def createAge(pop):
ageInitOps = [
# InitInfo(lambda: random.randint(0, cfg.ages-2), infoFields='age'),
sp.IdTagger(),
# PyOperator(func=outputAge,at=[0]),
sp.PyOperator(func=setAge, at=[0]),
]
agePreOps = [
sp.InfoExec("age += 1"),
sp.InfoExec("mate = -1"),
sp.InfoExec("force_skip = 0"),
sp.PyOperator(func=outputAge),
]
mySubPops = []
for age in range(cfg.ages - 2):
mySubPops.append((0, age + 1))
mateOp = sp.HeteroMating([
sp.HomoMating(
sp.PyParentsChooser(fitnessGenerator if cfg.doNegBinom
else (litterSkipGenerator if cfg.Nb is None else
restrictedGenerator)),
sp.OffspringGenerator(numOffspring=1, ops=[
sp.MendelianGenoTransmitter(), sp.IdTagger(),
sp.PedigreeTagger()],
sexMode=(sp.PROB_OF_MALES, cfg.maleProb)), weight=1),
sp.CloneMating(subPops=mySubPops, weight=-1)],
subPopSize=calcDemo)
agePostOps = [
sp.PyOperator(func=outputMega),
sp.PyOperator(func=cull),
]
pop.setVirtualSplitter(sp.InfoSplitter(field='age',
cutoff=list(range(1, cfg.ages))))
return ageInitOps, agePreOps, mateOp, agePostOps
def LC_evolve(popSize, alleleFreq, diseaseModel):
'''
'''
pop = sim.Population(
size=popSize,
loci=[1] * len(alleleFreq),
infoFields=['age', 'smoking', 'age_death', 'age_LC', 'LC'])
pop.setVirtualSplitter(
sim.CombinedSplitter(splitters=[
sim.InfoSplitter(field='age',
cutoff=[20, 40],
names=['youngster', 'adult', 'senior']),
sim.SexSplitter(),
sim.InfoSplitter(field='smoking',
values=[0, 1, 2],
names=['nonSmoker', 'smoker', 'formerSmoker'])
]))
pop.evolve(
initOps=[sim.InitSex(),
sim.InitInfo(range(75), infoFields='age')] + [
sim.InitGenotype(freq=[1 - f, f], loci=i)
for i, f in enumerate(alleleFreq)
] + [
sim.PyOperator(func=diseaseModel.initialize),
],
preOps=[
sim.InfoExec('age += 1'),
# die of lung cancer or natural death
sim.DiscardIf('age > age_death')
],
matingScheme=sim.HeteroMating(
[
sim.CloneMating(weight=-1),
sim.RandomMating(ops=[
sim.MendelianGenoTransmitter(),
sim.PyOperator(func=diseaseModel.initialize)
],
subPops=[(0, 'adult')])
],
subPopSize=lambda pop: pop.popSize() + popSize / 75),
postOps=[
# update individual, currently ding nothing.
sim.PyOperator(func=diseaseModel.updateStatus),
# determine if someone has LC at his or her age
sim.InfoExec('LC = age >= age_LC'),
# get statistics about COPD and LC prevalence
sim.Stat(pop,
meanOfInfo='LC',
subPops=[(0, sim.ALL_AVAIL)],
vars=['meanOfInfo', 'meanOfInfo_sp']),
sim.PyEval(
r"'Year %d: Overall %.2f%% M: %.2f%% F: %.2f%% "
r"NS: %.1f%%, S: %.2f%%\n' % (gen, meanOfInfo['LC']*100, "
r"subPop[(0,3)]['meanOfInfo']['LC']*100,"
r"subPop[(0,4)]['meanOfInfo']['LC']*100,"
r"subPop[(0,5)]['meanOfInfo']['LC']*100,"
r"subPop[(0,6)]['meanOfInfo']['LC']*100)"),
],
gen=100)
def simulation(self):
self.pop = sim.Population(size = [500, 500], loci=[1]*20,
infoFields = ["age",'ind_id', 'father_idx', 'mother_idx', "hc", "ywc",'migrate_to'],
subPopNames = ["croatia", "slovenia"])
sim.initInfo(pop = self.pop, values = list(map(int, np.random.negative_binomial(n = 1, p = 0.25, size=500))), infoFields="age")
self.pop.setVirtualSplitter(sim.CombinedSplitter([
sim.ProductSplitter([
sim.SexSplitter(),
sim.InfoSplitter(field = "age", cutoff = [1,3,6,10])])],
vspMap = [[0,1], [2], [3], [4], [5,6,7,8], [9] ]))
# Age groups: from 0 to 1 - cubs, from 1 to 3 - prereproductive, from 3 to 6 - reproductive class, from 6 to 10 - dominant
