def HuberThreeEpoch():
id = "QC-African3Epoch_1H18"
populations = [
stdpopsim.Population(id="SouthMiddleAtlas", description="A. thalina"),
]
# Time of second epoch
T_2 = 7420
T_3 = 14534
# population sizes
N_ANC = 161744
N_2 = 24076
N_3 = 203077
return stdpopsim.DemographicModel(
id=id,
description=id,
long_description=id,
generation_time=_species.generation_time,
populations=populations,
population_configurations=[
msprime.PopulationConfiguration(initial_size=N_3,
metadata=populations[0].asdict()),
],
demographic_events=[
msprime.PopulationParametersChange(time=T_3,
initial_size=N_2,
population_id=0),
msprime.PopulationParametersChange(time=T_2 + T_3,
initial_size=N_ANC,
population_id=0),
],
population_id_map=[{
"SouthMiddleAtlas": 0
}] * 3,
)
def _afr_2epoch():
N_A = 746148
N_0 = 100218
t_1 = 568344
populations = [
stdpopsim.Population(
id="SouthMiddleAtlas",
description="Arabidopsis Thaliana South Middle Atlas population")
]
return stdpopsim.DemographicModel(
id="African2Epoch_1H18",
description="South Middle Atlas African two epoch model",
long_description="""
Model estimated from site frequency spectrum of synonymous
SNPs from African South Middle Atlas samples using
Williamson et al. 2005 methodology. Values come from supplementary
table 1 of Huber et al 2018. Sizes change from N_A -> N_0 and t_1 is
time of the second epoch.
""",
populations=populations,
citations=[stdpopsim.Citation(
author="Huber et al.",
year=2018,
doi="https://doi.org/10.1038/s41467-018-05281-7",
reasons={stdpopsim.CiteReason.DEM_MODEL})
],
generation_time=1,
population_configurations=[
msprime.PopulationConfiguration(
initial_size=N_0, metadata=populations[0].asdict())
],
demographic_events=[
msprime.PopulationParametersChange(
time=t_1, initial_size=N_A, population_id=0)
]
)
def _GAS_sp():
# the size during the interval times[k] to times[k+1] = sizes[k]
times = np.array([
1.08221472e01,
1.08221472e01,
1.77815418e01,
1.77815418e01,
2.43957877e01,
2.43957877e01,
2.62194838e01,
2.62194838e01,
2.80715527e01,
2.80715527e01,
2.99059882e01,
2.99059882e01,
3.17590591e01,
3.17590591e01,
3.36213534e01,
3.36213534e01,
3.55031410e01,
3.55031410e01,
3.74218265e01,
3.74218265e01,
3.93692837e01,
3.93692837e01,
4.13556900e01,
4.13556900e01,
4.33822257e01,
4.33822257e01,
4.54501193e01,
4.54501193e01,
4.75606499e01,
4.75606499e01,
4.97151498e01,
4.97151498e01,
5.19150077e01,
5.19150077e01,
5.41616710e01,
5.41616710e01,
5.64566497e01,
5.64566497e01,
5.88015192e01,
5.88015192e01,
6.11979243e01,
6.11979243e01,
6.36475829e01,
6.36475829e01,
6.61522900e01,
6.61522900e01,
6.87139222e01,
6.87139222e01,
7.13468519e01,
7.13468519e01,
7.40546632e01,
7.40546632e01,
7.69788992e01,
7.69788992e01,
8.00932536e01,
8.00932536e01,
8.33883683e01,
8.33883683e01,
9.33624771e01,
9.33624771e01,
1.05867180e02,
1.05867180e02,
1.19778311e02,
1.19778311e02,
1.34317538e02,
1.34317538e02,
1.49486416e02,
1.49486416e02,
1.65081438e02,
1.65081438e02,
1.81069660e02,
1.81069660e02,
1.96884420e02,
1.96884420e02,
2.13010501e02,
2.13010501e02,
2.28659684e02,
2.28659684e02,
2.43883726e02,
2.43883726e02,
2.58403062e02,
2.58403062e02,
2.70573820e02,
2.70573820e02,
2.75664382e02,
2.75664382e02,
2.80730912e02,
2.80730912e02,
2.85888999e02,
2.85888999e02,
2.91187697e02,
2.91187697e02,
2.96632655e02,
2.96632655e02,
3.02218199e02,
3.02218199e02,
3.07951917e02,
3.07951917e02,
3.13860240e02,
3.13860240e02,
3.19955507e02,
3.19955507e02,
3.26244790e02,
3.26244790e02,
3.32745865e02,
3.32745865e02,
3.39465956e02,
3.39465956e02,
3.46421839e02,
3.46421839e02,
3.53626147e02,
3.53626147e02,
3.61092430e02,
3.61092430e02,
3.68835241e02,
3.68835241e02,
3.76870235e02,
3.76870235e02,
3.85214266e02,
3.85214266e02,
3.93885515e02,
3.93885515e02,
4.02903613e02,
4.02903613e02,
4.12289797e02,
4.12289797e02,
4.22067072e02,
4.22067072e02,
4.32260401e02,
4.32260401e02,
4.42896919e02,
4.42896919e02,
4.54006170e02,
4.54006170e02,
4.65620388e02,
4.65620388e02,
4.77774801e02,
4.77774801e02,
4.90490498e02,
4.90490498e02,
5.03826473e02,
5.03826473e02,
5.17829247e02,
5.17829247e02,
5.32550112e02,
5.32550112e02,
5.48045759e02,
5.48045759e02,
5.64379009e02,
5.64379009e02,
5.81588679e02,
5.81588679e02,
5.99781759e02,
5.99781759e02,
6.19045020e02,
6.19045020e02,
6.39454764e02,
6.39454764e02,
6.61129552e02,
6.61129552e02,
6.84129309e02,
6.84129309e02,
7.08609246e02,
7.08609246e02,
7.34719738e02,
7.34719738e02,
7.62657505e02,
7.62657505e02,
7.92661043e02,
7.92661043e02,
8.24963147e02,
8.24963147e02,
8.59801339e02,
8.59801339e02,
8.97519058e02,
8.97519058e02,
9.38516578e02,
9.38516578e02,
9.83196252e02,
9.83196252e02,
1.03213113e03,
1.03213113e03,
1.08595950e03,
1.08595950e03,
1.14545402e03,
1.14545402e03,
1.21187638e03,
1.21187638e03,
1.42580467e03,
1.42580467e03,
1.66762718e03,
1.66762718e03,
1.94206662e03,
1.94206662e03,
2.25673855e03,
2.25673855e03,
2.61982155e03,
2.61982155e03,
3.04325043e03,
3.04325043e03,
3.54366638e03,
3.54366638e03,
4.14416552e03,
4.14416552e03,
4.86878040e03,
4.86878040e03,
5.74277111e03,
5.74277111e03,
6.06376420e03,
6.06376420e03,
6.48115044e03,
6.48115044e03,
7.16974182e03,
7.16974182e03,
9.94668728e03,
9.94668728e03,
1.08341379e04,
1.08341379e04,
1.12602248e04,
1.12602248e04,
1.80292683e04,
])
sizes = np.array([
4069863.26413754,
2570469.78880419,
2570469.78880419,
2398970.46335773,
2398970.46335773,
649423.64089016,
649423.64089016,
647425.87204344,
647425.87204344,
629386.55333025,
629386.55333025,
623896.55036331,
623896.55036331,
615171.6634657,
615171.6634657,
609770.71698915,
609770.71698915,
609770.71698915,
609770.71698915,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
606896.78477367,
609770.71698915,
609770.71698915,
612860.29930411,
612860.29930411,
646628.97921259,
646628.97921259,
672653.84034255,
672653.84034255,
694949.57011815,
694949.57011815,
2053484.48904133,
2053484.48904133,
2512451.15337624,
2512451.15337624,
2726859.95574373,
2726859.95574373,
2779609.28386339,
2779609.28386339,
2827486.59107239,
2827486.59107239,
2833326.90480091,
2833326.90480091,
2830283.03259496,
2830283.03259496,
2726859.95574373,
2726859.95574373,
2707367.42825967,
2707367.42825967,
2557240.92209016,
2557240.92209016,
2420531.19687943,
2420531.19687943,
2245241.1128414,
2245241.1128414,
1829782.22978925,
1829782.22978925,
743769.25832605,
743769.25832605,
719107.79112942,
719107.79112942,
710882.47527273,
710882.47527273,
708782.9061541,
708782.9061541,
706605.83914068,
706605.83914068,
702884.83678823,
702884.83678823,
699330.14877435,
699330.14877435,
698106.79217618,
698106.79217618,
697332.03584512,
697332.03584512,
696317.93467906,
696317.93467906,
696167.64460966,
696167.64460966,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
695633.55941906,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
694677.60526595,
693429.24269629,
693429.24269629,
693429.24269629,
693429.24269629,
693429.24269629,
693429.24269629,
692716.93526192,
692716.93526192,
692379.41939462,
692379.41939462,
690176.6983718,
690176.6983718,
688681.83478372,
688681.83478372,
687162.86783629,
687162.86783629,
686235.36369922,
686235.36369922,
686150.90910499,
686150.90910499,
685951.34145025,
685951.34145025,
685005.95081601,
685005.95081601,
684576.59330447,
684576.59330447,
684576.59330447,
684576.59330447,
683889.43627665,
683889.43627665,
683889.43627665,
683889.43627665,
683889.43627665,
683889.43627665,
683889.43627665,
683889.43627665,
687172.58638815,
687172.58638815,
1980227.21562606,
1980227.21562606,
1989715.58702617,
1989715.58702617,
1992430.35210815,
1992430.35210815,
1998953.44050118,
1998953.44050118,
1998953.44050118,
1998953.44050118,
1998160.88128216,
1998160.88128216,
1998160.88128216,
1998160.88128216,
1998160.88128216,
1998160.88128216,
1972764.03439268,
1972764.03439268,
1903551.76041575,
1903551.76041575,
543762.2932594,
543762.2932594,
530289.21416049,
530289.21416049,
624896.67937685,
624896.67937685,
1680052.006021,
1680052.006021,
322144.56539297,
322144.56539297,
77334.77862527,
77334.77862527,
409527.12801416,
409527.12801416,
])
demographic_events = []
for sz, t in zip(sizes, times):
demographic_events.append(
msprime.PopulationParametersChange(time=t,
initial_size=sz,
population_id=0))
populations = [
stdpopsim.Population(
id="GAS",
description="Gabon gabiae population",
)
]
return stdpopsim.DemographicModel(
id="GAS_1A17",
description="Stairwayplot estimates of N(t) for Gabon sample",
long_description="""
These estimates were done as part of the Ag1000G 2017 Consortium
paper. Stairwayplot was run with the addition of a misorientation
parameter using SFS information from each population.
