def test_new_implementation_round_trip(self):
p = {
"c_shape": 2,
"c_scale": 5,
"theta_shape": 2.0,
"theta_scale": 0.002,
"tau_shape": 1.0,
"tau_scale": 5.0,
"a_theta_shape": 1,
"a_theta_scale": 0.01,
"theta_params": "001",
"migration_shape": 1.5,
"migration_scale": 0.1,
"recombination_shape": 0.0,
"recombination_scale": 0.0,
"psi": 0,
}
self._update_config(self.cfg, p, new_impl=True)
_LOG.debug("testing config:\n\n{0}\n".format(self.cfg.getvalue()))
c = MsBayesConfig(self.cfg)
s = StringIO()
c.write(s)
_LOG.debug("written config:\n\n{0}\n".format(s.getvalue()))
s.seek(0)
c2 = MsBayesConfig(s)
self.assertSameConfigs([c, c2])
def test_new_implementation_round_trip(self):
p = {
'c_shape': 2,
'c_scale': 5,
'theta_shape': 2.0,
'theta_scale': 0.002,
'tau_shape': 1.0,
'tau_scale': 5.0,
'a_theta_shape': 1,
'a_theta_scale': 0.01,
'theta_params': '001',
'migration_shape': 1.5,
'migration_scale': 0.1,
'recombination_shape': 0.0,
'recombination_scale': 0.0,
'psi': 0,
}
self._update_config(self.cfg, p, new_impl=True)
_LOG.debug('testing config:\n\n{0}\n'.format(self.cfg.getvalue()))
c = MsBayesConfig(self.cfg)
s = StringIO()
c.write(s)
_LOG.debug('written config:\n\n{0}\n'.format(s.getvalue()))
s.seek(0)
c2 = MsBayesConfig(s)
self.assertSameConfigs([c, c2])
def test_multi_locus_round_trip(self):
p = {"ltheta": 0.0001, "utheta": 0.1, "utau": 10.0, "psi": 0, "umig": 0.0, "urec": 0.0, "atheta": 1.0}
self._update_config(self.cfg, p, multi_locus=True)
_LOG.debug("testing config:\n\n{0}\n".format(self.cfg.getvalue()))
c = MsBayesConfig(self.cfg)
s = StringIO()
c.write(s)
_LOG.debug("written config:\n\n{0}\n".format(s.getvalue()))
s.seek(0)
c2 = MsBayesConfig(s)
self.assertSameConfigs([c, c2])
def test_multi_locus_round_trip(self):
p = {
'ltheta': 0.0001,
'utheta': 0.1,
'utau': 10.0,
'psi': 0,
'umig': 0.0,
'urec': 0.0,
'atheta': 1.0,
}
self._update_config(self.cfg, p, multi_locus=True)
_LOG.debug('testing config:\n\n{0}\n'.format(self.cfg.getvalue()))
c = MsBayesConfig(self.cfg)
s = StringIO()
c.write(s)
_LOG.debug('written config:\n\n{0}\n'.format(s.getvalue()))
s.seek(0)
c2 = MsBayesConfig(s)
self.assertSameConfigs([c, c2])
def main_cli():
description = '{name} {version}'.format(**_program_info)
parser = argparse.ArgumentParser(description = description)
parser.add_argument('-c', '--config',
type = arg_is_config,
required = True,
help = ('msBayes config file to be used to generate saturation '
'plot.'))
parser.add_argument('-n', '--num-prior-samples',
action = 'store',
type = int,
default = 1000,
help = ('The number of prior samples to simulate for the '
'saturation plot.'))
parser.add_argument('--np',
action = 'store',
type = int,
default = multiprocessing.cpu_count(),
help = ('The maximum number of processes to run in parallel. The '
'default is the number of CPUs available on the machine.'))
parser.add_argument('-o', '--output-dir',
action = 'store',
type = arg_is_dir,
help = ('The directory in which all output files will be written. '
'The default is to use the directory of the first observed '
'config file.'))
parser.add_argument('--temp-dir',
action = 'store',
type = arg_is_dir,
help = ('A directory to temporarily stage files. The default is to '
'use the output directory.'))
