def sampleAndExport(pop):
sz = pop.subPopSizes()
new_sz = [x//200 for x in sz]
sample = drawRandomSample(pop, new_sz)
export(sample, format='genepop', output='sim15_time_mono_1of2_a_041400_sample_%d.gen' % pop.dvars().gen, adjust = 0, gui = False)
export(sample, format='fstat', output='sim15_time_mono_1of2_a_041400_sample_%d.dat' % pop.dvars().gen, gui = False)
return True
def sampleAndExport(pop):
sz = pop.subPopSizes()
new_sz = [x//1 for x in sz]
sample = drawRandomSample(pop, new_sz)
export(sample, format='genepop', output='sim16a_mono_1of2_%d_himu_sample_%d.gen' % (name,pop.dvars().gen), adjust = 0, gui = False)
export(sample, format='fstat', output='sim16a_mono_1of2_%s_himu_sample_%d.dat' % (name,pop.dvars().gen), gui = False)
return True
def sampleAndExport(pop):
sz = pop.subPopSizes()
new_sz = [x//2000 for x in sz]
sample = drawRandomSample(pop, new_sz)
export(sample, format = 'fstat', output = 'sim16b_4allele_mono_1of2_realmu_%d_000000_sample_%d.dat' % (iter,pop.dvars().gen), gui=False),
export(sample, format = 'phylip', output = 'sim16b_4allele_mono_1of2_realmu_%d_000000_sample_%d.phy' % (iter,pop.dvars().gen), alleleNames = ('A','C','G','T'), gui=False),
os.system('perl convert_diploid.pl N sim16b_4allele_mono_1of2_realmu_%d_000000_sample_%d.phy sim16b_4allele_mono_1of2_realmu_%d_000000_merged_sample_%d.phy' % (iter,pop.dvars().gen)),
os.system('perl phylip_to_fasta.pl sim16b_4allele_mono_1of2_realmu_%d_000000_sample_%d.phy sim16b_4allele_mono_1of2_realmu_%d_000000_sample_%d.fas' % (iter,pop.dvars().gen)),
return True
file_to_open_4 = os.path.join(PATHF, "Resultats", "savepop0.pop")
shutil.copy(file_to_open_3, file_to_open_4)
for g in range(nb_gen):
file_to_open_5 = os.path.join(PATHF, "Resultats", "savepop%d.pop" % g)
pop = sim.loadPopulation(file_to_open_5)
file_to_open_6 = os.path.join(PATHF, "Resultats",
"myPOP_Generation_" + str(g) + ".tmp")
export(pop,
format='csv',
infoFields=['sum_g1', 'z1', 'fitness'],
output=file_to_open_6,
gui=False)
#====================== Execute the script wich paste the gen column ======================#
file_to_open_7 = os.path.join(PATHF, "Script", "GenCol_Script.sh %s %s")
subprocess.call([file_to_open_7 % (nb_indiv, nb_gen)], shell=True)
#====================== Time ======================#
start_time = time.time()
interval = time.time() - start_time
print '== > Total time in seconds = ', interval
def main():
# Check for arguments passed
try:
opts, args = getopt.getopt(sys.argv[1:], shortopts="vhd:p1:p2:s:n:l:e:f:i:m:g:r:", longopts=["verbose", "help", "distribution=", "parameter1=", "parameter2=", "size=",
"number=", "loci=", "effect=", "mean=", "filename=", "heritability=", "gen=", "rrate="])
except getopt.GetoptError as err:
print(err)
usage()
sys.exit()
verbose = False
filename = "my"
size = 1000
number = 100
heritability = 0.2
mean = 2.0
gen = 5
rrate = 0.0
print "\n"
for o in opts:
if o[0] in ("-v", "--verbose"):
verbose = True
print ("Verbose mode")
for o in opts:
if o[0] in ("-d", "--distribution"):
distribution = float(o[1])
if distribution == 0:
parameter1 = None
parameter2 = None
if verbose:
print "Simulation will occur with Normal Distribution"
elif distribution == 1:
if verbose:
print "Simulation will occur with Gamma Distribution"
for o in opts:
if o[0] in ("-p1", "--parameter1"):
parameter1 = float(o[1])
if verbose:
print "Gamma distribution will occur with alpha:", parameter1
for o in opts:
if o[0] in ("-p2", "--parameter2"):
parameter2 = float(o[1])
if verbose:
print "Gamma distribution will occur with beta:", parameter2
elif distribution != 0 or distribution != 1:
sys.exit("Error message: Distribution option must be either 0 or 1")
for o in opts:
if o[0] in ("-p2", "--parameter2"):