""",
populations=populations,
citations=[
stdpopsim.Citation(
doi="https://doi.org/10.1038/nature24995",
year=2017,
author="Ag1000G Consortium",
reasons={stdpopsim.CiteReason.DEM_MODEL},
)
],
# generation times and mutation rate given at bottom of page 32 in supp info
generation_time=1 / 11,
mutation_rate=3.5e-9,
demographic_events=demographic_events,
population_configurations=[
msprime.PopulationConfiguration(initial_size=4069863,
metadata=populations[0].asdict())
],
)
def _afr_3epoch():
id = "African3Epoch_1S16"
description = "Three epoch African population"
long_description = """
The three epoch (modern, bottleneck, ancestral) model estimated for a
single African Drosophila Melanogaster population from Sheehan and Song (2016).
Population sizes are estimated by a
deep learning model trained on simulation data. NOTE: Due to differences in
coalescence units between PSMC (2N) and msms (4N) the number of generations were
doubled from PSMC estimates when simulating data from msms in the original
publication. We have faithfully represented the published model here.
"""
populations = [_afr_population]
citations = [
stdpopsim.Citation(
author="Sheehan and Song",
year=2016,
doi="https://doi.org/10.1371/journal.pcbi.1004845",
reasons={stdpopsim.CiteReason.DEM_MODEL},
)
]
generation_time = _species.generation_time
mutation_rate = 8.4e-9
# Parameter values from "Simulating Data" section
# these are assumptions, not estimates
N_ref = 100000
t_1_coal = 0.5
t_2_coal = 5.0
# estimates from the ANN
N_R = 544200
N_B = 145300
N_A = 652700
# Times are provided in 4N_ref generations, so we convert into generations.
# generation_time = 10 / year
t_1 = t_1_coal * 4 * N_ref
t_2 = (t_1_coal + t_2_coal) * 4 * N_ref
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
mutation_rate=mutation_rate,
population_configurations=[
msprime.PopulationConfiguration(initial_size=N_R,
metadata=populations[0].asdict())
],
demographic_events=[
# Size change at bottleneck (back in time; BIT)
msprime.PopulationParametersChange(time=t_1,
initial_size=N_B,
population_id=0),
# Size change at recovery (BIT)
msprime.PopulationParametersChange(time=t_2,
initial_size=N_A,
population_id=0),
],
)
def LiStephanTwoPopulation():
id = "QC-OutOfAfrica_2L06"
populations = [
stdpopsim.Population("AFR", ""),
stdpopsim.Population("EUR", ""),
]
# Parameters for the African population are taken from the section Demographic
# History of the African Population
generation_time = 0.1 # 10 generations per year
N_A0 = 8.603e6 # modern African pop. size
N_A1 = N_A0 / 5.0 # African pop. size before expansion
# Parameters for the European population are taken from the section Demographic
# History of the European Population
N_E0 = 1.075e6 # modern European pop. size
N_E1 = 2.2e3 # European founder pop. size
# Times from from the section Demographic History of the * Population
T_A0 = 6e4 / generation_time # time of 1st expansion in African pop.
T_E_A = 15.8e3 / generation_time # European/African divergence time
T_EE = T_E_A - 340 / generation_time # Time of European pop. re-expansion
return stdpopsim.DemographicModel(
id=id,
description=id,
long_description=id,
generation_time=generation_time,
populations=populations,
# Set population sizes at T=0
# pop0 is Africa, pop1 is Europe
population_configurations=[
msprime.PopulationConfiguration(initial_size=N_A0, growth_rate=0),
msprime.PopulationConfiguration(initial_size=N_E0, growth_rate=0),
],
# Now we add the demographic events working backwards in time.
demographic_events=[
# OOA bottleneck
msprime.PopulationParametersChange(time=T_EE,
initial_size=N_E1,
population_id=1),
# E and A coalesce
msprime.MassMigration(time=T_E_A,
source=1,
destination=0,
proportion=1.0),
# Pre-expansion Africa
msprime.PopulationParametersChange(time=T_A0,
initial_size=N_A1,
population_id=0),
],
population_id_map=[
{
"AFR": 0,
"EUR": 1
},
{
"AFR": 0,
"EUR": 1
},
{
"AFR": 0,
"EUR": 1
},
{
"AFR": 0,
"EUR": 1
},
],
)
class TestNoQCWarning:
species = stdpopsim.get_species("EscCol")
model = stdpopsim.DemographicModel(
id="FakeModel",
description="FakeModel",
long_description="FakeModel",
citations=[
stdpopsim.Citation(
author="Farnsworth",
year=3000,
doi="https://doi.org/10.1000/xyz123",
reasons={stdpopsim.CiteReason.DEM_MODEL},
)
],
generation_time=10,
populations=[stdpopsim.Population("Omicronians", "Popplers, etc.")],
population_configurations=[msprime.PopulationConfiguration(initial_size=1000)],
)
@classmethod
def setup_class(cls):
cls.species.add_demographic_model(cls.model)
@classmethod
def teardown_class(cls):
cls.species.demographic_models.remove(cls.model)
def verify_noQC_warning(self, cmd):
# setup_logging() interferes with pytest.warns().
with mock.patch("stdpopsim.cli.setup_logging", autospec=True):
with pytest.warns(stdpopsim.QCMissingWarning):
capture_output(stdpopsim.cli.stdpopsim_main, cmd.split())
def test_noQC_warning(self):
self.verify_noQC_warning("EscCol -d FakeModel -D 10 -L 10")
def test_noQC_warning_quiet(self):
self.verify_noQC_warning("-q EscCol -d FakeModel -D 10 -L 10")
def verify_noQC_citations_not_written(self, cmd, caplog):
# Non-QCed models shouldn't be used in publications, so citations
# shouldn't be offered to the user.
out, err = capture_output(stdpopsim.cli.stdpopsim_main, cmd.split())
log_output = caplog.text
for citation in self.model.citations:
assert not (citation.author in out)
assert not (citation.doi in out)
assert not (citation.author in err)
assert not (citation.doi in err)
assert not (citation.author in log_output)
assert not (citation.doi in log_output)
# The following two tests use the "caplog" pytest fixture, which captures
# the logging output. The caplog param is automatically passed to test_*()
# methods by pytest, which we pass through to verify_noQC_citations_not_written().
@pytest.mark.filterwarnings("ignore::stdpopsim.QCMissingWarning")
@pytest.mark.usefixtures("caplog")
def test_noQC_citations_not_written(self, caplog):
self.verify_noQC_citations_not_written(
"EscCol -d FakeModel -D 10 -L 10", caplog
)
@pytest.mark.filterwarnings("ignore::stdpopsim.QCMissingWarning")
@pytest.mark.usefixtures("caplog")
def test_noQC_citations_not_written_verbose(self, caplog):
self.verify_noQC_citations_not_written(
"-vv EscCol -d FakeModel -D 10 -L 10", caplog
)
def hominin_composite():
id = "HomininComposite_4G20"
description = "Four population out of Africa with Neandertal admixture"
long_description = """
A composite of demographic parameters from multiple sources
"""
# samples:
# T_Altai = 115e3
# T_Vindija = 55e3
# n_YRI = 108
# n_CEU = 99
populations = [
stdpopsim.Population(id="YRI", description="1000 Genomes YRI (Yorubans)"),
stdpopsim.Population(
id="CEU",
description=(
"1000 Genomes CEU (Utah Residents (CEPH) with Northern and "
"Western European Ancestry"
),
),
stdpopsim.Population(id="Nea", description="Neandertal lineage"),
stdpopsim.Population(
id="Anc", description="Ancestral hominins", sampling_time=None
),
]
pop = {p.id: i for i, p in enumerate(populations)}
citations = [
stdpopsim.Citation(
author="Kuhlwilm et al.",
year=2016,
doi="https://doi.org/10.1038/nature16544",
),
stdpopsim.Citation(
author="Prüfer et al.",
year=2017,
doi="https://doi.org/10.1126/science.aao1887",
),
stdpopsim.Citation(
author="Ragsdale and Gravel",
year=2019,
doi="https://doi.org/10.1371/journal.pgen.1008204",
),
]
generation_time = 29
# Kuhlwilm et al. 2016
N_YRI = 27000
N_Nea = 3400
N_Anc = 18500
# Ragsdale & Gravel 2019
N_CEU0 = 1450
r_CEU = 0.00202
T_CEU_exp = 31.9e3 / generation_time
N_CEU = N_CEU0 * math.exp(r_CEU * T_CEU_exp)
T_YRI_CEU_split = 65.7e3 / generation_time
N_ooa_bottleneck = 1080
# Prüfer et al. 2017
T_Nea_human_split = 550e3 / generation_time
T_Nea_CEU_mig = 55e3 / generation_time
m_Nea_CEU = 0.0225
pop_meta = (p.asdict() for p in populations)
population_configurations = [
msprime.PopulationConfiguration(initial_size=N_YRI, metadata=next(pop_meta)),
msprime.PopulationConfiguration(
initial_size=N_CEU, growth_rate=r_CEU, metadata=next(pop_meta)
),
msprime.PopulationConfiguration(initial_size=N_Nea, metadata=next(pop_meta)),
msprime.PopulationConfiguration(initial_size=N_Anc, metadata=next(pop_meta)),
]
demographic_events = [
# out-of-Africa bottleneck
msprime.PopulationParametersChange(
time=T_CEU_exp,
initial_size=N_ooa_bottleneck,
growth_rate=0,
population_id=pop["CEU"],
),
# Neandertal -> CEU admixture
msprime.MassMigration(
time=T_Nea_CEU_mig,
proportion=m_Nea_CEU,
source=pop["CEU"],
destination=pop["Nea"],
),
# population splits
msprime.MassMigration(
time=T_YRI_CEU_split, source=pop["CEU"], destination=pop["Anc"]
),
msprime.MassMigration(
time=T_YRI_CEU_split, source=pop["YRI"], destination=pop["Anc"]
),
msprime.MassMigration(
time=T_Nea_human_split, source=pop["Nea"], destination=pop["Anc"]
),
]
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
population_configurations=population_configurations,
demographic_events=demographic_events,
)
def test_population_construction_no_popconfig(self):
dm = stdpopsim.DemographicModel(
id="A", description="A", long_description="A", generation_time=1
)
self.assertEqual(dm.populations, [])
def _ooa_archaic():
id = "OutOfAfricaArchaicAdmixture_5R19"
description = "Three population out-of-Africa with archaic admixture"
long_description = """
The three population out-of-African model popularized by Gutenkunst et al. (2009)
and augmented by archaic contributions to both Eurasian and African populations.