parser.add_argument('-s', '--stat-prefixes',
nargs = '*',
type = str,
default = ['pi', 'pi.net', 'wattTheta', 'tajD.denom'],
help = ('Prefixes of summary statistics to use in the analyses. '
'The prefixes should be separated by spaces. '
'Default: `-s pi pi.net wattTheta tajD.denom`.'))
parser.add_argument('--vertical-lines',
nargs = '*',
type = float,
default = [],
help = ('Positions along x-axis where vertical lines are to be '
'drawn. Default is to draw no vertical lines.'))
parser.add_argument('--compress',
action = 'store_true',
help = 'Compress plot data file.')
parser.add_argument('--keep-temps',
action = 'store_true',
help = 'Keep all temporary files.')
parser.add_argument('--seed',
action = 'store',
type = int,
help = 'Random number seed to use for the analysis.')
parser.add_argument('--version',
action = 'version',
version = '%(prog)s ' + _program_info['version'],
help = 'Report version and exit.')
parser.add_argument('--quiet',
action = 'store_true',
help = 'Run without verbose messaging.')
parser.add_argument('--debug',
action = 'store_true',
help = 'Run in debugging mode.')
args = parser.parse_args()
##########################################################################
## handle args
from pymsbayes.utils.messaging import (LoggingControl,
InfoLogger)
LoggingControl.set_logging_level("INFO")
if args.quiet:
LoggingControl.set_logging_level("WARNING")
if args.debug:
LoggingControl.set_logging_level("DEBUG")
log = LoggingControl.get_logger(__name__)
from pymsbayes.workers import MsBayesWorker
from pymsbayes.utils.parsing import (get_patterns_from_prefixes,
DEFAULT_STAT_PATTERNS, get_stats_by_time, dict_line_iter)
from pymsbayes.manager import Manager
from pymsbayes.utils.tempfs import TempFileSystem
from pymsbayes.utils import probability
from pymsbayes.utils.functions import long_division
from pymsbayes.config import MsBayesConfig
from pymsbayes.utils import GLOBAL_RNG, MSBAYES_SORT_INDEX, ToolPathManager
from pymsbayes.fileio import process_file_arg
from pymsbayes.plotting import MATPLOTLIB_AVAILABLE, SaturationPlotGrid
MSBAYES_SORT_INDEX.set_index(0)
# get full paths to tools
msbayes_path = ToolPathManager.get_tool_full_path('msbayes.pl')
dpp_msbayes_path = ToolPathManager.get_tool_full_path('dpp-msbayes.pl')
if not args.output_dir:
args.output_dir = os.path.dirname(args.config)
info = InfoLogger(os.path.join(args.output_dir, 'pymsbayes-info.txt'))
stats_by_time_path = os.path.join(args.output_dir, 'stats-by-time.txt')
if args.compress:
stats_by_time_path += '.gz'
plot_path = os.path.join(args.output_dir, 'saturation-plot.pdf')
if not args.temp_dir:
args.temp_dir = args.output_dir
temp_fs = TempFileSystem(parent=args.temp_dir, prefix='temp-files-')
args.stat_prefixes = [s.rstrip('.') for s in args.stat_prefixes]
stat_patterns = get_patterns_from_prefixes(
[s + '.' for s in args.stat_prefixes],
ignore_case=True)
if not args.seed:
args.seed = random.randint(1, 999999999)
GLOBAL_RNG.seed(args.seed)
compress_level = None
if args.compress:
compress_level = 9
cfg = MsBayesConfig(args.config)
num_taxon_pairs = cfg.npairs
cfg.div_model_prior = 'constrained'
cfg.psi = probability.DiscreteUniformDistribution(num_taxon_pairs,
num_taxon_pairs)
config_path = temp_fs.get_file_path(prefix='cfg-')
cfg.write(config_path)
info.write('[pymsbayes]', log.info)
info.write('\tprogram_name = {name}'.format(**_program_info), log.info)
info.write('\tversion = {version}'.format(**_program_info), log.info)
info.write('\tinvocation = {0!r}'.format(' '.join(sys.argv)), log.info)