bbeta = float(o[1])
if verbose:
print "Gamma distribution will occur with beta:", bbeta
for o in opts:
if o[0] in ("-s", "--size"):
individuals = o[1].split(",")
individuals = map(int, individuals)
if verbose:
print "Population size/s is set at", individuals
for o in opts:
if o[0] in ("-h", "--help"):
usage()
sys.exit()
elif o[0] in ("-n", "--number"):
number = o[1]
if verbose:
print "Number of loci per individual is set at", number
elif o[0] in ("-l", "--loci"):
global loci
loci = o[1].split(",")
loci = map(int, loci)
if verbose:
print "Loci positions per individual are:", loci
elif o[0] in ("-e", "--effect"):
global effects
effects = o[1].split(",")
effects = map(float, effects)
if verbose:
print "Effects for loci per individual are:", effects
elif o[0] in ("-f", "--filename"):
filename = o[1]
if verbose:
print "File will be saved as:", filename
elif o[0] in ("-i", "--heritability"):
heritability = float(o[1])
if verbose:
print "Heritability for simulation specified as:", heritability
elif o[0] in ("-m", "--mean"):
mean = o[1].split(",")
mean = map(float, mean)
if len(mean) == 1 and len(individuals) > 1:
mean = numpy.array(mean)
mean = numpy.repeat(mean, len(individuals), axis=0)
mean = list(mean)
if verbose:
print "Population mean/s specified as:", mean
elif o[0] in ("-g", "--gen"):
gen = int(o[1])
if verbose:
print "Generations to evolve specified as:", gen
elif o[0] in ("-r", "--rrate"):
rrate = float(o[1])
if verbose:
print "Recombination will occur with rate:", rrate
## Start quantitative trait simulation
if verbose:
print "Creating population..."
pop = sim.Population(size=individuals, loci=int(number), infoFields=["qtrait"])
if verbose:
print "Evolving population..."
pop.evolve(initOps=[sim.InitSex(), sim.InitGenotype(prop=[0.7, 0.3])],
matingScheme=sim.RandomMating(ops=sim.Recombinator(rates=rrate)),
postOps=[sim.PyQuanTrait(loci=loci, func=additive_model, infoFields=["qtrait"])],
gen=gen)
if verbose:
print "Coalescent process complete. Population evolved with", pop.numSubPop(),"sub-populations."
genotypes = list()
for i in pop.individuals():
genotypes.append(i.genotype())
phenotypes = list()
for i in pop.individuals():
phenotypes.append(i.qtrait)
# fun() obtains the heritability equation set to zero for various settings of sigma (standard deviation)
#NOTE: May need to tweak gamma distribution parameters to be appropriate for data!
def fun(sigma, h):
x_exact = list()
count = 0
for i in phenotypes:
current_mean = mean[pop.subPopIndPair(count)[0]]
x_exact.append(current_mean + i)
count += 1
x_random = list()
#bbeta=((sigma**2)/current_mean) #Set up approximate beta variable for gamma distribution
count = 0
for each in phenotypes:
current_mean = mean[pop.subPopIndPair(count)[0]]
x_random.append(random.normalvariate(current_mean + each, sigma))
count += 1
r = pearsonr(x_exact, x_random)[0]
return r - math.sqrt(h)
if verbose:
print "Building polynomial model for variance tuning..."
# Fits a polynomial model in numpy to the values obtained from the fun() function
points = list()
for i in drange(0, max(effects)*10, 0.001):
points.append(i)
y_points = list()
for i in points:
y_points.append(fun(i, heritability))
z = numpy.polyfit(x=points, y=y_points, deg=3)
p = numpy.poly1d(z)
# Netwon's method finds the polynomial model's roots
def newton(p):
xn = 100
p_d = p.deriv()
count = 0
while abs(p(xn)) > 0.01:
if count > 1000:
print "Unable to converge after 1000 iterations...\nPlease choose different settings."
usage()
sys.exit()
count += 1
xn = xn - p(xn)/p_d(xn)
if xn < 0.0:
xn = 0.0
if verbose:
print "Estimated variance of phenotypes for specified heriability: ", xn
return xn
if verbose:
print "Using Newton's method to find polynomial roots..."