Two archaic populations split early in human history, before the African
expansion, and contribute to Eurasian populations (putative Neanderthal branch)
and to the African branch (a deep diverging branch within Africa). Admixture
is modeled as symmetric migration between the archaic and modern human branches,
with contribution ending at a given time in the past.
"""
populations = [
_yri_population,
_ceu_population,
_chb_population,
stdpopsim.Population("Neanderthal",
"Putative Neanderthals",
sampling_time=None),
stdpopsim.Population("ArchaicAFR",
"Putative Archaic Africans",
sampling_time=None),
]
citations = [
stdpopsim.Citation(author="Ragsdale and Gravel",
year=2019,
doi="https://doi.org/10.1371/journal.pgen.1008204",
reasons={stdpopsim.CiteReason.DEM_MODEL})
]
# First we set out the maximum likelihood values of the various parameters
# given in Table 1 (under archaic admixture).
N_0 = 3600
N_YRI = 13900
N_B = 880
N_CEU0 = 2300
N_CHB0 = 650
# Times are provided in years, so we convert into generations.
# In the published model, the authors used a generation time of 29 years to
# convert from genetic to physical units
generation_time = 29
T_AF = 300e3 / generation_time
T_B = 60.7e3 / generation_time
T_EU_AS = 36.0e3 / generation_time
T_arch_afr_split = 499e3 / generation_time
T_arch_afr_mig = 125e3 / generation_time
T_nean_split = 559e3 / generation_time
T_arch_adm_end = 18.7e3 / generation_time
# We need to work out the starting (diploid) population sizes based on
# the growth rates provided for these two populations
r_CEU = 0.00125
r_CHB = 0.00372
N_CEU = N_CEU0 / math.exp(-r_CEU * T_EU_AS)
N_CHB = N_CHB0 / math.exp(-r_CHB * T_EU_AS)
# Migration rates during the various epochs.
m_AF_B = 52.2e-5
m_YRI_CEU = 2.48e-5
m_YRI_CHB = 0e-5
m_CEU_CHB = 11.3e-5
m_AF_arch_af = 1.98e-5
m_OOA_nean = 0.825e-5
# Population IDs correspond to their indexes in the population
# configuration array. Therefore, we have 0=YRI, 1=CEU and 2=CHB
# initially.
# We also have two archaic populations, putative Neanderthals and
# archaicAfrican, which are population indices 3=Nean and 4=arch_afr.
# Their sizes are equal to the ancestral reference population size N_0.
population_configurations = [
msprime.PopulationConfiguration(initial_size=N_YRI,
metadata=populations[0].asdict()),
msprime.PopulationConfiguration(initial_size=N_CEU,
growth_rate=r_CEU,
metadata=populations[1].asdict()),
msprime.PopulationConfiguration(initial_size=N_CHB,
growth_rate=r_CHB,
metadata=populations[2].asdict()),
msprime.PopulationConfiguration(initial_size=N_0,
metadata=populations[3].asdict()),
msprime.PopulationConfiguration(initial_size=N_0,
metadata=populations[4].asdict())
]
migration_matrix = [ # noqa
[0, m_YRI_CEU, m_YRI_CHB, 0, 0], # noqa
[m_YRI_CEU, 0, m_CEU_CHB, 0, 0], # noqa
[m_YRI_CHB, m_CEU_CHB, 0, 0, 0], # noqa
[0, 0, 0, 0, 0], # noqa
[0, 0, 0, 0, 0] # noqa
] # noqa
demographic_events = [
# first event is migration turned on between modern and archaic humans
msprime.MigrationRateChange(time=T_arch_adm_end,
rate=m_AF_arch_af,
matrix_index=(0, 4)),
msprime.MigrationRateChange(time=T_arch_adm_end,
rate=m_AF_arch_af,
matrix_index=(4, 0)),
msprime.MigrationRateChange(time=T_arch_adm_end,
rate=m_OOA_nean,
matrix_index=(1, 3)),
msprime.MigrationRateChange(time=T_arch_adm_end,
rate=m_OOA_nean,
matrix_index=(3, 1)),
msprime.MigrationRateChange(time=T_arch_adm_end,
rate=m_OOA_nean,
matrix_index=(2, 3)),
msprime.MigrationRateChange(time=T_arch_adm_end,
rate=m_OOA_nean,
matrix_index=(3, 2)),
# CEU and CHB merge into B with rate changes at T_EU_AS
msprime.MassMigration(time=T_EU_AS,
source=2,
destination=1,
proportion=1.0),
msprime.MigrationRateChange(time=T_EU_AS, rate=0),
msprime.MigrationRateChange(time=T_EU_AS,
rate=m_AF_B,
matrix_index=(0, 1)),
msprime.MigrationRateChange(time=T_EU_AS,
rate=m_AF_B,
matrix_index=(1, 0)),
msprime.MigrationRateChange(time=T_EU_AS,
rate=m_AF_arch_af,
matrix_index=(0, 4)),
msprime.MigrationRateChange(time=T_EU_AS,
rate=m_AF_arch_af,
matrix_index=(4, 0)),
msprime.MigrationRateChange(time=T_EU_AS,
rate=m_OOA_nean,
matrix_index=(1, 3)),
msprime.MigrationRateChange(time=T_EU_AS,
rate=m_OOA_nean,
matrix_index=(3, 1)),
msprime.PopulationParametersChange(time=T_EU_AS,
initial_size=N_B,
growth_rate=0,
population_id=1),
# Population B merges into YRI at T_B
msprime.MassMigration(time=T_B,
source=1,
destination=0,
proportion=1.0),
msprime.MigrationRateChange(time=T_B, rate=0),
msprime.MigrationRateChange(time=T_B,
rate=m_AF_arch_af,
matrix_index=(0, 4)),
msprime.MigrationRateChange(time=T_B,
rate=m_AF_arch_af,
matrix_index=(4, 0)),
# Beginning of migration between African and archaic African populations
msprime.MigrationRateChange(time=T_arch_afr_mig, rate=0),
# Size changes to N_0 at T_AF
msprime.PopulationParametersChange(time=T_AF,
initial_size=N_0,
population_id=0),
# Archaic African merges with moderns
msprime.MassMigration(time=T_arch_afr_split,
source=4,
destination=0,
proportion=1.0),
# Neanderthal merges with moderns
msprime.MassMigration(time=T_nean_split,
source=3,
destination=0,
proportion=1.0)
]
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
population_configurations=population_configurations,
migration_matrix=migration_matrix,
demographic_events=demographic_events,
)
def _america():
id = "AmericanAdmixture_4B11"
description = "American admixture"
long_description = """
Demographic model for American admixture, taken from Browning et al. 2011.
This model extends the Gravel et al. (2011) model of African/European/Asian
demographic history to simulate an admixed population with admixture
occurring 12 generations ago. The admixed population had an initial size
of 30,000 and grew at a rate of 5% per generation, with 1/6 of the
population of African ancestry, 1/3 European, and 1/2 Asian.