info.write('\toutput_directory = {0!r}'.format(args.output_dir), log.info)
info.write('\ttemp_directory = {0!r}'.format(temp_fs.base_dir), log.info)
info.write('\tsort_index = {0}'.format(
MSBAYES_SORT_INDEX.current_value()), log.info)
info.write('\tstat_patterns = {0!r}'.format(
', '.join([p.pattern for p in stat_patterns])), log.info)
info.write('\tseed = {0}'.format(args.seed), log.info)
info.write('\tnum_prior_samples = {0}'.format(args.num_prior_samples),
log.info)
info.write('\tstats_by_time_path = {0!r}'.format(stats_by_time_path),
log.info)
info.write('\t[[tool_paths]]', log.info)
info.write('\t\tdpp_msbayes = {0}'.format(dpp_msbayes_path), log.info)
info.write('\t\tmsbayes = {0}'.format(msbayes_path), log.info)
info.write('\t[[config]]', log.debug)
info.write('{0}'.format(str(cfg)), log.debug)
##########################################################################
## begin analysis --- generate samples
start_time = datetime.datetime.now()
if args.np > args.num_prior_samples:
args.np = args.num_prior_samples
batch_size, remainder = long_division(args.num_prior_samples, args.np)
schema = 'abctoolbox'
workers = []
for i in range(args.np):
sample_size = batch_size
if i == (args.np - 1):
sample_size += remainder
w = MsBayesWorker(
temp_fs = temp_fs,
sample_size = sample_size,
config_path = config_path,
report_parameters = True,
schema = schema,
include_header = True,
stat_patterns = stat_patterns,
write_stats_file = False)
workers.append(w)
log.info('Generating samples...')
workers = Manager.run_workers(
workers = workers,
num_processors = args.np)
log.info('Parsing samples...')
stats_by_time = get_stats_by_time([w.prior_path for w in workers])
stat_keys = stats_by_time.keys()
stat_keys.remove('PRI.t')
for prefix in args.stat_prefixes:
if not prefix in stat_keys:
raise Exception('stat prefix {0!r} not found in simulated stats:'
'\n\t{1}'.format(prefix, ', '.join(stat_keys)))
header = ['PRI.t'] + args.stat_prefixes
log.info('Writing stats-by-time matrix...')
out, close = process_file_arg(stats_by_time_path, 'w',
compresslevel = compress_level)
for row in dict_line_iter(stats_by_time, sep = '\t', header = header):
out.write(row)
if close:
out.close()
log.info('Creating plots...')
if not MATPLOTLIB_AVAILABLE:
log.warning(
'`matplotlib` could not be imported, so the plot can not be\n'
'produced. The data to create the plot can be found in:\n\t'
'{0!r}'.format(stats_by_time_path))
else:
y_labels = {'pi': r'$\pi$',
'pi.net': r'$\pi_{net}$',
'wattTheta': r'$\theta_W$',
'tajD.denom': r'$SD(\pi - \theta_W)$'}
spg = SaturationPlotGrid(stats_by_time,
x_key = 'PRI.t',
y_keys = args.stat_prefixes,
y_labels = y_labels,
num_columns = 2,
vertical_line_positions = args.vertical_lines)
fig = spg.create_grid()
fig.savefig(plot_path)
stop_time = datetime.datetime.now()
log.info('Done!')
info.write('\t[[run_stats]]', log.info)
info.write('\t\tstart_time = {0}'.format(str(start_time)), log.info)
info.write('\t\tstop_time = {0}'.format(str(stop_time)), log.info)
info.write('\t\ttotal_duration = {0}'.format(str(stop_time - start_time)),
log.info)
if not args.keep_temps:
log.debug('purging temps...')
temp_fs.purge()
def main_cli():
description = '{name} {version}'.format(**_program_info)
parser = argparse.ArgumentParser(description=description)
parser.add_argument(
'-c',
'--config',
type=arg_is_config,
required=True,
help=('msBayes config file to be used to generate saturation '
'plot.'))
parser.add_argument(
'-n',
'--num-prior-samples',
action='store',
type=int,
default=1000,
help=('The number of prior samples to simulate for the '
'saturation plot.'))
