 -*-
"run as python estimate_SFS_stats.py $inputfile"
"Appends to the given output file"
import dendropy
from dendropy.calculate import popgenstat
import os
import sys
seqs = dendropy.DnaCharacterMatrix.get(path=sys.argv[1], schema="fasta")
out = open("popgen_stats.txt", "a")
pop = sys.argv[1].split("/")[1].split("_")[2].split(".")[0]
gene = sys.argv[1].split("/")[1].split("_")[0]
fbtr = sys.argv[1].split("/")[1].split("_")[1]
td = popgenstat.tajimas_d(seqs)
tw = popgenstat.wattersons_theta(seqs)
tp = popgenstat.average_number_of_pairwise_differences(seqs)
ss = popgenstat.num_segregating_sites(seqs)
out.write("\t".join(
[str(pop),
str(gene),
str(fbtr),
str(td),
str(tw),
str(tp),
str(ss)]) + "\n")