"""
populations = [
stdpopsim.Population(id="AFR",
description="Contemporary African population"),
stdpopsim.Population(id="EUR",
description="Contemporary European population"),
stdpopsim.Population(id="ASIA",
description="Contemporary Asian population"),
stdpopsim.Population(id="ADMIX",
description="Modern admixed population"),
]
citations = [
stdpopsim.Citation(
author="Browning et al.",
year=2011,
doi="http://dx.doi.org/10.1371/journal.pgen.1007385",
reasons={stdpopsim.CiteReason.DEM_MODEL})
]
generation_time = 25
# Model code was ported from Supplementary File 1.
N0 = 7310 # initial population size
Thum = 5920 # time (gens) of advent of modern humans
Naf = 14474 # size of african population
Tooa = 2040 # number of generations back to Out of Africa
Nb = 1861 # size of out of Africa population
mafb = 1.5e-4 # migration rate Africa and Out-of-Africa
Teu = 920 # number generations back to Asia-Europe split
Neu = 1032 # bottleneck population sizes
Nas = 554
mafeu = 2.5e-5 # mig. rates
mafas = 7.8e-6
meuas = 3.11e-5
reu = 0.0038 # growth rate per generation in Europe
ras = 0.0048 # growth rate per generation in Asia
Tadmix = 12 # time of admixture
Nadmix = 30000 # initial size of admixed population
radmix = .05 # growth rate of admixed population
# pop0 is Africa, pop1 is Europe, pop2 is Asia, pop3 is admixed
population_configurations = [
msprime.PopulationConfiguration(initial_size=Naf,
growth_rate=0.0,
metadata=populations[0].asdict()),
msprime.PopulationConfiguration(initial_size=Neu * math.exp(reu * Teu),
growth_rate=reu,
metadata=populations[1].asdict()),
msprime.PopulationConfiguration(initial_size=Nas * math.exp(ras * Teu),
growth_rate=ras,
metadata=populations[2].asdict()),
msprime.PopulationConfiguration(initial_size=Nadmix *
math.exp(radmix * Tadmix),
growth_rate=radmix,
metadata=populations[3].asdict())
]
migration_matrix = [[0, mafeu, mafas, 0], [mafeu, 0, meuas, 0],
[mafas, meuas, 0, 0], [0, 0, 0, 0]]
# Admixture event, 1/6 Africa, 2/6 Europe, 3/6 Asia
admixture_event = [
msprime.MassMigration(time=Tadmix,
source=3,
destination=0,
proportion=1.0 / 6.0),
msprime.MassMigration(time=Tadmix + 0.0001,
source=3,
destination=1,
proportion=2.0 / 5.0),
msprime.MassMigration(time=Tadmix + 0.0002,
source=3,
destination=2,
proportion=1.0)
]
# Asia and Europe split
eu_event = [
msprime.MigrationRateChange(time=Teu, rate=0.0),
msprime.MassMigration(time=Teu + 0.0001,
source=2,
destination=1,
proportion=1.0),
msprime.PopulationParametersChange(time=Teu + 0.0002,
initial_size=Nb,
growth_rate=0.0,
population_id=1),
msprime.MigrationRateChange(time=Teu + 0.0003,
rate=mafb,
matrix_index=(0, 1)),
msprime.MigrationRateChange(time=Teu + 0.0003,
rate=mafb,
matrix_index=(1, 0))
]
# Out of Africa event
ooa_event = [
msprime.MigrationRateChange(time=Tooa, rate=0.0),
msprime.MassMigration(time=Tooa + 0.0001,
source=1,
destination=0,
proportion=1.0)
]
# initial population size
init_event = [
msprime.PopulationParametersChange(time=Thum,
initial_size=N0,
population_id=0)
]
demographic_events = admixture_event + eu_event + ooa_event + init_event
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
population_configurations=population_configurations,
migration_matrix=migration_matrix,
demographic_events=demographic_events,
)
def _ooa_2():
id = "OutOfAfrica_2T12"
description = "Two population out-of-Africa"
long_description = """
The model is derived from the Tennesen et al. analysis of the
jSFS from European Americans and African Americans.
It describes the ancestral human population in Africa, the out of Africa event,
and two distinct periods of subsequent European population growth over the past
23kya. Model parameters are taken from Fig. S5 in Fu et al.
"""
populations = [
stdpopsim.Population(id="AFR", description="African Americans"),
stdpopsim.Population(id="EUR", description="European Americans")
]
citations = [
_tennessen_et_al,
stdpopsim.Citation(author="Fu et al.",
year=2013,
doi="https://doi.org/10.1038/nature11690",
reasons={stdpopsim.CiteReason.DEM_MODEL})
]
generation_time = 25
T_AF = 148e3 / generation_time
T_OOA = 51e3 / generation_time
T_EU0 = 23e3 / generation_time
T_EG = 5115 / generation_time
# Growth rates
r_EU0 = 0.00307
r_EU = 0.0195
r_AF = 0.0166
# population sizes
N_A = 7310
N_AF1 = 14474
N_B = 1861
N_EU0 = 1032
N_EU1 = N_EU0 / math.exp(-r_EU0 * (T_EU0 - T_EG))
# migration rates
m_AF_B = 15e-5
m_AF_EU = 2.5e-5
# present Ne
N_EU = N_EU1 / math.exp(-r_EU * T_EG)
N_AF = N_AF1 / math.exp(-r_AF * T_EG)
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
population_configurations=[
msprime.PopulationConfiguration(initial_size=N_AF,
growth_rate=r_AF,
metadata=populations[0].asdict()),
msprime.PopulationConfiguration(initial_size=N_EU,
growth_rate=r_EU,
metadata=populations[1].asdict())
],
migration_matrix=[
[0, m_AF_EU],
[m_AF_EU, 0],
],
demographic_events=[
msprime.MigrationRateChange(time=T_EG,
rate=m_AF_EU,
matrix_index=(0, 1)),
msprime.MigrationRateChange(time=T_EG,
rate=m_AF_EU,
matrix_index=(1, 0)),
msprime.PopulationParametersChange(time=T_EG,
growth_rate=r_EU0,
initial_size=N_EU1,
population_id=1),
msprime.PopulationParametersChange(time=T_EG,
growth_rate=0,
initial_size=N_AF1,
population_id=0),
msprime.MigrationRateChange(time=T_EU0,
rate=m_AF_B,
matrix_index=(0, 1)),
msprime.MigrationRateChange(time=T_EU0,
rate=m_AF_B,
matrix_index=(1, 0)),
msprime.PopulationParametersChange(time=T_EU0,
initial_size=N_B,
growth_rate=0,
population_id=1),
msprime.MassMigration(time=T_OOA,
source=1,
destination=0,
proportion=1.0),
msprime.PopulationParametersChange(time=T_AF,
initial_size=N_A,
population_id=0)
],
)
def _ooa_3():
id = "OutOfAfrica_3G09"
description = "Three population out-of-Africa"
long_description = """
The three population Out-of-Africa model from Gutenkunst et al. 2009.
It describes the ancestral human population in Africa, the out of Africa
event, and the subsequent European-Asian population split.
Model parameters are the maximum likelihood values of the
various parameters given in Table 1 of Gutenkunst et al.
"""
populations = [_yri_population, _ceu_population, _chb_population]
citations = [
stdpopsim.Citation(author="Gutenkunst et al.",
year=2009,
doi="https://doi.org/10.1371/journal.pgen.1000695",
reasons={stdpopsim.CiteReason.DEM_MODEL})
]
generation_time = 25
# First we set out the maximum likelihood values of the various parameters
# given in Table 1.
N_A = 7300
N_B = 2100
N_AF = 12300
N_EU0 = 1000
N_AS0 = 510
# Times are provided in years, so we convert into generations.
T_AF = 220e3 / generation_time
T_B = 140e3 / generation_time
T_EU_AS = 21.2e3 / generation_time
# We need to work out the starting (diploid) population sizes based on
# the growth rates provided for these two populations
r_EU = 0.004
r_AS = 0.0055
N_EU = N_EU0 / math.exp(-r_EU * T_EU_AS)
N_AS = N_AS0 / math.exp(-r_AS * T_EU_AS)
# Migration rates during the various epochs.
m_AF_B = 25e-5
m_AF_EU = 3e-5
m_AF_AS = 1.9e-5
m_EU_AS = 9.6e-5
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
# Population IDs correspond to their indexes in the population
# configuration array. Therefore, we have 0=YRI, 1=CEU and 2=CHB
# initially.
population_configurations=[
msprime.PopulationConfiguration(initial_size=N_AF,
metadata=populations[0].asdict()),
msprime.PopulationConfiguration(initial_size=N_EU,
growth_rate=r_EU,
metadata=populations[1].asdict()),
msprime.PopulationConfiguration(initial_size=N_AS,
growth_rate=r_AS,
metadata=populations[2].asdict()),
],
migration_matrix=[
[0, m_AF_EU, m_AF_AS], # noqa
[m_AF_EU, 0, m_EU_AS], # noqa
[m_AF_AS, m_EU_AS, 0], # noqa
],
demographic_events=[
# CEU and CHB merge into B with rate changes at T_EU_AS
msprime.MassMigration(time=T_EU_AS,
source=2,
destination=1,
proportion=1.0),
msprime.MigrationRateChange(time=T_EU_AS, rate=0),
msprime.MigrationRateChange(time=T_EU_AS,
rate=m_AF_B,
matrix_index=(0, 1)),
msprime.MigrationRateChange(time=T_EU_AS,
rate=m_AF_B,
matrix_index=(1, 0)),
msprime.PopulationParametersChange(time=T_EU_AS,
initial_size=N_B,
growth_rate=0,
population_id=1),
# Population B merges into YRI at T_B
msprime.MassMigration(time=T_B,
source=1,
destination=0,
proportion=1.0),
# Size changes to N_A at T_AF
msprime.PopulationParametersChange(time=T_AF,
initial_size=N_A,
population_id=0)
],
)
def papuans_10j19():
id = "PapuansOutOfAfrica_10J19"
description = "Out-of-Africa with archaic admixture into Papuans"
long_description = """
A ten population model of out-of-Africa, including two pulses of
Denisovan admixture into Papuans, and several pulses of Neandertal
admixture into non-Africans.
Most parameters are from Jacobs et al. (2019), Table S5 and Figure S5.
This model is an extension of one from Malaspinas et al. (2016), thus
some parameters are inherited from there.