self.pop.evolve(
initOps=[
sim.InitSex(),
# random genotype
sim.InitGenotype(freq=[0.01]*2 + [0.03]*2 + [0.23]*4),
# assign an unique ID to everyone.
sim.IdTagger(),
],
# increase the age of everyone by 1 before mating.
preOps=[sim.InfoExec('age += 1'),
sim.InfoExec("hc +=1 if 0 < hc < 3 else 0"), # Mother bear can't have cubs for two years after pregnancy
sim.Migrator(rate=[[self.cro_to_slo]],
mode=sim.BY_PROPORTION,
subPops=[(0, 0)],
toSubPops=[1]), # reproductive males migrate from Cro to Slo
sim.Migrator(rate=[[self.slo_to_cro]],
mode=sim.BY_PROPORTION,
subPops=[(1, 0)],
toSubPops=[0]),
sim.Stat(effectiveSize=sim.ALL_AVAIL, subPops=[(0,1),(0,2),(0,4), (1,1), (1,2), (1,4)], vars='Ne_demo_base'),
sim.Stat(effectiveSize=sim.ALL_AVAIL,subPops=[(0,1),(0,2),(0,4), (1,1), (1,2), (1,4)], vars='Ne_demo_base_sp')
#sim.PyEval(r'"Cro %d, Slo %d' ' % (Cro, Slo)', "Cro = pop.subPopSize(0)" "Slo = pop.subPopSize(1)",exposePop='pop'),
],
matingScheme=sim.HeteroMating([
# CloneMating will keep individual sex and all
# information fields (by default).
# The age of offspring will be zero.
sim.HomoMating(subPops=sim.ALL_AVAIL,
chooser=sim.CombinedParentsChooser(
fatherChooser=sim.PyParentsChooser(generator=self.bearFather),
motherChooser=sim.PyParentsChooser(generator=self.bearMother)
),
generator=sim.OffspringGenerator(ops=[
sim.InfoExec("age = 0"),
sim.IdTagger(),
#sim.PedigreeTagger(),
sim.ParentsTagger(),
sim.MendelianGenoTransmitter()
], numOffspring=(sim.UNIFORM_DISTRIBUTION, 1, 3))),
sim.CloneMating(subPops=[(0,0), (0,1), (0,2), (0,4), (1,0), (1,1), (1,2), (1,4)], weight=-1),
], subPopSize=popmodel.demoModel),
# number of individuals?
postOps = [
#sim.PyOperator(func=popmodel.NaturalMortality),
sim.PyOperator(func = popmodel.CalcNe, param={"me":self.me, "Ne":self.Ne}, begin=int(0.2*self.generations)),
sim.PyOperator(func = popmodel.CalcLDNe, param={"me":self.me, "x":self.x}, begin=int(0.2*self.generations)),
sim.PyOperator(func=popmodel.cullCountry,param={"slo_cull": self.slo_cull, "cro_cull": self.cro_cull}),
],
gen = self.generations
)
def runSimulation(scenario_id, sub_population_size, minMatingAge, maxMatingAge,
gen):
'''
sub_population_size A vector giving the population sizes for each sub-population. The subpopulations determine which breeding ground an individual belongs to
minMatingAge minimal mating age.
maxMatingAge maximal mating age. Individuals older than this are effectively dead
years number of years to simulate
'''