# Files are saved to the specified location
estimated_variance = newton(p)
new_phenotypes = list()
count = 0
for o in opts:
if o[0] in ("-d", "--distribution"):
if distribution == 0:
for each in phenotypes:
current_mean = mean[pop.subPopIndPair(count)[0]]
new_phenotypes.append(random.normalvariate(current_mean + each, estimated_variance))
count += 1
elif distribution == 1:
for each in phenotypes:
current_mean = mean[pop.subPopIndPair(count)[0]]
new_phenotypes.append(random.gammavariate((current_mean + each)/parameter2, ((estimated_variance/parameter1)**0.5)))
count += 1
f = open(filename + "_qtrait.txt", "w")
f.write("\n".join(map(lambda x: str(x), new_phenotypes)))
f.close()
numpy.savetxt(filename + "_kt_ote2.txt", numpy.column_stack((loci, numpy.array(effects))))
export(pop, format='ped', output=filename+'_genomes.ped')
export(pop, format='map', output=filename+'_genomes.map')
print "\n\n"
format_data(filename, new_phenotypes)
print "\n\n"
def main():
# Check for arguments passed
try:
opts, args = getopt.getopt(sys.argv[1:],
shortopts="vhd:p1:p2:s:n:l:e:f:i:m:g:r:",
longopts=[
"verbose", "help", "distribution=",
"parameter1=", "parameter2=", "size=",
"number=", "loci=", "effect=", "mean=",
"filename=", "heritability=", "gen=",
"rrate="
])
except getopt.GetoptError as err:
print(err)
usage()
sys.exit()
verbose = False
filename = "my"
size = 1000
number = 100
heritability = 0.2
mean = 2.0
gen = 5
rrate = 0.0
print "\n"
for o in opts:
if o[0] in ("-v", "--verbose"):
verbose = True
print("Verbose mode")
for o in opts:
if o[0] in ("-d", "--distribution"):
distribution = float(o[1])
if distribution == 0:
parameter1 = None
parameter2 = None
if verbose:
print "Simulation will occur with Normal Distribution"
elif distribution == 1:
if verbose:
print "Simulation will occur with Gamma Distribution"
for o in opts:
if o[0] in ("-p1", "--parameter1"):
parameter1 = float(o[1])
if verbose:
print "Gamma distribution will occur with alpha:", parameter1
for o in opts:
if o[0] in ("-p2", "--parameter2"):
parameter2 = float(o[1])
if verbose:
print "Gamma distribution will occur with beta:", parameter2
elif distribution != 0 or distribution != 1:
sys.exit(
"Error message: Distribution option must be either 0 or 1")
for o in opts:
if o[0] in ("-p2", "--parameter2"):
bbeta = float(o[1])
if verbose:
print "Gamma distribution will occur with beta:", bbeta
for o in opts:
if o[0] in ("-s", "--size"):
individuals = o[1].split(",")
individuals = map(int, individuals)
if verbose:
print "Population size/s is set at", individuals
for o in opts:
if o[0] in ("-h", "--help"):
usage()
sys.exit()
elif o[0] in ("-n", "--number"):
number = o[1]
if verbose:
print "Number of loci per individual is set at", number
elif o[0] in ("-l", "--loci"):
global loci
loci = o[1].split(",")
loci = map(int, loci)
if verbose:
print "Loci positions per individual are:", loci
elif o[0] in ("-e", "--effect"):
global effects
effects = o[1].split(",")
effects = map(float, effects)
if verbose:
print "Effects for loci per individual are:", effects
elif o[0] in ("-f", "--filename"):
filename = o[1]
if verbose:
print "File will be saved as:", filename
elif o[0] in ("-i", "--heritability"):
heritability = float(o[1])
if verbose:
print "Heritability for simulation specified as:", heritability
elif o[0] in ("-m", "--mean"):
mean = o[1].split(",")
mean = map(float, mean)
if len(mean) == 1 and len(individuals) > 1:
mean = numpy.array(mean)
mean = numpy.repeat(mean, len(individuals), axis=0)
mean = list(mean)
if verbose:
print "Population mean/s specified as:", mean
elif o[0] in ("-g", "--gen"):
gen = int(o[1])
if verbose:
print "Generations to evolve specified as:", gen
elif o[0] in ("-r", "--rrate"):
rrate = float(o[1])
if verbose:
print "Recombination will occur with rate:", rrate
## Start quantitative trait simulation
if verbose:
print "Creating population..."
pop = sim.Population(size=individuals,
loci=int(number),
infoFields=["qtrait"])
if verbose:
print "Evolving population..."
pop.evolve(
initOps=[sim.InitSex(),
sim.InitGenotype(prop=[0.7, 0.3])],
matingScheme=sim.RandomMating(ops=sim.Recombinator(rates=rrate)),
postOps=[
sim.PyQuanTrait(loci=loci,
func=additive_model,
infoFields=["qtrait"])
],
gen=gen)
if verbose:
print "Coalescent process complete. Population evolved with", pop.numSubPop(
), "sub-populations."
genotypes = list()
for i in pop.individuals():
genotypes.append(i.genotype())
phenotypes = list()
for i in pop.individuals():
phenotypes.append(i.qtrait)
# fun() obtains the heritability equation set to zero for various settings of sigma (standard deviation)
#NOTE: May need to tweak gamma distribution parameters to be appropriate for data!
def fun(sigma, h):
x_exact = list()
count = 0
for i in phenotypes:
current_mean = mean[pop.subPopIndPair(count)[0]]
x_exact.append(current_mean + i)
count += 1
x_random = list()
#bbeta=((sigma**2)/current_mean) #Set up approximate beta variable for gamma distribution
count = 0
for each in phenotypes:
current_mean = mean[pop.subPopIndPair(count)[0]]
x_random.append(random.normalvariate(current_mean + each, sigma))
count += 1
r = pearsonr(x_exact, x_random)[0]
return r - math.sqrt(h)
if verbose:
print "Building polynomial model for variance tuning..."