"""
# sampling times
T_DenA = 2203
T_NeaA = 3793
populations = [
# humans
stdpopsim.Population(id="YRI", description="1000 Genomes YRI (Yorubans)"),
stdpopsim.Population(
id="CEU",
description="1000 Genomes CEU (Utah Residents (CEPH) with Northern and Western European Ancestry",
),
stdpopsim.Population(
id="CHB", description="1000 Genomes CHB (Han Chinese in Beijing, China)"
),
stdpopsim.Population("Papuan", "Papuans from Indonesia and New Guinea"),
# archaics
stdpopsim.Population(
"DenA", "Altai Denisovan (sampling) lineage", sampling_time=T_DenA
),
stdpopsim.Population(
"NeaA", "Altai Neandertal (sampling) lineage", sampling_time=T_NeaA
),
stdpopsim.Population(
"Den1", "Denisovan D1 (introgressing) lineage", sampling_time=None
),
stdpopsim.Population(
"Den2", "Denisovan D2 (introgressing) lineage", sampling_time=None
),
stdpopsim.Population(
"Nea1", "Neandertal N1 (introgressing) lineage", sampling_time=1725
),
stdpopsim.Population("Ghost", "Out-of-Africa lineage", sampling_time=None),
]
pop = {p.id: i for i, p in enumerate(populations)}
citations = [
stdpopsim.Citation(
author="Jacobs et al.",
year=2019,
doi="https://doi.org/10.1016/j.cell.2019.02.035",
reasons={stdpopsim.CiteReason.DEM_MODEL},
),
stdpopsim.Citation(
author="Malaspinas et al.",
year=2016,
doi="https://doi.org/10.1038/nature18299",
reasons={stdpopsim.CiteReason.DEM_MODEL},
),
]
# Inherited from Malaspinas et al., which gives the following refs:
generation_time = 29 # Fenner 2005
# mutation_rate = 1.25e-8 # per gen per site, Scally & Durbin 2012
N_YRI = 48433
N_Ghost = 8516
N_CEU = 6962
N_CHB = 9025
N_Papuan = 8834
N_DenA = 5083
N_Den1 = 13249
N_Den2 = 13249
N_NeaA = 826
N_Nea1 = 13249
pop_meta = {p.id: p.asdict() for p in populations}
population_configurations = [
msprime.PopulationConfiguration(initial_size=N_YRI, metadata=pop_meta["YRI"]),
msprime.PopulationConfiguration(initial_size=N_CEU, metadata=pop_meta["CEU"]),
msprime.PopulationConfiguration(initial_size=N_CHB, metadata=pop_meta["CHB"]),
msprime.PopulationConfiguration(
initial_size=N_Papuan, metadata=pop_meta["Papuan"]
),
msprime.PopulationConfiguration(initial_size=N_DenA, metadata=pop_meta["DenA"]),
msprime.PopulationConfiguration(initial_size=N_NeaA, metadata=pop_meta["NeaA"]),
msprime.PopulationConfiguration(initial_size=N_Den1, metadata=pop_meta["Den1"]),
msprime.PopulationConfiguration(initial_size=N_Den2, metadata=pop_meta["Den2"]),
msprime.PopulationConfiguration(initial_size=N_Nea1, metadata=pop_meta["Nea1"]),
msprime.PopulationConfiguration(
initial_size=N_Ghost, metadata=pop_meta["Ghost"]
),
]
assert len(populations) == len(population_configurations)
# initial migrations
m_Ghost_Afr = 1.79e-4
m_Ghost_EU = 4.42e-4
m_EU_AS = 3.14e-5
m_AS_Papuan = 5.72e-5
# older migrations
m_Eurasian_Papuan = 5.72e-4
m_Eurasian_Ghost = 4.42e-4
migration_matrix = [[0] * len(populations) for _ in range(len(populations))]
migration_matrix[pop["Ghost"]][pop["YRI"]] = m_Ghost_Afr
migration_matrix[pop["YRI"]][pop["Ghost"]] = m_Ghost_Afr
migration_matrix[pop["Ghost"]][pop["CEU"]] = m_Ghost_EU
migration_matrix[pop["CEU"]][pop["Ghost"]] = m_Ghost_EU
migration_matrix[pop["CEU"]][pop["CHB"]] = m_EU_AS
migration_matrix[pop["CHB"]][pop["CEU"]] = m_EU_AS
migration_matrix[pop["CHB"]][pop["Papuan"]] = m_AS_Papuan
migration_matrix[pop["Papuan"]][pop["CHB"]] = m_AS_Papuan
# splits
T_EU_AS_split = 1293
T_Eurasian_Ghost_split = 1758
T_Papuan_Ghost_split = 1784
T_Ghost_Afr_split = 2218
T_NeaA_Nea1_split = 3375
T_DenA_Den1_split = 9750
T_DenA_Den2_split = 12500
T_Den_Nea_split = 15090
T_Afr_Archaic_split = 20225
# bottlenecks
Tb_Eurasia = 1659
Tb_Papua = 1685
Tb_Ghost = 2119
Nb_Eurasia = 2231
Nb_Papua = 243
Nb_Ghost = 1394
# internal branches
N_EU_AS = 12971
N_Ghost_Afr = 41563
N_NeaA_Nea1 = 13249
N_Den_Anc = 100 # S10.i p. 31/45
N_DenA_Den1 = N_Den_Anc
N_DenA_Den2 = N_Den_Anc
N_Den_Nea = 13249
N_Afr_Archaic = 32671
# admixture pulses
m_Den_Papuan = 0.04
p1 = 0.55 # S10.i p. 31/45
m_Den1_Papuan = p1 * m_Den_Papuan
m_Den2_Papuan = (1 - p1) * m_Den_Papuan
m_Nea1_Ghost = 0.024
m_Nea1_Eurasian = 0.011
m_Nea1_Papuan = 0.002
m_Nea1_AS = 0.002
T_Nea1_Ghost_mig = 1853
T_Nea1_Eurasian_mig = 1566
T_Nea1_Papuan_mig = 1412
T_Nea1_AS_mig = 883
# Fig. 4B, and S10.h p. 30/45
T_Den1_Papuan_mig = 29.8e3 / generation_time
T_Den2_Papuan_mig = 45.7e3 / generation_time
demographic_events = [
# human lineage splits
msprime.MassMigration(
time=T_EU_AS_split, source=pop["CEU"], destination=pop["CHB"]
),
msprime.PopulationParametersChange(
time=T_EU_AS_split, initial_size=N_EU_AS, population_id=pop["CHB"]
),
msprime.MassMigration(
time=T_Eurasian_Ghost_split, source=pop["CHB"], destination=pop["Ghost"]
),
msprime.MassMigration(
time=T_Papuan_Ghost_split, source=pop["Papuan"], destination=pop["Ghost"]
),
msprime.MassMigration(
time=T_Ghost_Afr_split, source=pop["Ghost"], destination=pop["YRI"]
),
msprime.PopulationParametersChange(
time=T_Ghost_Afr_split, initial_size=N_Ghost_Afr, population_id=pop["YRI"]
),
# archaic lineage splits
msprime.MassMigration(
time=T_DenA_Den1_split, source=pop["Den1"], destination=pop["DenA"]
),
msprime.PopulationParametersChange(
time=T_DenA_Den1_split, initial_size=N_DenA_Den1, population_id=pop["DenA"]
),
msprime.MassMigration(
time=T_DenA_Den2_split, source=pop["Den2"], destination=pop["DenA"]
),
msprime.PopulationParametersChange(
time=T_DenA_Den2_split, initial_size=N_DenA_Den2, population_id=pop["DenA"]
),
msprime.MassMigration(
time=T_NeaA_Nea1_split, source=pop["Nea1"], destination=pop["NeaA"]
),
msprime.PopulationParametersChange(
time=T_NeaA_Nea1_split, initial_size=N_NeaA_Nea1, population_id=pop["NeaA"]
),
msprime.MassMigration(
time=T_Den_Nea_split, source=pop["NeaA"], destination=pop["DenA"]
),
msprime.PopulationParametersChange(
time=T_Den_Nea_split, initial_size=N_Den_Nea, population_id=pop["DenA"]
),
msprime.MassMigration(
time=T_Afr_Archaic_split, source=pop["DenA"], destination=pop["YRI"]
),
msprime.PopulationParametersChange(
time=T_Afr_Archaic_split,
initial_size=N_Afr_Archaic,
population_id=pop["YRI"],
),
# bottlenecks
msprime.PopulationParametersChange(
time=Tb_Eurasia, initial_size=Nb_Eurasia, population_id=pop["CHB"]
),
msprime.PopulationParametersChange(
time=Tb_Papua, initial_size=Nb_Papua, population_id=pop["Papuan"]
),
msprime.PopulationParametersChange(
time=Tb_Ghost, initial_size=Nb_Ghost, population_id=pop["Ghost"]
),
# migration changes
msprime.MigrationRateChange(
time=T_EU_AS_split, rate=0, matrix_index=(pop["CHB"], pop["CEU"])
),
msprime.MigrationRateChange(
time=T_EU_AS_split, rate=0, matrix_index=(pop["CEU"], pop["CHB"])
),
msprime.MigrationRateChange(
time=T_EU_AS_split, rate=0, matrix_index=(pop["Papuan"], pop["CHB"])
),
msprime.MigrationRateChange(
time=T_EU_AS_split, rate=0, matrix_index=(pop["CHB"], pop["Papuan"])
),
msprime.MigrationRateChange(
time=T_EU_AS_split, rate=0, matrix_index=(pop["Ghost"], pop["CEU"])
),
msprime.MigrationRateChange(
time=T_EU_AS_split, rate=0, matrix_index=(pop["CEU"], pop["Ghost"])
),
msprime.MigrationRateChange(
time=T_EU_AS_split,
rate=m_Eurasian_Papuan,
matrix_index=(pop["CHB"], pop["Papuan"]),
),
msprime.MigrationRateChange(
time=T_EU_AS_split,
rate=m_Eurasian_Papuan,
matrix_index=(pop["Papuan"], pop["CHB"]),
),
msprime.MigrationRateChange(
time=T_EU_AS_split,
rate=m_Eurasian_Ghost,
matrix_index=(pop["CHB"], pop["Ghost"]),
),
msprime.MigrationRateChange(
time=T_EU_AS_split,
rate=m_Eurasian_Ghost,
matrix_index=(pop["Ghost"], pop["CHB"]),
),
# all migrations off
msprime.MigrationRateChange(time=Tb_Eurasia, rate=0),
# admixture pulses
msprime.MassMigration(
time=T_Den1_Papuan_mig,
proportion=m_Den1_Papuan,
source=pop["Papuan"],
destination=pop["Den1"],
),
msprime.MassMigration(
time=T_Den2_Papuan_mig,
proportion=m_Den2_Papuan,
source=pop["Papuan"],
destination=pop["Den2"],
),
msprime.MassMigration(
time=T_Nea1_Ghost_mig,
proportion=m_Nea1_Ghost,
source=pop["Ghost"],
destination=pop["Nea1"],
),
msprime.MassMigration(
time=T_Nea1_Eurasian_mig,
proportion=m_Nea1_Eurasian,
source=pop["CHB"],
destination=pop["Nea1"],
),
msprime.MassMigration(
time=T_Nea1_Papuan_mig,
proportion=m_Nea1_Papuan,
source=pop["Papuan"],
destination=pop["Nea1"],
),
msprime.MassMigration(
time=T_Nea1_AS_mig,
proportion=m_Nea1_AS,
source=pop["CHB"],
destination=pop["Nea1"],
),
]
demographic_events.sort(key=lambda x: x.time)
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
population_configurations=population_configurations,
migration_matrix=migration_matrix,
demographic_events=demographic_events,
)
def _HolsteinFriesan_1M13():
id = "HolsteinFriesian_1M13"
description = "Piecewise size changes in Holstein-Friesian cattle."
long_description = """
The piecewise-constant population size model of Holstein-Friesian cattle
from MacLeod et al. 2013. Population sizes were estimated from inferred
runs of homozygosity, with parameter values taken from Figure 4A by visual
inspection of the plots.