# scenario_id describes the batch of files to load
# The mitochondrial DNA will be in mtdna_<scenario_id>
# The SNP DNA will be in snp_<scenario_id>
# Read the mitochondrial haplotype frequencies. There's a bit to unpack here
# We read the lines into an array, and for each one, call split() on it to get one element per column.
# However, we do not want this - we want the transpose, where haplotype_frequencies[0] is a vector of
# all the frequencies for population 0, and haplotype_frequencies[1] is the corresponding vector for
# population 2. list(map(list, zip(*t))) will achieve this transformation for us.
# While we are at it, we also convert the strings into floats.
mitochondrial_file = "mtdna_" + scenario_id + ".txt"
with open(mitochondrial_file, "r") as fd:
haplotype_frequencies = list(
map(list,
zip(*[list(map(float, line[0:-1].split())) for line in fd])))
if len(haplotype_frequencies) != len(sub_population_size):
raise ValueError(
'The number of populations in the population size vector and the number of populations deduced from the haplotype file are different'
)
# Now read the SNP data. This builds a 2D array indexed as snp[locus][population]
snp_file = "snp_" + scenario_id + ".txt"
with open(snp_file, "r") as fd:
snp = [list(map(float, line[0:-1].split())) for line in fd]
sub_population_count = len(sub_population_size)
print()
print(sub_population_count, "subpopulations detected")
# Now we can create the population. We want to give each population a population name, starting from A
sub_population_names = list(map(chr, range(65, 65 + sub_population_count)))
# We have two chromosomes. The first is an autosome with nb_loci loci, and the second is the mitochondrial chromosome with 1 locus
pop = simuPOP.Population(
sub_population_size,
ploidy=2,
loci=[nb_loci, 1],
ancGen=2,
infoFields=[
'age', 'ind_id', 'father_id', 'mother_id', 'nitrogen', 'carbon',
'feeding_ground', 'native_breeding_ground', 'migrate_to'
],
subPopNames=sub_population_names,
chromTypes=[simuPOP.AUTOSOME, simuPOP.MITOCHONDRIAL])
sub_population_names = tuple(sub_population_names)
# Create an attribute on each individual called 'age'. Set it to a random number between 0 and maxMatingAge
# Note that size is a vector - the size of each population. We have to sum these to get the total number of individuals
individual_count = sum(sub_population_size)
# Assign a random age to each individual
pop.setIndInfo(
[random.randint(0, maxMatingAge) for x in range(individual_count)],
'age')
# Assign a random feeding ground to each individual
pop.setIndInfo([
random.randint(0, numberOfFeedingGrounds - 1)
for x in range(individual_count)
], 'feeding_ground')
# Currently we have these virtual subpopulations:
# age < minMatingAge (juvenile)
# age >= minMatingAge and age < maxMatingAge + 0.1 (age <= maxMatingAge) (mature)
# age >= maxMatingAge (dead)
#
# Ideally we would want something like this:
# 1) Immature
# 2) Receptive female (every 3 years)
# 3) Non-receptive female
# 4) Mature male
# 5) Dead
#
# Note that we use a cutoff InfoSplitter here, it is also possible to
# provide a list of values, each corresponding to a virtual subpopulation.
pop.setVirtualSplitter(
simuPOP.CombinedSplitter([
simuPOP.ProductSplitter([
simuPOP.SexSplitter(),
simuPOP.InfoSplitter('age',
cutoff=[minMatingAge, maxMatingAge + 0.1],
names=['juvenile', 'mature', 'dead'])
])
],
vspMap=[[0], [1], [2], [3], [4], [5],
[0, 1, 3, 4], [1, 4]],
names=[
'Juvenile Male', 'Mature Male',
'Dead Male', 'Juvenile Female',
'Mature Female', 'Dead Female',
'Not dead yet', 'Active'
]))
pop.evolve(
initOps=[
simuPOP.InitSex(),
simuPOP.IdTagger(),
simuPOP.PyOperator(func=init_native_breeding_grounds)
] + [
simuPOP.InitGenotype(subPops=sub_population_names[i],
freq=haplotype_frequencies[i],
loci=[nb_loci])
for i in range(0, sub_population_count)
] + [
simuPOP.InitGenotype(subPops=sub_population_names[i],
freq=[snp[n][i], 1 - snp[n][i]],
loci=[n])
for i in range(0, sub_population_count)
for n in range(0, nb_loci - 1)
],
# increase age by 1
preOps=[simuPOP.InfoExec('age += 1')],
matingScheme=simuPOP.HeteroMating(
[
# age <= maxAge, copy to the next generation (weight=-1)
# subPops is a list of tuples that will participate in mating. The tuple is a pair (subPopulation, virtualSubPopulation)
# First, we propagate (clone) all individuals in all subpopulations (and all VSPs except the ones who are now in the VSP of deceased individuals) to the next generation
simuPOP.CloneMating(
