def get_contig(self, chromosome, genetic_map=None, length_multiplier=1):
"""
Returns a :class:`.Contig` instance describing a section of genome that
is to be simulated based on empirical information for a given species
and chromosome.
:param str chromosome: The ID of the chromosome to simulate.
:param str genetic_map: If specified, obtain recombination rate information
from the genetic map with the specified ID. If None, simulate
using a default uniform recombination rate on a region with the length of
the specified chromosome. The default rates are species- and chromosome-
specific, and can be found in the :ref:`sec_catalog`. (Default: None)
:param float length_multiplier: If specified, simulate a region of length
`length_multiplier` times the length of the specified chromosome with the
same chromosome-specific mutation and recombination rates.
This option cannot currently be used in conjunction with the
``genetic_map`` argument.
:rtype: :class:`.Contig`
:return: A :class:`.Contig` describing a simulation of the section of genome.
"""
# TODO: add non-autosomal support
if (chromosome is not None and chromosome.lower()
in ("x", "y", "m", "mt", "chrx", "chry", "chrm")):
warnings.warn(
stdpopsim.NonAutosomalWarning(
"Non-autosomal simulations are not yet supported. See "
"https://github.com/popsim-consortium/stdpopsim/issues/383 and "
"https://github.com/popsim-consortium/stdpopsim/issues/406"
))
chrom = self.genome.get_chromosome(chromosome)
if genetic_map is None:
logger.debug(
f"Making flat chromosome {length_multiplier} * {chrom.id}")
gm = None
recomb_map = msprime.RecombinationMap.uniform_map(
chrom.length * length_multiplier, chrom.recombination_rate)
else:
if length_multiplier != 1:
raise ValueError(
"Cannot use length multiplier with empirical maps")
logger.debug(f"Getting map for {chrom.id} from {genetic_map}")
gm = self.get_genetic_map(genetic_map)
recomb_map = gm.get_chromosome_map(chrom.id)
ret = stdpopsim.Contig(recombination_map=recomb_map,
mutation_rate=chrom.mutation_rate,
genetic_map=gm)
return ret
def get_contig(
self,
chromosome=None,
genetic_map=None,
length_multiplier=1,
length=None,
inclusion_mask=None,
exclusion_mask=None,
):
"""
Returns a :class:`.Contig` instance describing a section of genome that
is to be simulated based on empirical information for a given species
and chromosome.
:param str chromosome: The ID of the chromosome to simulate.
A complete list of chromosome IDs for each species can be found in the
"Genome" subsection for the species in the :ref:`sec_catalog`.
If the chromosome is not given, we specify a "generic" contig with given
``length``.
:param str genetic_map: If specified, obtain recombination rate information
from the genetic map with the specified ID. If None, simulate
using a default uniform recombination rate on a region with the length of
the specified chromosome. The default rates are species- and chromosome-
specific, and can be found in the :ref:`sec_catalog`. (Default: None)
:param float length_multiplier: If specified, simulate a region of length
`length_multiplier` times the length of the specified chromosome with the
same chromosome-specific mutation and recombination rates.
This option cannot currently be used in conjunction with the
``genetic_map`` argument.
:param inclusion_mask: If specified, simulated genomes are subset to only
inlude regions given by the mask. The mask can be specified by the
path and file name of a bed file or as a list or array of intervals
given by the left and right end points of the intervals.
:param exclusion_mask: If specified, simulated genomes are subset to exclude
regions given by the mask. The mask can be specified by the
path and file name of a bed file or as a list or array of intervals
given by the left and right end points of the intervals.
:param float length: Used with a "generic" contig, specifies the
length of genome sequence for this contig. For a generic contig, mutation
and recombination rates are equal to the genome-wide average across all
autosomal chromosomes.
:rtype: :class:`.Contig`
:return: A :class:`.Contig` describing the section of the genome.