parser.add_argument(
'--np',
action='store',
type=int,
default=multiprocessing.cpu_count(),
help=('The maximum number of processes to run in parallel. The '
'default is the number of CPUs available on the machine.'))
parser.add_argument(
'-o',
'--output-dir',
action='store',
type=arg_is_dir,
help=('The directory in which all output files will be written. '
'The default is to use the directory of the first observed '
'config file.'))
parser.add_argument(
'--temp-dir',
action='store',
type=arg_is_dir,
help=('A directory to temporarily stage files. The default is to '
'use the output directory.'))
parser.add_argument(
'-s',
'--stat-prefixes',
nargs='*',
type=str,
default=['pi', 'pi.net', 'wattTheta', 'tajD.denom'],
help=('Prefixes of summary statistics to use in the analyses. '
'The prefixes should be separated by spaces. '
'Default: `-s pi pi.net wattTheta tajD.denom`.'))
parser.add_argument(
'--vertical-lines',
nargs='*',
type=float,
default=[],
help=('Positions along x-axis where vertical lines are to be '
'drawn. Default is to draw no vertical lines.'))
parser.add_argument('--compress',
action='store_true',
help='Compress plot data file.')
parser.add_argument('--keep-temps',
action='store_true',
help='Keep all temporary files.')
parser.add_argument('--seed',
action='store',
type=int,
help='Random number seed to use for the analysis.')
parser.add_argument('--version',
action='version',
version='%(prog)s ' + _program_info['version'],
help='Report version and exit.')
parser.add_argument('--quiet',
action='store_true',
help='Run without verbose messaging.')
parser.add_argument('--debug',
action='store_true',
help='Run in debugging mode.')
args = parser.parse_args()
##########################################################################
## handle args
from pymsbayes.utils.messaging import (LoggingControl, InfoLogger)
LoggingControl.set_logging_level("INFO")
if args.quiet:
LoggingControl.set_logging_level("WARNING")
if args.debug:
LoggingControl.set_logging_level("DEBUG")
log = LoggingControl.get_logger(__name__)
from pymsbayes.workers import MsBayesWorker
from pymsbayes.utils.parsing import (get_patterns_from_prefixes,
DEFAULT_STAT_PATTERNS,
get_stats_by_time, dict_line_iter)
from pymsbayes.manager import Manager
from pymsbayes.utils.tempfs import TempFileSystem
from pymsbayes.utils import probability
from pymsbayes.utils.functions import long_division
from pymsbayes.config import MsBayesConfig
from pymsbayes.utils import GLOBAL_RNG, MSBAYES_SORT_INDEX, ToolPathManager
from pymsbayes.fileio import process_file_arg
from pymsbayes.plotting import MATPLOTLIB_AVAILABLE, SaturationPlotGrid
MSBAYES_SORT_INDEX.set_index(0)
# get full paths to tools
msbayes_path = ToolPathManager.get_tool_full_path('msbayes.pl')
dpp_msbayes_path = ToolPathManager.get_tool_full_path('dpp-msbayes.pl')
if not args.output_dir:
args.output_dir = os.path.dirname(args.config)
info = InfoLogger(os.path.join(args.output_dir, 'pymsbayes-info.txt'))
stats_by_time_path = os.path.join(args.output_dir, 'stats-by-time.txt')
if args.compress:
stats_by_time_path += '.gz'
plot_path = os.path.join(args.output_dir, 'saturation-plot.pdf')
if not args.temp_dir:
args.temp_dir = args.output_dir
temp_fs = TempFileSystem(parent=args.temp_dir, prefix='temp-files-')
args.stat_prefixes = [s.rstrip('.') for s in args.stat_prefixes]
stat_patterns = get_patterns_from_prefixes(
[s + '.' for s in args.stat_prefixes], ignore_case=True)
if not args.seed:
args.seed = random.randint(1, 999999999)
GLOBAL_RNG.seed(args.seed)
compress_level = None
if args.compress:
compress_level = 9
cfg = MsBayesConfig(args.config)