ops=[simuPOP.CloneGenoTransmitter(chroms=[0, 1])],
subPops=[
(sub_population, 6)
for sub_population in range(0, sub_population_count)
],
weight=-1),
# Then we simulate random mating only in VSP 1 (ie reproductively mature individuals) within subpopulation (breeding/winter grounds)
simuPOP.RandomMating(
ops=[
simuPOP.MitochondrialGenoTransmitter(),
simuPOP.MendelianGenoTransmitter(),
simuPOP.IdTagger(),
simuPOP.InheritTagger(mode=simuPOP.MATERNAL,
infoFields=['feeding_ground']),
simuPOP.InheritTagger(
mode=simuPOP.MATERNAL,
infoFields=['native_breeding_ground']),
simuPOP.PedigreeTagger()
],
subPops=[
(sub_population, 7)
for sub_population in range(0, sub_population_count)
],
weight=1)
],
subPopSize=configure_new_population_size),
postOps=[
# Determine the isotopic ratios in individuals
simuPOP.PyOperator(func=postop_processing),
simuPOP.Migrator(mode=simuPOP.BY_IND_INFO),
# count the individuals in each virtual subpopulation
#simuPOP.Stat(popSize=True, subPops=[(0,0), (0,1), (0,2), (1,0), (1, 1), (1, 2)]),
# print virtual subpopulation sizes (there is no individual with age > maxAge after mating)
#simuPOP.PyEval(r"'Size of age groups: %s\n' % (','.join(['%d' % x for x in subPopSize]))")
# Alternatively, calculate the Fst
# FIXME: How does this actually work? Does it work for > 2 populations? I don't really understand it yet
# ELC: it is a calculation that partitions variance among and between populations, and can be calculated as a
# global statistic or on a pairwise basis. We use it as an indication of genetic differentiation.
simuPOP.Stat(structure=range(1),
subPops=sub_population_names,
suffix='_AB',
step=10),
simuPOP.PyEval(r"'Fst=%.3f \n' % (F_st_AB)", step=10)
],
gen=years)
#simuPOP.dump(pop, width=3, loci=[], subPops=[(simuPOP.ALL_AVAIL, simuPOP.ALL_AVAIL)], max=1000, structure=False);
#return
ped = simuPOP.Pedigree(pop)
print("This is the pedigree stuff")
simuPOP.dump(pop)
# Now sample the individuals
sample = drawRandomSample(pop, sizes=[sample_count] * sub_population_count)
# Print out the allele frequency data
simuPOP.stat(sample, alleleFreq=simuPOP.ALL_AVAIL)
frequencies = sample.dvars().alleleFreq
with open('freq.txt', 'w') as freqfile:
index = 0
for locus in frequencies:
if (locus == nb_loci):
continue
if (len(frequencies[locus]) < 2):
continue
print(index, end=' ', file=freqfile)
index = index + 1
for allele in frequencies[locus]:
print(frequencies[locus][allele], end=' ', file=freqfile)
print(file=freqfile)
# We want to remove monoallelic loci. This means a position in the genotype for which all individuals have the same value in both alleles
# To implement this we will build up a list of loci that get ignored when we dump out the file. Generally speaking, if we add all the values up
# then either they will sum to 0 (if all individuals have type 0) or to the number of individuals * 2 (if all individuals have type 1)
geno_sum = [0] * (nb_loci + 1) * 2
for individual in sample.individuals():
geno_sum = list(map(add, geno_sum, individual.genotype()))
final_sum = list(
map(add, geno_sum[:(nb_loci + 1)], geno_sum[(nb_loci + 1):]))
monoallelic_loci = []
for i in range(0, nb_loci):
if final_sum[i] == 0 or final_sum[
i] == sample_count * sub_population_count * 2:
monoallelic_loci = [i] + monoallelic_loci
monoallelic_loci = sorted(monoallelic_loci, reverse=True)
nb_ignored_loci = len(monoallelic_loci)
# Generate the two files
with open('mixfile.txt', 'w') as mixfile:
with open('haploiso.txt', 'w') as haplofile:
print(sub_population_count,
nb_loci - nb_ignored_loci,
2,
1,
file=mixfile)
print("sex, haplotype, iso1, iso2, native_ground", file=haplofile)
for i in range(0, nb_loci - nb_ignored_loci):
print('Loc', i + 1, sep='_', file=mixfile)
for individual in sample.individuals():
genotype = individual.genotype()
print(
1 if individual.sex() == 1 else 0,
genotype[nb_loci],
individual.info('carbon'),
individual.info('nitrogen'),
# int(individual.info('native_breeding_ground')),
file=haplofile,
sep=' ')
print(int(individual.info('native_breeding_ground') + 1),
end=' ',
file=mixfile)
for i in range(0, nb_loci):
if i not in monoallelic_loci:
print(genotype[i] + 1,
genotype[i + nb_loci + 1] + 1,
' ',
end='',
sep='',
file=mixfile)
print(file=mixfile)