# Fits a polynomial model in numpy to the values obtained from the fun() function
points = list()
for i in drange(0, max(effects) * 10, 0.001):
points.append(i)
y_points = list()
for i in points:
y_points.append(fun(i, heritability))
z = numpy.polyfit(x=points, y=y_points, deg=3)
p = numpy.poly1d(z)
# Netwon's method finds the polynomial model's roots
def newton(p):
xn = 100
p_d = p.deriv()
count = 0
while abs(p(xn)) > 0.01:
if count > 1000:
print "Unable to converge after 1000 iterations...\nPlease choose different settings."
usage()
sys.exit()
count += 1
xn = xn - p(xn) / p_d(xn)
if xn < 0.0:
xn = 0.0
if verbose:
print "Estimated variance of phenotypes for specified heriability: ", xn
return xn
if verbose:
print "Using Newton's method to find polynomial roots..."
# Files are saved to the specified location
estimated_variance = newton(p)
new_phenotypes = list()
count = 0
for o in opts:
if o[0] in ("-d", "--distribution"):
if distribution == 0:
for each in phenotypes:
current_mean = mean[pop.subPopIndPair(count)[0]]
new_phenotypes.append(
random.normalvariate(current_mean + each,
estimated_variance))
count += 1
elif distribution == 1:
for each in phenotypes:
current_mean = mean[pop.subPopIndPair(count)[0]]
new_phenotypes.append(
random.gammavariate(
(current_mean + each) / parameter2,
((estimated_variance / parameter1)**0.5)))
count += 1
f = open(filename + "_qtrait.txt", "w")
f.write("\n".join(map(lambda x: str(x), new_phenotypes)))
f.close()
numpy.savetxt(filename + "_kt_ote2.txt",
numpy.column_stack((loci, numpy.array(effects))))
export(pop, format='ped', output=filename + '_genomes.ped')
export(pop, format='map', output=filename + '_genomes.map')
print "\n\n"
def sampleAndExport(pop):
sz = pop.subPopSizes()
new_sz = [x//200 for x in sz]
sample = drawRandomSample(pop, new_sz)
export(sample, format='fstat', output='sim10n_mono_a_000000_sample_%d.dat' % pop.dvars().gen, gui = False)
return True
],
matingScheme=sim.RandomMating(
ops=[sim.Recombinator(intensity=args.recomb_rate)]),
postOps=[
sim.Stat(numOfSegSites=sim.ALL_AVAIL, step=50),
sim.PyEval(r"'Gen: %2d #seg sites: %d\n' % (gen, numOfSegSites)",
step=50)
],
gen=args.generations)
logfile.write("Done simulating!\n")
logfile.write(time.strftime('%X %x %Z') + "\n")
logfile.write("----------\n")
logfile.flush()
sample = drawRandomSample(pop, sizes=args.nsamples)
if args.outfile is None:
print("NOT writing out genotype data.\n")
else:
print("Writing out genotype data to " + args.outfile + "\n")
export(sample, format='PED', output=args.outfile)
logfile.write("Done writing out!\n")
logfile.write(time.strftime('%X %x %Z') + "\n")
logfile.write("----------\n")
logfile.flush()
logfile.write("All done!\n")
logfile.close()
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
#
# This script is an example in the simuPOP user's guide. Please refer to
# the user's guide (http://simupop.sourceforge.net/manual) for a detailed
# description of this example.
#
import simuPOP as sim
from simuPOP.utils import importPopulation, export
pop = sim.Population([2, 4],
loci=5,
lociNames=['a1', 'a2', 'a3', 'a4', 'a5'],
infoFields='BMI')
sim.initGenotype(pop, freq=[0.3, 0.5, 0.2])
sim.initSex(pop)
sim.initInfo(pop, [20, 30, 40, 50, 30, 25], infoFields='BMI')
export(pop, format='fstat', output='fstat.txt')
print(open('fstat.txt').read())
export(pop, format='structure', phenotype='BMI', output='stru.txt')
print(open('stru.txt').read())
pop1 = importPopulation(format='fstat', filename='fstat.txt')
sim.dump(pop1)