num_taxon_pairs = cfg.npairs
cfg.div_model_prior = 'constrained'
cfg.psi = probability.DiscreteUniformDistribution(num_taxon_pairs,
num_taxon_pairs)
config_path = temp_fs.get_file_path(prefix='cfg-')
cfg.write(config_path)
info.write('[pymsbayes]', log.info)
info.write('\tprogram_name = {name}'.format(**_program_info), log.info)
info.write('\tversion = {version}'.format(**_program_info), log.info)
info.write('\tinvocation = {0!r}'.format(' '.join(sys.argv)), log.info)
info.write('\toutput_directory = {0!r}'.format(args.output_dir), log.info)
info.write('\ttemp_directory = {0!r}'.format(temp_fs.base_dir), log.info)
info.write('\tsort_index = {0}'.format(MSBAYES_SORT_INDEX.current_value()),
log.info)
info.write(
'\tstat_patterns = {0!r}'.format(', '.join(
[p.pattern for p in stat_patterns])), log.info)
info.write('\tseed = {0}'.format(args.seed), log.info)
info.write('\tnum_prior_samples = {0}'.format(args.num_prior_samples),
log.info)
info.write('\tstats_by_time_path = {0!r}'.format(stats_by_time_path),
log.info)
info.write('\t[[tool_paths]]', log.info)
info.write('\t\tdpp_msbayes = {0}'.format(dpp_msbayes_path), log.info)
info.write('\t\tmsbayes = {0}'.format(msbayes_path), log.info)
info.write('\t[[config]]', log.debug)
info.write('{0}'.format(str(cfg)), log.debug)
##########################################################################
## begin analysis --- generate samples
start_time = datetime.datetime.now()
if args.np > args.num_prior_samples:
args.np = args.num_prior_samples
batch_size, remainder = long_division(args.num_prior_samples, args.np)
schema = 'abctoolbox'
workers = []
for i in range(args.np):
sample_size = batch_size
if i == (args.np - 1):
sample_size += remainder
w = MsBayesWorker(temp_fs=temp_fs,
sample_size=sample_size,
config_path=config_path,
report_parameters=True,
schema=schema,
include_header=True,
stat_patterns=stat_patterns,
write_stats_file=False)
workers.append(w)
log.info('Generating samples...')
workers = Manager.run_workers(workers=workers, num_processors=args.np)
log.info('Parsing samples...')
stats_by_time = get_stats_by_time([w.prior_path for w in workers])
stat_keys = stats_by_time.keys()
stat_keys.remove('PRI.t')
for prefix in args.stat_prefixes:
if not prefix in stat_keys:
raise Exception('stat prefix {0!r} not found in simulated stats:'
'\n\t{1}'.format(prefix, ', '.join(stat_keys)))
header = ['PRI.t'] + args.stat_prefixes
log.info('Writing stats-by-time matrix...')
out, close = process_file_arg(stats_by_time_path,
'w',
compresslevel=compress_level)
for row in dict_line_iter(stats_by_time, sep='\t', header=header):
out.write(row)
if close:
out.close()
log.info('Creating plots...')
if not MATPLOTLIB_AVAILABLE:
log.warning(
'`matplotlib` could not be imported, so the plot can not be\n'
'produced. The data to create the plot can be found in:\n\t'
'{0!r}'.format(stats_by_time_path))
else:
y_labels = {
'pi': r'$\pi$',
'pi.net': r'$\pi_{net}$',
'wattTheta': r'$\theta_W$',
'tajD.denom': r'$SD(\pi - \theta_W)$'
}
spg = SaturationPlotGrid(stats_by_time,
x_key='PRI.t',
y_keys=args.stat_prefixes,
y_labels=y_labels,
num_columns=2,
vertical_line_positions=args.vertical_lines)
fig = spg.create_grid()
fig.savefig(plot_path)
stop_time = datetime.datetime.now()
log.info('Done!')
info.write('\t[[run_stats]]', log.info)
info.write('\t\tstart_time = {0}'.format(str(start_time)), log.info)
info.write('\t\tstop_time = {0}'.format(str(stop_time)), log.info)
info.write('\t\ttotal_duration = {0}'.format(str(stop_time - start_time)),
log.info)
if not args.keep_temps:
log.debug('purging temps...')
temp_fs.purge()