return sample
sim.InitSex(),
# random assign age
sim.InitInfo(lambda: random.randint(0, 75), infoFields='age'),
# random genotype
sim.InitGenotype(freq=[0.5, 0.5]),
# assign an unique ID to everyone.
sim.IdTagger(),
sim.PyOutput('Prevalence of disease in each age group:\n'),
],
# increase the age of everyone by 1 before mating.
preOps=sim.InfoExec('age += 1'),
matingScheme=sim.HeteroMating([
# all individuals with age < 75 will be kept. Note that
# CloneMating will keep individual sex, affection status and all
# information fields (by default).
sim.CloneMating(subPops=[(0,0), (0,1), (0,2)], weight=-1),
# only individuals with age between 20 and 50 will mate and produce
# offspring. The age of offspring will be zero.
sim.RandomMating(ops=[
sim.IdTagger(), # give new born an ID
sim.PedigreeTagger(), # track parents of each individual
sim.MendelianGenoTransmitter(), # transmit genotype
],
numOffspring=(sim.UNIFORM_DISTRIBUTION, 1, 3),
subPops=[(0,1)]),],
subPopSize=demoModel),
# number of individuals?
postOps=[
sim.PyPenetrance(func=pene, loci=0),
sim.PyOperator(func=outputstat, step=20)
],
assert (node_times[p] >= last_time)
last_time = node_times[p]
assert (p < args.tables.nodes.num_rows)
for ch in edges.child:
assert (ch < args.tables.nodes.num_rows)
@pytest.fixture(
scope="function",
params=[
lambda recombinator, popsize, id_tagger: sim.RandomMating(
ops=[id_tagger, recombinator]),
# Overlapping generations mating system -- popsize grows by 2x
lambda recombinator, popsize, id_tagger: sim.HeteroMating([
sim.RandomMating(ops=[id_tagger, recombinator]),
sim.CloneMating()
],
subPopSize=
popsize * 2),
# Overlapping generations mating system -- popsize grows by 5x
lambda recombinator, popsize, id_tagger: sim.HeteroMating([
sim.RandomMating(ops=[id_tagger, recombinator]),
sim.CloneMating()
],
subPopSize=
popsize * 5),
])
def make_pop(request):
# request.param stores a lambda function to make mating scheme
# each test that uses this fixture will be run for both entries in 'params'
mating_scheme_factory = request.param
initOps = [
sim.InitInfo([0], infoFields = 'age'),
sim.InitInfo([args.a], infoFields = 'a'),
sim.InitInfo([args.b], infoFields = 'b'),
sim.InitInfo(lambda: random.random(), infoFields = 'luck'),
sim.InfoExec("t0 = -ind.b / ind.a", exposeInd = 'ind'),
sim.InfoExec("smurf = 1 if (model == 'two_phases' and ind.age > ind.t0 and ind.luck <= 1.0 - math.exp(-ind.a * ind.age + ind.a * ind.t0 - ind.a / 2.0)) else 0", exposeInd = 'ind'),
sim.PyExec("Surviving = {'larvae': [], 'adults': [], 'smurfs': []}")
],
preOps = [
sim.InfoExec("luck = random.random()"),
sim.InfoExec("smurf = 1 if ((ind.smurf == 1) or (model == 'two_phases' and ind.age > ind.t0 and ind.luck <= 1.0 - math.exp(-ind.a * ind.age + ind.a * ind.t0 - ind.a / 2.0))) else 0", exposeInd='ind'),
sim.DiscardIf(aging_model(args.model)),
sim.InfoExec("age += 1")
],
matingScheme = sim.CloneMating(subPops = sim.ALL_AVAIL, subPopSize = demo),
postOps = [
sim.Stat(popSize=True, subPops=[(0,0), (0,1), (0,2)]),
sim.PyExec("Surviving['larvae'].append(subPopSize[0])"),
sim.PyExec("Surviving['adults'].append(subPopSize[1])"),
sim.PyExec("Surviving['smurfs'].append(subPopSize[2])"),
# sim.PyEval(r'"{:d}\t{:d}\t{:d}\t{:d}\n".format(gen, subPopSize[0], subPopSize[1], subPopSize[2])', step=1),
sim.TerminateIf('popSize == 0')
],
gen=args.G
)
args.output.write("#Gen\t" + "\t".join(['Larvae\tAdults\tSmurfs' for a in range(args.replicates)]) + "\n")
for day in range(args.G):
line = "{:3d}".format(day + 1)
for rep in range(args.replicates):