"""
populations = [
stdpopsim.Population(id="Holstein-Friesian", description="Holstein-Friesian"),
]
citations = [_MacLeodEtAl.because(stdpopsim.CiteReason.DEM_MODEL)]
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=_species.generation_time,
population_configurations=[
msprime.PopulationConfiguration(
initial_size=90, growth_rate=0.0166, metadata=populations[0].asdict()
)
],
# Here 'time' should be in generation notation ie. how many
# generations ago were that Ne (effective population size)
# and growth rate.
# Growth rate is "per generation exponential growth rate":
# -alpha= [ln(initial_pop_size/next_stage_pop_size)/generation_span_in_years]
# For example: ln(90/120)/3= -0.095894024
demographic_events=[
msprime.PopulationParametersChange(
time=1,
initial_size=90,
growth_rate=-0.095894024,
population_id=0,
), # Ne 90 to 120
msprime.PopulationParametersChange(
time=4, growth_rate=-0.24465639, population_id=0
), # Ne 120 to 250
msprime.PopulationParametersChange(
time=7, growth_rate=-0.0560787, population_id=0
), # Ne 250 to 350
msprime.PopulationParametersChange(
time=13, growth_rate=-0.1749704, population_id=0
), # Ne 350 to 1000
msprime.PopulationParametersChange(
time=19, growth_rate=-0.0675775, population_id=0
), # Ne 1000 to 1500
msprime.PopulationParametersChange(
time=25, growth_rate=-0.0022129, population_id=0
), # Ne 1500 to 2000
msprime.PopulationParametersChange(
time=155, growth_rate=-0.0007438, population_id=0
), # Ne 2000 to 2500
msprime.PopulationParametersChange(
time=455, growth_rate=-0.0016824, population_id=0
), # Ne 2500 to 3500
msprime.PopulationParametersChange(
time=655, growth_rate=-0.0006301, population_id=0
), # Ne 3500 to 7000
msprime.PopulationParametersChange(
time=1755, growth_rate=-0.0005945, population_id=0
), # Ne 7000 to 10000
msprime.PopulationParametersChange(
time=2355, growth_rate=-0.0005306, population_id=0
), # Ne 10000 to 17000
msprime.PopulationParametersChange(
time=3355, growth_rate=-0.0000434, population_id=0
), # Ne 17000 to 62000
msprime.PopulationParametersChange(
time=33155, growth_rate=-0.0000, population_id=0
), # Ne 62000 (model has "coalesced")
msprime.PopulationParametersChange(
time=933155, growth_rate=-0.0, population_id=0
),
],
)
def _zigzag():
id = "Zigzag_1S14"
description = "Periodic growth and decline."
long_description = """
A validation model used by Schiffels and Durbin (2014) and Terhorst and
Terhorst, Kamm, and Song (2017) with periods of exponential growth and
decline in a single population.
"""
populations = [
stdpopsim.Population("generic",
"Generic expanding and contracting population"),
]
citations = [
stdpopsim.Citation(author="Schiffels and Durbin",
year=2014,
doi="https://doi.org/10.1038/ng.3015",
reasons={stdpopsim.CiteReason.DEM_MODEL})
]
generation_time = 29
N0 = 14312
g_1 = 0.023025
t_1 = 33.333
n_1 = N0
g_2 = -0.005756
t_2 = 133.33
n_2 = N0 / 10
g_3 = 0.0014391
t_3 = 533.33
n_3 = N0
g_4 = -0.00035977
t_4 = 2133.33
n_4 = N0 / 10
g_5 = 8.99448e-5
t_5 = 8533.33
n_5 = N0
n_ancient = N0 / 10
t_ancient = 34133.31
population_configurations = [
msprime.PopulationConfiguration(initial_size=N0,
metadata=populations[0].asdict())
]
demographic_events = [
msprime.PopulationParametersChange(initial_size=n_1,
time=t_1,
growth_rate=g_1),
msprime.PopulationParametersChange(initial_size=n_2,
time=t_2,
growth_rate=g_2),
msprime.PopulationParametersChange(initial_size=n_3,
time=t_3,
growth_rate=g_3),
msprime.PopulationParametersChange(initial_size=n_4,
time=t_4,
growth_rate=g_4),
msprime.PopulationParametersChange(initial_size=n_5,
time=t_5,
growth_rate=g_5),
msprime.PopulationParametersChange(time=t_ancient,
initial_size=n_ancient,
growth_rate=0)
]
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
population_configurations=population_configurations,
demographic_events=demographic_events,
)
def Durvasula2017MSMC():
id = "QC-SouthMiddleAtlas_1D17"
populations = [stdpopsim.Population("SouthMiddleAtlas", "")]
# Both of the following are directly
# converted from MSMC output scaled by A.Thaliana
# mutation rate 7e-9 and 1 generation
# per year.
times = np.array([
6.990000e02,
2.796000e03,
6.068000e03,
9.894000e03,
1.437000e04,
1.960600e04,
2.573000e04,
3.289400e04,
4.127500e04,
5.107700e04,
6.254400e04,
7.595800e04,
9.164800e04,
1.100010e05,
1.314710e05,
1.565840e05,
1.859600e05,
2.203240e05,
2.605200e05,
3.075400e05,
3.625410e05,
4.268790e05,
5.021390e05,
5.901730e05,
6.931510e05,
8.136100e05,
9.545170e05,
1.119341e06,
1.312147e06,
1.537686e06,
1.801500e06,
2.110100e06,
])
sizes = np.array([
4.2252426e07,
4.2252426e07,
6.0323000e04,
7.2174000e04,
4.0591000e04,
2.1158000e04,
2.1442000e04,
3.9942000e04,
7.8908000e04,
1.1113200e05,
1.1074500e05,
9.6283000e04,
8.7661000e04,
8.3932000e04,
8.3829000e04,
9.1813000e04,
1.1164400e05,
1.4345600e05,
1.8157100e05,
2.1733100e05,
2.4140000e05,
2.4698400e05,
2.3859300e05,
2.2822200e05,
2.1775200e05,
1.9801900e05,
1.6521000e05,
1.2179600e05,
1.2179600e05,
7.3989000e04,
7.3989000e04,
7.3989000e04,
])
# The first 8 epochs are "masked" to
# the last Ne at 40kya due to
# the limitations of MSMC to infer
# population size in this range.
#
# Similarly, the last 2 entries
# are set to equal the third last.
#
# Durvasula et al 2017 shows that
# MSMC has power in A.Thaliana
# between 40kya and 1.6Mya.
sizes[:8] = sizes[8]
sizes[30:32] = sizes[30]
demographic_events = []
population_configurations = [
msprime.PopulationConfiguration(initial_size=sizes[0])
]
for i, t in enumerate(times):
demographic_events.append(
msprime.PopulationParametersChange(time=t,
initial_size=sizes[i],
population_id=0))
return stdpopsim.DemographicModel(
id=id,
description=id,
long_description=id,
generation_time=_species.generation_time,
populations=populations,
population_configurations=population_configurations,
demographic_events=demographic_events,
population_id_map=[{
"SouthMiddleAtlas": 0
}] * 33,
mutation_rate=7.1e-9,
)
def _kamm_ancient_eurasia():
id = "AncientEurasia_9K19"
description = "Multi-population model of ancient Eurasia"
long_description = """
This is the best-fitting model of a history of
multiple ancient and present-day human populations
sampled across Eurasia over the past 120,000 years.
The fitting was performed using momi2 (Kamm et al. 2019),
which uses the multi-population site-frequency spectrum
as input data. The model includes a ghost admixture event
(from unsampled basal Eurasians into early European
farmers), and two admixture events where the source is
approximately well-known (from Neanderthals into
Non-Africans and from Western European hunter-gatherers
into modern Sardinians. There are three present-day
populations: Sardinians, Han Chinese and African Mbuti.
Additionally, there are several ancient samples
obtained from fossils dated at different times in
the past: the Altai Neanderthal (Prufer et al. 2014),
a Mesolithic hunter-gatherer (Lazaridis et al. 2014),
a Neolithic early European sample (Lazaridis et al. 2014),
and two Palaeolithic modern humans from Siberia - MA1
(Raghavan et al. 2014) and Ust'Ishim (Fu et al. 2014).
All the ancient samples are represented by a single diploid
genome.