"""
# TODO: add non-autosomal support
non_autosomal_lower = ["x", "y", "m", "mt", "chrx", "chry", "chrm"]
if chromosome is not None and chromosome.lower(
) in non_autosomal_lower:
warnings.warn(
stdpopsim.NonAutosomalWarning(
"Non-autosomal simulations are not yet supported. See "
"https://github.com/popsim-consortium/stdpopsim/issues/383 and "
"https://github.com/popsim-consortium/stdpopsim/issues/406"
))
if chromosome is None:
if genetic_map is not None:
raise ValueError("Cannot use genetic map with generic contic")
if length_multiplier != 1:
raise ValueError(
"Cannot use length multiplier for generic contig")
if inclusion_mask is not None or exclusion_mask is not None:
raise ValueError("Cannot use mask with generic contig")
if length is None:
raise ValueError(
"Must specify sequence length of generic contig")
L_tot = 0
r_tot = 0
u_tot = 0
for chrom_data in self.genome.chromosomes:
if chrom_data.id.lower() not in non_autosomal_lower:
L_tot += chrom_data.length
r_tot += chrom_data.length * chrom_data.recombination_rate
u_tot += chrom_data.length * chrom_data.mutation_rate
u = u_tot / L_tot
r = r_tot / L_tot
recomb_map = msprime.RateMap.uniform(length, r)
ret = stdpopsim.Contig(recombination_map=recomb_map,
mutation_rate=u)
else:
if length is not None:
raise ValueError(
"Cannot specify sequence length for named contig")
if inclusion_mask is not None and exclusion_mask is not None:
raise ValueError(
"Cannot specify both inclusion and exclusion masks")
chrom = self.genome.get_chromosome(chromosome)
if genetic_map is None:
logger.debug(
f"Making flat chromosome {length_multiplier} * {chrom.id}")
gm = None
recomb_map = msprime.RateMap.uniform(
round(chrom.length * length_multiplier),
chrom.recombination_rate)
else:
if length_multiplier != 1:
raise ValueError(
"Cannot use length multiplier with empirical maps")
logger.debug(f"Getting map for {chrom.id} from {genetic_map}")
gm = self.get_genetic_map(genetic_map)
recomb_map = gm.get_chromosome_map(chrom.id)
inclusion_intervals = None
exclusion_intervals = None
if inclusion_mask is not None:
if length_multiplier != 1:
raise ValueError("Cannot use length multiplier with mask")
if isinstance(inclusion_mask, str):
inclusion_intervals = stdpopsim.utils.read_bed(
inclusion_mask, chromosome)
else:
inclusion_intervals = inclusion_mask
if exclusion_mask is not None:
if length_multiplier != 1:
raise ValueError("Cannot use length multiplier with mask")
if isinstance(exclusion_mask, str):
exclusion_intervals = stdpopsim.utils.read_bed(
exclusion_mask, chromosome)
else:
exclusion_intervals = exclusion_mask
ret = stdpopsim.Contig(
recombination_map=recomb_map,
mutation_rate=chrom.mutation_rate,
genetic_map=gm,
inclusion_mask=inclusion_intervals,
exclusion_mask=exclusion_intervals,
)
return ret
def species_contig(
*,
species,
chromosome=None,
genetic_map=None,
length_multiplier=1,
length=None,
mutation_rate=None,
inclusion_mask=None,
exclusion_mask=None,
):
"""
Build a Contig for a species.
"""
# TODO: add non-autosomal support
non_autosomal_lower = ["x", "y", "m", "mt", "chrx", "chry", "chrm"]
if chromosome is not None and chromosome.lower(
) in non_autosomal_lower:
warnings.warn(
stdpopsim.NonAutosomalWarning(
"Non-autosomal simulations are not yet supported. See "
"https://github.com/popsim-consortium/stdpopsim/issues/383 and "
"https://github.com/popsim-consortium/stdpopsim/issues/406"
))
if chromosome is None:
if genetic_map is not None:
raise ValueError("Cannot use genetic map with generic contig")
if length_multiplier != 1:
raise ValueError(
"Cannot use length multiplier for generic contig")
if inclusion_mask is not None or exclusion_mask is not None:
raise ValueError("Cannot use mask with generic contig")
if length is None:
raise ValueError(
"Must specify sequence length of generic contig")
L_tot = 0
r_tot = 0
u_tot = 0
for chrom_data in species.genome.chromosomes:
if chrom_data.id.lower() not in non_autosomal_lower:
L_tot += chrom_data.length
r_tot += chrom_data.length * chrom_data.recombination_rate
u_tot += chrom_data.length * chrom_data.mutation_rate
if mutation_rate is None:
mutation_rate = u_tot / L_tot
r = r_tot / L_tot
contig = Contig.basic_contig(
length=length,
mutation_rate=mutation_rate,
recombination_rate=r,
)
else:
if length is not None:
raise ValueError(
"Cannot specify sequence length for named contig")
if inclusion_mask is not None and exclusion_mask is not None:
raise ValueError(
"Cannot specify both inclusion and exclusion masks")
chrom = species.genome.get_chromosome(chromosome)
if genetic_map is None:
logger.debug(
f"Making flat chromosome {length_multiplier} * {chrom.id}")
gm = None
recomb_map = msprime.RateMap.uniform(
round(chrom.length * length_multiplier),
chrom.recombination_rate)
else:
if length_multiplier != 1:
raise ValueError(
"Cannot use length multiplier with empirical maps")
logger.debug(f"Getting map for {chrom.id} from {genetic_map}")
gm = species.get_genetic_map(genetic_map)
recomb_map = gm.get_chromosome_map(chrom.id)
inclusion_intervals = None
exclusion_intervals = None
if inclusion_mask is not None:
if length_multiplier != 1:
raise ValueError("Cannot use length multiplier with mask")
if isinstance(inclusion_mask, str):
inclusion_intervals = stdpopsim.utils.read_bed(
inclusion_mask, chromosome)
else:
inclusion_intervals = inclusion_mask
if exclusion_mask is not None:
if length_multiplier != 1:
raise ValueError("Cannot use length multiplier with mask")
if isinstance(exclusion_mask, str):
exclusion_intervals = stdpopsim.utils.read_bed(
exclusion_mask, chromosome)
else:
exclusion_intervals = exclusion_mask
if mutation_rate is None:
mutation_rate = chrom.mutation_rate
contig = stdpopsim.Contig(
recombination_map=recomb_map,
mutation_rate=mutation_rate,
genetic_map=gm,
inclusion_mask=inclusion_intervals,
exclusion_mask=exclusion_intervals,
)
return contig