def gen_manhattan(path_list, threshold=None, titles=None):
"""Generate separated manhattan plots and one combined QQ plot.
Input: - path_list is a Python list containing paths to the plink .glm.logistic or .glm.linear files.
- threshold: draw a threshold line, 0.0 < threshold < 1.0
- titles: the title are only recognized for amino acid with filenames in the 'xxx.glm.logistic.ADD' format"""
################# COLORS
chrs = [str(i) for i in range(1, 23)]
chrs_names = np.array([str(i) for i in range(1, 23)])
chrs_names[1::2] = ''
colors = ['#1b9e77', "#d95f02", '#7570b3', '#e7298a']
# Converting from HEX into RGB
colors = [hex2color(colors[i]) for i in range(len(colors))]
################# LOAD DAT
# TODO: manage better paths and amino acid name
aa_names = [file.split('.')[-4] for file in path_list]
if titles != None:
aa_names = titles
list_p_series = []
for i, file in enumerate(path_list):
df = pd.read_csv(file, sep='\t')
list_p_series.append(df.P)
################# PLOT
manhattan.manhattan(
df['P'],
df['POS'],
df['#CHROM'].astype(str),
'',
plot_type='single',
chrs_plot=[str(i) for i in range(1, 23)],
chrs_names=chrs_names,
cut=0,
title=aa_names[i],
xlabel='chromosome',
ylabel='-log10(p-value)',
lines=[],
#lines_colors=['r'],
colors=colors)
if threshold != None:
minlogthreshold = -np.log10(threshold)
plt.plot([0, 5000000000], [minlogthreshold, minlogthreshold])
plt.show()
## QQ PLOTS
N = len(aa_names)
fill_dens = [0.2] * N
# ['#1b9e77', "#d95f02", '#7570b3']
colors = [
'#c0392b', '#884ea0', '#2471a3', '#17a589', '#229954', '#d4ac0d',
'#ca6f1e', '#839192', '#17202a'
]
qqplot(list_p_series,
aa_names,
color=colors[0:N],
fill_dens=fill_dens,
error_type='theoretical',
distribution='beta',
title='QQ plots')
plt.gca().legend(bbox_to_anchor=(1.1, 0.1 * N), loc=2, borderaxespad=0.)
plt.grid(linestyle='--')
import matplotlib.pyplot as plt
import numpy as np
from assocplots.qqplot import *
# data = np.genfromtxt('C:\\Users\\Ekaterina\\Desktop\\file_transfer\\fem_assoc_dos_ocd_occc_eur_sr-qc.hg19.dosage.assoc.dosage.QCed')
# data = data[:,-1]
# data2 = np.genfromtxt('C:\\Users\\Ekaterina\\Desktop\\file_transfer\\mal_assoc_dos_ocd_occc_eur_sr-qc.hg19.dosage.assoc.dosage.QCed')
# data2 = data2[:,-1]
q_data, q_th, q_err = qqplot([data, data2], labels=['female', 'male'], error_type='experimental', n_quantiles=1000, alpha=0.95, color=['r', 'b'])
data_male = np.genfromtxt('sample_data\GIANT_Randall2013PlosGenet_stage1_publicrelease_HapMapCeuFreq_BMI_MEN_N.txt', dtype=None, names=True)
data_female = np.genfromtxt('sample_data\GIANT_Randall2013PlosGenet_stage1_publicrelease_HapMapCeuFreq_BMI_WOMEN_N.txt', dtype=None, names=True)
q_data, q_th, q_err = qqplot([data_female['P2gc'], data_male['P2gc']], labels=['female', 'male'], error_type='experimental', n_quantiles=1000, alpha=0.95, color=['r', 'b'])
data = mock_data_generation(1, 100000)
data[0]['chr']=1
manhattan(data[0]['pval'], data[0]['pos'], data[0]['chr'], 'abc',
p2=None, pos2=None, chr2=None, label2=None,
type='single',
import matplotlib.pyplot as plt
import numpy as np
from assocplots.qqplot import *