],
matingScheme=sim.HeteroMating(
[
# only adult individuals with age >=3 will mate and produce
# offspring. The age of offspring will be zero.
sim.RandomMating(ops=[
sim.MendelianGenoTransmitter(),
sim.Recombinator(intensity=0.1)
],
subPops=[(sim.ALL_AVAIL, '3 <= age < 9'),
(sim.ALL_AVAIL, '9 <= age < 17')],
weight=-0.1),
# individuals with age < 17 will be kept, but might be removed due to
# population size decline
sim.CloneMating(subPops=[(
sim.ALL_AVAIL,
'age < 3'), (sim.ALL_AVAIL,
'3 <= age < 9'), (sim.ALL_AVAIL, '9 <= age < 17')]
),
],
subPopSize=demoModel),
postOps=[
sim.Stat(popSize=True),
sim.PyEval(r'f"{gen} {subPopSize}\n"'),
sim.utils.Exporter(format='GENEPOP',
step=10,
output='!f"{gen}.pop"',
gui='batch')
],
gen=150)
def simulateBySimuPOP():
#starting variables
directory = '/data/new/javi/toxo/simulations4/'
input_path = 'Toxo20.txt'
output_path = 'SimulatedToxo.txt'
input_path = directory + input_path
output_path = directory + output_path
parents_path = directory + '/parents.txt'
pedigree_path = directory + 'pedigree.txt'
number_of_ancestors = 3
expansion_pop_size = 15
offsprings_sampled = number_of_ancestors + expansion_pop_size
gen = 3
translate_mode = 'toxoplasma'
structure_mode = 'simupop'
#parsing input
init_info = parseSNPInput(input_path, number_of_ancestors)
ancestral_genomes = init_info[0]
ancestor_names = ancestral_genomes.keys()
loci_positions = init_info[1]
chromosome_names = sorted(loci_positions.keys(),
key=lambda x: cns.getValue(x, translate_mode))
list_of_loci = [len(loci_positions[chr]) for chr in chromosome_names]
lociPos = fc.reduce(lambda x, y: x + y,
[loci_positions[x] for x in chromosome_names])
sp.turnOnDebug(code="DBG_GENERAL")
#initializing
print('Initializaing Population')
population = sp.Population(size=[number_of_ancestors], loci=list_of_loci, ancGen = 5, lociPos = lociPos, \
chromNames = chromosome_names, lociNames = [], alleleNames = ['A','T','G','C'],\
infoFields=['name', 'ind_id', 'father_id', 'mother_id'])
for individual, sample, ind_id in zip(population.individuals(),
ancestral_genomes,
range(len(ancestral_genomes))):
individual.setInfo(ancestor_names.index(sample), 'name')
individual.setInfo(ind_id, 'ind_id')
for ind, chr in enumerate(chromosome_names):
individual.setGenotype(ancestral_genomes[sample][chr],
chroms=[ind])
#Alternating rounds of recombination with clonal expansion. Clonal expansion gives + 2.
#Mutation prior to each round
simulator = sp.Simulator(population)
rate_matrix = createRateMatrix(len(ancestor_names),
0.0002) #10,000 times the mutation rate.
id_tagger = sp.IdTagger()
ped_tagger = sp.PedigreeTagger(output='>>' + pedigree_path,
outputFields=['name', 'ind_id'])
inherit_tagger = sp.InheritTagger(infoFields='name')
initOps1 = [sp.PyExec('print("Starting random selection")'), ped_tagger]
initOps2 = [sp.PyExec('print("Starting random mating")'), ped_tagger]
preOps1 = [sp.MatrixMutator(rate=rate_matrix)]
preOps2 = [sp.InitSex(sex=[sp.MALE, sp.FEMALE])]
matingScheme1 = sp.RandomSelection(
ops=[sp.CloneGenoTransmitter(), inherit_tagger, id_tagger, ped_tagger],
subPopSize=expansion_pop_size)
matingScheme2 = sp.RandomMating(
ops=[
sp.Recombinator(intensity=0.01 / 105000,
convMode=(sp.GEOMETRIC_DISTRIBUTION, 0.001,
0.01)), #10x normal
sp.PyTagger(func=addNames),
id_tagger,
ped_tagger
],
subPopSize=expansion_pop_size)
postOps = []
finalOps = []
print('Starting Evolution Cycles.')