"""
# Sampling times are assuming 25 years per generation
populations = [
stdpopsim.Population(id="Mbuti",
description="Present-day African Mbuti",
sampling_time=0),
# LBK: 8,000 years ago
stdpopsim.Population(id="LBK",
description="Early European farmer (EEF)",
sampling_time=320),
stdpopsim.Population(id="Sardinian",
description="Present-day Sardinian",
sampling_time=0),
# Loschbour: 7,500 years ago
stdpopsim.Population(id="Loschbour",
description="Western hunter-gatherer (WHG)",
sampling_time=300),
# MA1: 24,000 years ago
stdpopsim.Population(id="MA1",
description="Upper Palaeolithic MAl'ta culture",
sampling_time=960),
stdpopsim.Population(id="Han",
description="Present-day Han Chinese",
sampling_time=0),
# Ust Ishim: 45,000 years ago
stdpopsim.Population(id="UstIshim",
description="early Palaeolithic Ust'-Ishim",
sampling_time=1800),
# Altai Neanderthal: 50,000 years ago
stdpopsim.Population(id="Neanderthal",
description="Altai Neanderthal from Siberia",
sampling_time=2000),
stdpopsim.Population(id="BasalEurasian",
description="Basal Eurasians",
sampling_time=None),
]
citations = [
stdpopsim.Citation(author="Kamm et al.",
year=2019,
doi="https://doi.org/10.1080/01621459.2019.1635482",
reasons={stdpopsim.CiteReason.DEM_MODEL})
]
# Times are provided in years, so we convert into generations.
generation_time = 25
# Mutation_rate in Kamm et al. = 1.22e-8
# Effective population sizes
N_Losch = 1920
N_Mbu = 17300
N_Mbu_Losch = 29100
N_Han = 6300
N_Han_Losch = 2340
N_Nean_Losch = 18200
N_Nean = 86.9
N_LBK = 75.7
N_Sard = 15000
N_Sard_LBK = 12000
# Table A.1 has Altai at 50,000 years ago
t_NeaPopSizeChange = 50000 / generation_time
# Unknown but suspected parameters...
N_Basal = N_Losch
N_MA1 = N_Losch
N_Ust = N_Losch
# Population split times
t_Mbu_Losch = 95800 / generation_time
t_Han_Losch = 50400 / generation_time
t_Ust_Losch = 51500 / generation_time
t_Nean_Losch = 696000 / generation_time
t_MA1_Losch = 44900 / generation_time
t_LBK_Losch = 37700 / generation_time
t_Basal_Losch = 79800 / generation_time
t_Sard_LBK = 7690 / generation_time
# Given that we're using best model estimate,
# ghost WHG is directly descended from Loschbour,
# so this parameter is not used
# t_GhostWHG_Losch = 1560 / generation_time
# Admixture times
t_Nean_to_Eurasian = 56800 / generation_time
t_Basal_to_EEF = 33700 / generation_time
t_GhostWHG_to_Sard = 1230 / generation_time
t_NeanGrowth = t_Mbu_Losch - t_NeaPopSizeChange
logdiffNeanGrowth = math.log(N_Nean / N_Nean_Losch)
r_NeanGrowth = logdiffNeanGrowth / t_NeanGrowth
p_Nean_to_Eurasian = 0.0296
p_Basal_to_EEF = 0.0936
p_GhostWHG_to_Sard = 0.0317
# Population IDs: Mbuti = 0; LBK = 1;
# Sardinian = 2; Loschbour = 3; MA1 = 4;
# Han = 5; Ust Ishim = 6; Neanderthal = 7;
# Basal Eurasian = 8
population_configurations = [
msprime.PopulationConfiguration(initial_size=N_Mbu,
metadata=populations[0].asdict()),
msprime.PopulationConfiguration(initial_size=N_LBK,
metadata=populations[1].asdict()),
msprime.PopulationConfiguration(initial_size=N_Sard,
metadata=populations[2].asdict()),
msprime.PopulationConfiguration(initial_size=N_Losch,
metadata=populations[3].asdict()),
msprime.PopulationConfiguration(initial_size=N_MA1,
metadata=populations[4].asdict()),
msprime.PopulationConfiguration(initial_size=N_Han,
metadata=populations[5].asdict()),
msprime.PopulationConfiguration(initial_size=N_Ust,
metadata=populations[6].asdict()),
msprime.PopulationConfiguration(initial_size=N_Nean,
metadata=populations[7].asdict()),
msprime.PopulationConfiguration(initial_size=N_Basal,
metadata=populations[8].asdict())
]
demographic_events = [
# Sardinian receives admixture from Loschbour / WHG
msprime.MassMigration(time=t_GhostWHG_to_Sard,
source=2,
destination=3,
proportion=p_GhostWHG_to_Sard),
# Sardinian merges into LBK / EEF
# Now pop 1: Sardinian-LBK ancestral pop
msprime.MassMigration(time=t_Sard_LBK,
source=2,
destination=1,
proportion=1.0),
# Sardinian-LBK ancestral pop size change
msprime.PopulationParametersChange(time=t_Sard_LBK,
initial_size=N_Sard_LBK,
population_id=1),
# LBK / EEF receives admixture from Basal Eurasians
msprime.MassMigration(time=t_Basal_to_EEF,
source=1,
destination=8,
proportion=p_Basal_to_EEF),
# LBK / EEF merges into Loschbour
msprime.MassMigration(time=t_LBK_Losch,
source=1,
destination=3,
proportion=1.0),
# MA1 merges into Loschbour
msprime.MassMigration(time=t_MA1_Losch,
source=4,
destination=3,
proportion=1.0),
# Neanderthal start change in population size
msprime.PopulationParametersChange(time=t_NeaPopSizeChange,
initial_size=N_Nean,
growth_rate=r_NeanGrowth,
population_id=7),
# Han merges into Loschbour
msprime.MassMigration(time=t_Han_Losch,
source=5,
destination=3,
proportion=1.0),
# Change in population size in Han-Losch ancestral pop
msprime.PopulationParametersChange(time=t_Han_Losch,
initial_size=N_Han_Losch,
population_id=3),
# UstIshim merges into Loschbour
msprime.MassMigration(time=t_Ust_Losch,
source=6,
destination=3,
proportion=1.0),
# Loschbour / Non-Africans receive admixture from Neanderthals
msprime.MassMigration(time=t_Nean_to_Eurasian,
source=3,
destination=7,
proportion=p_Nean_to_Eurasian),
# Basal Eurasians merge into Loschbour / Non-Africans
msprime.MassMigration(time=t_Basal_Losch,
source=8,
destination=3,
proportion=1.0),
# Mbuti merge into Loschbour / Non-Africans
msprime.MassMigration(time=t_Mbu_Losch,
source=0,
destination=3,
proportion=1.0),
# Change in population size in Mbuti-Losch ancestral pop
msprime.PopulationParametersChange(time=t_Mbu_Losch,
initial_size=N_Mbu_Losch,
population_id=3),
# Change in population size in Neanderthal, growth rate 0
msprime.PopulationParametersChange(time=t_Mbu_Losch,
initial_size=N_Nean_Losch,
growth_rate=0,
population_id=7),
# Neanderthal merge into Loschbour / modern humans
msprime.MassMigration(time=t_Nean_Losch,
source=7,
destination=3,
proportion=1.0),
# Ancestral hominin population size change
msprime.PopulationParametersChange(time=t_Nean_Losch,
initial_size=N_Nean_Losch,
population_id=3),
]
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=generation_time,
population_configurations=population_configurations,
demographic_events=demographic_events,
)
def _orangutan():
id = "TwoSpecies_2L11"
description = "Two population orangutan model"
long_description = """
The two orang-utan species, Sumatran (Pongo abelii) and Bornean (Pongo
pygmaeus) inferred from the joint-site frequency spectrum with ten
individuals from each population. This model is an isolation-with-
migration model, with exponential growth or decay in each population
after the split. The Sumatran population grows in size, while the
Bornean population slightly declines.
"""
citations = [_locke2011.because(stdpopsim.CiteReason.DEM_MODEL)]
populations = [
stdpopsim.Population("Bornean", "Pongo pygmaeus (Bornean) population"),
stdpopsim.Population("Sumatran", "Pongo abelii (Sumatran) population"),
]
# Parameters from paper:
# ancestral size, before split
Na = 17934
# time of split
T_split_years = 403149
# get split time in units of generations
generation_time = _species.generation_time
T_split = T_split_years / generation_time
# proportion of ancestral pop to branch B
s = 0.592
# sizes at split
Na_B = 17934 * s
Na_S = 17934 * (1 - s)
# present sizes
N_B = 8805
N_S = 37661
# get growth rates
r_B = -1 * math.log(Na_B / N_B) / T_split
r_S = -1 * math.log(Na_S / N_S) / T_split
# migration rates
m_S_B = 0.395 / 2 / Na
m_B_S = 0.239 / 2 / Na
# Population IDs correspond to their indexes in the population
# configuration array. Therefore, we have 0=B and 1=S
# initially.
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
citations=citations,
populations=populations,
generation_time=generation_time,
population_configurations=[
msprime.PopulationConfiguration(
initial_size=N_B,
growth_rate=r_B,
metadata=populations[0].asdict()), # NOQA
msprime.PopulationConfiguration(
initial_size=N_S,
growth_rate=r_S,
metadata=populations[1].asdict()), # NOQA
],
migration_matrix=[
[0, m_B_S], # NOQA
[m_S_B, 0], # NOQA
],
demographic_events=[
# Merge populations and turn off migration, change to size Na
msprime.MassMigration(time=T_split,
source=1,
destination=0,
proportion=1.0),
msprime.MigrationRateChange(time=T_split, rate=0),
msprime.PopulationParametersChange(time=T_split,
initial_size=Na,
growth_rate=0,
population_id=0),
],
)
def _HolsteinFriesian_1M13():
id = "HolsteinFriesian_1M13"
description = "Piecewise size changes in Holstein-Friesian cattle."
long_description = """
The piecewise-constant population size model of Holstein-Friesian cattle
from MacLeod et al. (2013). Population sizes were estimated from inferred
runs of homozygosity, with parameter values taken from Figure 4A and Table S1.