# data = np.genfromtxt('C:\\Users\\Ekaterina\\Desktop\\file_transfer\\fem_assoc_dos_ocd_occc_eur_sr-qc.hg19.dosage.assoc.dosage.QCed')
# data = data[:,-1]
# data2 = np.genfromtxt('C:\\Users\\Ekaterina\\Desktop\\file_transfer\\mal_assoc_dos_ocd_occc_eur_sr-qc.hg19.dosage.assoc.dosage.QCed')
# data2 = data2[:,-1]
q_data, q_th, q_err = qqplot([data, data2],
labels=['female', 'male'],
error_type='experimental',
n_quantiles=1000,
alpha=0.95,
color=['r', 'b'])
data_male = np.genfromtxt(
'sample_data\GIANT_Randall2013PlosGenet_stage1_publicrelease_HapMapCeuFreq_BMI_MEN_N.txt',
dtype=None,
names=True)
data_female = np.genfromtxt(
'sample_data\GIANT_Randall2013PlosGenet_stage1_publicrelease_HapMapCeuFreq_BMI_WOMEN_N.txt',
dtype=None,
names=True)
q_data, q_th, q_err = qqplot([data_female['P2gc'], data_male['P2gc']],
labels=['female', 'male'],
error_type='experimental',
n_quantiles=1000,
alpha=0.95,
data = pd.read_fwf(datafile)
manhattan(data['P'], data["BP"], data["CHR"].apply(str), '', \
type='single', \
chrs_plot=[str(i) for i in range(1,23)], \
chrs_names=chrs_names, \
cut = 0, \
title='Eye color', \
xlabel='chromosome', \
ylabel='-log10(p-value)', \
lines= [], \
colors = colors, \
scaling = '-log10')
plt.savefig('%s_manhattan.png' % prefix, dpi=300)
############# QQ plot #############
from assocplots.qqplot import *
mpl.rcParams['figure.dpi'] = 100
mpl.rcParams['savefig.dpi'] = 100
mpl.rcParams['figure.figsize'] = 5.375, 5.375
qqplot([data["P"]], ['eyecolor'],
color=['b'],
fill_dens=[0.2],
error_type='theoretical',
distribution='beta',
title='')
plt.savefig('%s_qq.png' % prefix, dpi=300)
from assocplots.manhattan import *
datf = np.genfromtxt('OCD_ADHD_female.META.CHR.POS', dtype=None, skip_header=1)
datm = np.genfromtxt('OCD_ADHD_male.META.CHR.POS', dtype=None, skip_header=1)
##### QQ plot
import matplotlib as mpl
mpl.rcParams['figure.dpi'] = 150
mpl.rcParams['savefig.dpi'] = 150
mpl.rcParams['figure.figsize'] = 6.375, 6.375
plt.clf()
qqplot([datf['f2'], datm['f2']],
['Female $\lambda=1.104$', 'Male $\lambda=1.150$'],
color=['r', 'b'],
fill_dens=[0.2, 0.2],
error_type='theoretical',
distribution='beta',
title='')
plt.ylim([0, 8])
#plt.show()
plt.savefig('OCD_ADHD_mal_fem_QQ.png', dpi=300)
get_lambda(datf['f2'], definition='median')
#1.1038305474906456
get_lambda(datm['f2'], definition='median')
#1.1502208100009044
##### Manhattan Plot
import matplotlib as mpl
[1.2755015584185816,
1.3253828247438937,
1.5075929863128841,
1.4956062686095888,
1.2630205643665071,
1.3120597064714479,
1.4892568965957846,
1.4884128058714166]
mpl.rcParams['figure.dpi']=300
mpl.rcParams['savefig.dpi']=300
mpl.rcParams['figure.figsize']=5.375, 5.375
qqplot([Baseline_Log2_NoReg['f3'], Baseline_Log2_RegOut['f3']],
['-RegOut(1.27)', '+RegOut(1.32)'],
color=['b','r'],
fill_dens=[0.2,0.2],
error_type='experimental',
distribution='beta',
title='QQ Plot of Baseline_Log2 eQTLs')
plt.savefig('QQ2_Baseline_Log2.png', dpi=300)
mpl.rcParams['figure.dpi']=300
mpl.rcParams['savefig.dpi']=300
mpl.rcParams['figure.figsize']=5.375, 5.375
qqplot([Baseline_PEER10_NoReg['f3'], Baseline_PEER10_RegOut['f3']],
['-RegOut(1.51)', '+RegOut(1.49)'],
color=['b','r'],
fill_dens=[0.2,0.2],
error_type='experimental',
distribution='beta',
title='QQ Plot of Baseline_PEER10 eQTLs')