try:
os.remove(pedigree_path)
except:
pass
simulator.evolve(
initOps=[id_tagger, ped_tagger],
matingScheme=sp.CloneMating(ops=[
sp.CloneGenoTransmitter(), ped_tagger, id_tagger, inherit_tagger
]),
gen=1)
for x in range(gen):
simulator.evolve(initOps=initOps1,
preOps=preOps1,
matingScheme=matingScheme1,
postOps=postOps,
finalOps=finalOps,
gen=1)
simulator.evolve(initOps=initOps2,
preOps=preOps2,
matingScheme=matingScheme2,
postOps=postOps,
finalOps=finalOps,
gen=1)
offsprings = {
''.join([
str(int(x.info('name'))),
generateID(3),
str(int(x.info('ind_id')))
]): x.info('ind_id')
for x in simulator.population(0).individuals()
}
sampled_ones = rand.sample(offsprings.keys(), offsprings_sampled)
#reorganizes the offspring genome. Extract info by chr.
offspring_genomes = {name: {} for name in sampled_ones}
for name in sampled_ones:
for ind, chr in enumerate(chromosome_names):
offspring_genomes[name][chr] = simulator.population(0).indByID(
offsprings[name], idField='ind_id').genotype(ploidy=0,
chroms=[ind])
offspring_genomes.update(ancestral_genomes)
print('Parent Guide:')
for ind, id in enumerate(ancestor_names):
print(" : ".join([str(ind), str(id)]))
print('Complete. Generating Output.')
with open(parents_path, 'w') as parent_output:
parent_output.write('Parent Guide:\n')
for ind, id in enumerate(ancestor_names):
parent_output.write(" : ".join([str(ind), str(id)]) + '\n')
#output
offspring_genomes = snp.restructure((offspring_genomes, loci_positions),
structure_mode)
snp.outputGriggFormat(offspring_genomes, output_path)
print('Simulation Complete.')
sim.InitInfo(lambda: random.randint(0, 3), infoFields='age'),
# random genotype
sim.InitGenotype(freq=[0.5, 0.5]),
# assign an unique ID to everyone.
sim.IdTagger(),
],
# increase the age of everyone by 1 before mating.
preOps=sim.InfoExec('age += 1'),
matingScheme=sim.HeteroMating([
# age 1, 2 will be copied
sim.CloneMating(
ops=[
# This will set offspring ID
sim.CloneGenoTransmitter(),
# new ID for offspring in order to track pedigree
sim.IdTagger(),
# both offspring and parental IDs will be the same
sim.PedigreeTagger(output='>>structured.ped'),
],
subPops=[(0,1), (0,2)],
weight=-1
),
# age 2 produce offspring
sim.RandomMating(
ops=[
# new ID for offspring
sim.IdTagger(),
# record complete pedigree
sim.PedigreeTagger(output='>>structured.ped'),
sim.MendelianGenoTransmitter(), # transmit genotype
],
subPops=[(0,2)]
def MutationSelection(N=1000,
generations=10000,
X_loci=100,
A_loci=0,
AgingModel='two_phases',
seed=2001,
reps=1,
InitMutFreq=0.001,
aging_a1=0.003,
aging_a2=0.05,
aging_b=-0.019,
aging_k=0.1911,
MutRate=0.001,
StatsStep=100,
OutPopPrefix='z1',
PrintFreqs=False,
debug=False):
'''Creates and evolves a population to reach mutation-selection balance.'''