"""
populations = [
stdpopsim.Population(id="Holstein_Friesian",
description="Holstein-Friesian"),
]
citations = [
stdpopsim.Citation(
author="MacLeod et al.",
year=2013,
doi="https://doi.org/10.1093/molbev/mst125",
reasons={stdpopsim.CiteReason.DEM_MODEL},
)
]
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=long_description,
populations=populations,
citations=citations,
generation_time=_species.generation_time,
mutation_rate=1e-8,
population_configurations=[
# 3 generations at 90, 1- 3
msprime.PopulationConfiguration(initial_size=90,
metadata=populations[0].asdict())
],
# Here 'time' should be in generation notation ie. how many
# generations ago were that Ne (effective population size)
demographic_events=[
# 3 generations at 120, 4- 6
msprime.PopulationParametersChange(time=4,
initial_size=120,
population_id=0),
# 6 generations at 250, 7- 12
msprime.PopulationParametersChange(time=7,
initial_size=250,
population_id=0),
# 6 generations at 350, 13- 18
msprime.PopulationParametersChange(time=13,
initial_size=350,
population_id=0),
# 6 generations at 1000, 19- 24
msprime.PopulationParametersChange(time=19,
initial_size=1000,
population_id=0),
# 130 generations at 1500, 25- 154
msprime.PopulationParametersChange(time=25,
initial_size=1500,
population_id=0),
# 200 generations at 2000, 155- 454
msprime.PopulationParametersChange(time=155,
initial_size=2000,
population_id=0),
# 200 generations at 2500, 455- 654
msprime.PopulationParametersChange(time=455,
initial_size=2500,
population_id=0),
# 1100 generations at 3500, 655- 1754
msprime.PopulationParametersChange(time=655,
initial_size=3500,
population_id=0),
# 600 generations at 7000, 1755- 2354
msprime.PopulationParametersChange(time=1755,
initial_size=7000,
population_id=0),
# 1000 generations at 10000, 2355- 3354
msprime.PopulationParametersChange(time=2355,
initial_size=10000,
population_id=0),
# 29800 generations at 17000, 3355- 33154
msprime.PopulationParametersChange(time=3355,
initial_size=17000,
population_id=0),
# 900000 generations at 62000, 33155-933154
msprime.PopulationParametersChange(time=33155,
initial_size=62000,
population_id=0),
],
)
def hominin_composite_archaic_africa():
dm = hominin_composite()
id = "HomininComposite2_4G20"
description = "HomininComposite2_4G20 plus archaic lineage in Africa"
generation_time = 29
# Ragsdale & Gravel 2019
N_0 = 3600
T_arch_afr_split = 499e3 / generation_time
T_arch_afr_mig = 125e3 / generation_time
T_arch_adm_end = 18.7e3 / generation_time
m_AF_arch_af = 1.98e-5
T_YRI_CEU_split = 65.7e3 / generation_time
populations = dm.populations + [
stdpopsim.Population(
"ArchaicAFR", "Putative Archaic Africans", sampling_time=None
),
]
population_configurations = dm.population_configurations + [
msprime.PopulationConfiguration(
initial_size=N_0, metadata=populations[-1].asdict()
),
]
pop = {p.id: i for i, p in enumerate(populations)}
demographic_events = dm.demographic_events + [
# migration turned on between Yoruban and archaic African populations
msprime.MigrationRateChange(
time=T_arch_adm_end,
rate=m_AF_arch_af,
matrix_index=(pop["YRI"], pop["ArchaicAFR"]),
),
msprime.MigrationRateChange(
time=T_arch_adm_end,
rate=m_AF_arch_af,
matrix_index=(pop["ArchaicAFR"], pop["YRI"]),
),
# YRI merges into Anc: turn off migration between YRI and ArchaicAFR.
msprime.MigrationRateChange(
time=T_YRI_CEU_split,
rate=0,
matrix_index=(pop["YRI"], pop["ArchaicAFR"]),
),
msprime.MigrationRateChange(
time=T_YRI_CEU_split,
rate=0,
matrix_index=(pop["ArchaicAFR"], pop["YRI"]),
),
# migration turned on between African and archaic African populations
msprime.MigrationRateChange(
time=T_YRI_CEU_split,
rate=m_AF_arch_af,
matrix_index=(pop["Anc"], pop["ArchaicAFR"]),
),
msprime.MigrationRateChange(
time=T_YRI_CEU_split,
rate=m_AF_arch_af,
matrix_index=(pop["ArchaicAFR"], pop["Anc"]),
),
# Beginning of migration between African and archaic African populations
msprime.MigrationRateChange(
time=T_arch_afr_mig,
rate=0,
matrix_index=(pop["Anc"], pop["ArchaicAFR"]),
),
msprime.MigrationRateChange(
time=T_arch_afr_mig,
rate=0,
matrix_index=(pop["ArchaicAFR"], pop["Anc"]),
),
# Archaic African merges with moderns
msprime.MassMigration(
time=T_arch_afr_split,
source=pop["ArchaicAFR"],
destination=pop["Anc"],
proportion=1.0,
),
]
demographic_events.sort(key=lambda x: x.time)
return stdpopsim.DemographicModel(
id=id,
description=description,
long_description=dm.long_description,
populations=populations,
citations=dm.citations.copy(),
generation_time=generation_time,
population_configurations=population_configurations,
demographic_events=demographic_events,
)
def LockePongo():
id = "QC-TwoSpecies_2L11"
populations = [
stdpopsim.Population("Bornean", ""),
stdpopsim.Population("Sumatran", ""),
]
# This is a split-migration style model, with exponential growth or
# decay allowed in each population after the split. They assumed a
# generation time of 20 years and a mutation rate of 2e-8 per bp per gen
generation_time = 20
mutation_rate = 2e-8
# Parameters given in Table S21-2
Ne = 17934
s = 0.592
NB0 = s * Ne
NS0 = (1 - s) * Ne
NBF = 8805
NSF = 37661
mSB = 0.395 / 2 / Ne
mBS = 0.239 / 2 / Ne
T = 403149 / generation_time
rB = np.log(NBF / NB0) / T
rS = np.log(NSF / NS0) / T
return stdpopsim.DemographicModel(
id=id,
description=id,
long_description=id,
generation_time=generation_time,
populations=populations,
# pop 0 is Bornean, pop 1 is Sumatran
population_configurations=[
msprime.PopulationConfiguration(initial_size=NBF, growth_rate=rB),
msprime.PopulationConfiguration(initial_size=NSF, growth_rate=rS),
],
migration_matrix=[[0, mBS], [mSB, 0]],
demographic_events=[
# merge, turn off migration, change size and growth rate
msprime.MassMigration(source=1,
destination=0,
time=T,
proportion=1),
msprime.MigrationRateChange(time=T, rate=0),
msprime.PopulationParametersChange(time=T,
initial_size=Ne,
growth_rate=0,
population_id=0),
],
population_id_map=[
{
"Bornean": 0,
"Sumatran": 1
},
{
"Bornean": 0,
"Sumatran": 1
},
],
mutation_rate=mutation_rate,
)
def _sma_1pop():
# the size during the interval times[k] to times[k+1] = sizes[k]
times = np.array(
[
699,
2796,
6068,
9894,
14370,
19606,
25730,
32894,
41275,
51077,
62544,
75958,
91648,
110001,
131471,
156584,
185960,
220324,
260520,
307540,
362541,
426879,
502139,
590173,
693151,
813610,
954517,
1119341,
1312147,
1537686,
1801500,
2110100,
]
)
sizes = np.array(
[
42252426,
42252426,
60323,
72174,
40591,
21158,
21442,
39942,
78908,
111132,
110745,
96283,
87661,
83932,
83829,
91813,
111644,
143456,
181571,
217331,
241400,
246984,
238593,
228222,
217752,
198019,
165210,
121796,
121796,
73989,
73989,
73989,
]
)
# MSMC is accurate from 40Kya-1.6Mya for A.thaliana (Durvasula et al 2017)
# set the first 7 sizes
# equal to the size at 8 (~40Kya)
sizes[:8] = sizes[8]
# set the last 2 entries equal
# to the size at 30 (~1.6Mya)
sizes[30:32] = sizes[30]
demographic_events = []
for sz, t in zip(sizes, times):
demographic_events.append(
msprime.PopulationParametersChange(time=t, initial_size=sz, population_id=0)
)
populations = [
stdpopsim.Population(
id="SouthMiddleAtlas",
description="Arabidopsis Thaliana South Middle Atlas population",
)
]
return stdpopsim.DemographicModel(
id="SouthMiddleAtlas_1D17",
description="South Middle Atlas piecewise constant size",
long_description="""
This model comes from MSMC using two randomly sampled homozygous
individuals (Khe32 and Ifr4) from the South Middle Atlas region
from the Middle Atlas Mountains in Morocco. The model is estimated
with 32 time periods. Because estimates from the recent and ancient
past are less accurate, we set the population size in the first 7
time periods equal to the size at the 8th time period and the size
during last 2 time periods equal to the size in the 30th time
period.
""",
populations=populations,
citations=[
stdpopsim.Citation(
author="Durvasula et al.",
year=2017,
doi="https://doi.org/10.1073/pnas.1616736114",
reasons={stdpopsim.CiteReason.DEM_MODEL},
)
],
generation_time=1,
demographic_events=demographic_events,
population_configurations=[
msprime.PopulationConfiguration(
initial_size=sizes[0], metadata=populations[0].asdict()
)
],
)
def test_population_construction_popconfig(self):
pc = [msprime.PopulationConfiguration(initial_size=1, growth_rate=0.03)]
dm = stdpopsim.DemographicModel(
id="", description="", long_description="",
generation_time=1, population_configurations=pc)
self.assertEqual(dm.populations[0].id, "pop0")