if debug:
sim.turnOnDebug('DBG_ALL')
else:
sim.turnOffDebug('DBG_ALL')
sim.setRNG('mt19937', seed)
pop = sim.Population(N,
loci=[X_loci, A_loci],
ploidy=2,
chromTypes=[sim.CHROMOSOME_X, sim.AUTOSOME],
infoFields=[
'age', 'a', 'b', 'smurf', 'ind_id', 'father_id',
'mother_id', 'luck', 't0', 'fitness'
])
pop.setVirtualSplitter(
sim.CombinedSplitter(
splitters=[
sim.ProductSplitter(splitters=[
sim.InfoSplitter(field='age', cutoff=9),
sim.InfoSplitter(field='smurf', values=[0, 1])
]),
sim.SexSplitter(),
sim.InfoSplitter(field='age', values=0)
],
vspMap=[(0), (2), (1, 3), (4), (5), (6)],
names=['larvae', 'adults', 'smurfs', 'males', 'females', 'zero']))
pop.dvars().k = aging_k
pop.dvars().N = N
pop.dvars().seed = seed
pop.dvars().X_loci = X_loci
pop.dvars().A_loci = A_loci
pop.dvars().AgingModel = AgingModel
exec("import random\nrandom.seed(seed)", pop.vars(), pop.vars())
exec("import math", pop.vars(), pop.vars())
simu = sim.Simulator(pop, rep=reps)
simu.evolve(
initOps=[
sim.InitSex(),
sim.InitGenotype(freq=[1 - InitMutFreq, InitMutFreq]),
sim.InitInfo([0], infoFields='age'),
sim.InitInfo([aging_a1], infoFields='a'),
sim.InitInfo([aging_b], infoFields='b'),
sim.InitInfo(lambda: random.random(), infoFields='luck'),
sim.InfoExec('t0 = -ind.b / ind.a', exposeInd='ind'),
sim.InfoExec(
'smurf = 1.0 if AgingModel == "two_phases" and (ind.smurf == 1 or (ind.age > ind.t0 and ind.luck < 1.0 - math.exp(-ind.a * ind.age + ind.a * ind.t0 - ind.a / 2.0))) else 0.0',
exposeInd='ind'),
sim.IdTagger(),
sim.PyExec('XFreqChange={}'),
sim.PyExec('AFreqChange={}')
],
preOps=[
sim.InfoExec('luck = random.random()'),
sim.InfoExec(
'smurf = 1.0 if AgingModel == "two_phases" and (ind.smurf == 1 or (ind.age > ind.t0 and ind.luck < 1.0 - math.exp(-ind.a * ind.age + ind.a * ind.t0 - ind.a / 2.0))) else 0.0',
exposeInd='ind'),
sim.DiscardIf(natural_death(AgingModel)),
sim.InfoExec('age += 1'),
sim.PySelector(func=fitness_func1)
],
matingScheme=sim.HeteroMating([
sim.CloneMating(subPops=[(0, 0), (0, 1), (0, 2)], weight=-1),
sim.RandomMating(ops=[
sim.IdTagger(),
sim.PedigreeTagger(),
sim.InfoExec('smurf = 0.0'),
sexSpecificRecombinator(
rates=[0.75 / X_loci for x in range(X_loci)] +
[2.07 / A_loci for x in range(A_loci)],
maleRates=0.0),
sim.PyQuanTrait(loci=sim.ALL_AVAIL,
func=TweakAdditiveRecessive(
aging_a1, aging_a2, aging_b, X_loci),
infoFields=['a', 'b'])
],
weight=1,
subPops=[(0, 1)],
numOffspring=1)
],
subPopSize=demo),
postOps=[
sim.SNPMutator(u=MutRate, subPops=[(0, 5)]),
sim.Stat(alleleFreq=sim.ALL_AVAIL, step=StatsStep),
sim.IfElse(
'X_loci > 0',
ifOps=[
sim.PyExec(
'XFreqChange[gen] = [alleleFreq[x][1] for x in range(X_loci)]'
)
],
elseOps=[sim.PyExec('XFreqChange[gen] = []')],
step=StatsStep),
sim.IfElse(
'A_loci > 0',
ifOps=[
sim.PyExec(
'AFreqChange[gen] = [alleleFreq[a][1] for a in range(X_loci, pop.totNumLoci())]',
exposePop='pop')
],
elseOps=[sim.PyExec('AFreqChange[gen] = []')],
step=StatsStep),
sim.IfElse(
PrintFreqs,
ifOps=[
sim.PyEval(
r"str(rep) + '\t' + str(gen) + '\t' + '\t'.join(map('{0:.4f}'.format, XFreqChange[gen])) + '\t\t' + '\t'.join(map('{0:.4f}'.format, AFreqChange[gen])) + '\n'"
)
],
step=StatsStep),
sim.TerminateIf(
'sum([alleleFreq[x][0] * alleleFreq[x][1] for x in range(X_loci + A_loci)]) == 0'
)
],
gen=generations)
i = 0
for pop in simu.populations():
pop.save('{}_{}.pop'.format(OutPopPrefix, i))
i += 1
