def QQPlot(arguments,pv,unique_pv,fname):
font_size = 18
mpl.rcParams['font.family']="sans-serif"
mpl.rcParams['font.sans-serif']="Arial"
mpl.rcParams['font.size']=font_size
#mpl.rcParams['figure.dpi'] = 300
mpl.rcParams['font.weight']='medium'
mpl.rcParams['figure.facecolor'] = 'white'
mpl.rcParams['lines.linewidth'] = 1
mpl.rcParams['axes.facecolor'] = 'white'
mpl.rcParams['patch.edgecolor'] = 'white'
mpl.rcParams['grid.linestyle'] = '-'
mpl.rcParams['grid.color'] = 'LightGray'
if arguments.ignore!=None:
if arguments.ignore in fname:
return
if arguments.distinct:
pv = unique_pv
pl.figure(figsize=(5,5))
pv_uni = (sp.arange(1.0/float(pv.shape[0]),1,1.0/float(pv.shape[0]+1)))
pl.plot(-sp.log10(pv_uni),-sp.log10(sp.sort(pv_uni)),'b--')
pl.ylim(0,(-sp.log10(pv[:])).max()+1)
pl.plot(-sp.log10(pv_uni),-sp.log10(sp.sort(pv[:],axis=0)),'.',color="#F68E55",markersize=12,markeredgewidth=0,alpha=1)
#plot theoretical expectations
if arguments.estpv:
datapoints=10**(sp.arange(sp.log10(0.5),sp.log10(pv.shape[0]-0.5)+0.1,0.1))
beta_alpha=sp.zeros(datapoints.shape[0])
beta_nalpha=sp.zeros(datapoints.shape[0])
beta_tmp=sp.zeros(datapoints.shape[0])
for n in xrange(datapoints.shape[0]):
m=datapoints[n]
beta_tmp[n]=stats.beta.ppf(0.5,m,pv.shape[0]-m)
beta_alpha[n]=stats.beta.ppf(0.05,m,pv.shape[0]-m)
beta_nalpha[n]=stats.beta.ppf(1-0.05,m,pv.shape[0]-m)
estimated_pvals=datapoints/pv.shape[0]
lower_bound = -sp.log10(estimated_pvals-(beta_tmp-beta_alpha))
upper_bound = -sp.log10(estimated_pvals+(beta_nalpha-beta_tmp))
pl.fill_between(-sp.log10(estimated_pvals),lower_bound,upper_bound,color='#00BFF3',alpha=0.4,linewidth=0)
if arguments.title:
pl.title("Phenotype: %s"%(fname))
pl.xlabel('Expected $-log10(p-value)$')
pl.ylabel('Observed $-log10(p-value)$')
if arguments.gc:
gc = sp.median(stats.chi2.isf(pv,1))/0.456
pl.text(4,1,"$\hat \lambda=%.2f$"%(gc))
remove_border()
pl.subplots_adjust(left=0.14,bottom=0.13,right=0.97,top=0.95,wspace=0.45)
pl.savefig(os.path.join(arguments.out,'qqplot_' + fname + '.' + arguments.iformat) )
pl.close()
def LDPlot(arguments, identifiers, encoded, maf, pv, unique_pv, positions,
chromosomes, fname, pathogenicity_map):
font_size = 16
mpl.rcParams['font.family'] = "sans-serif"
mpl.rcParams['font.sans-serif'] = "Arial"
mpl.rcParams['font.size'] = font_size
#mpl.rcParams['figure.dpi'] = 150
mpl.rcParams['font.weight'] = 'medium'
mpl.rcParams['figure.facecolor'] = 'white'
mpl.rcParams['lines.linewidth'] = 1
mpl.rcParams['axes.facecolor'] = 'white'
mpl.rcParams['patch.edgecolor'] = 'white'
mpl.rcParams['grid.linestyle'] = '-'
mpl.rcParams['grid.color'] = 'LightGray'
if arguments.ignore != None:
if arguments.ignore in fname:
return
if not arguments.sql_gene == None:
sqlite = sqlite3.connect(arguments.sql_gene)
sqlite_cursor = sqlite.cursor()
else:
sqlite = None
if arguments.nr_hypothesis == -1:
if arguments.distinct:
bf_threshold = 0.05 / unique_pv.shape[0]
else:
bf_threshold = 0.05 / pv.shape[0]
else:
bf_threshold = 0.05 / arguments.nr_hypothesis
snp_distance = arguments.distance
r2_measure = arguments.r2_measure
pl.ion()
pl.figure(figsize=(20, 4))
color_list = [
'#F26C4F', '#F68E55', '#7CC576', '#00BFF3', '#605CA8', '#F06EA9',
'#F26C4F', '#F68E55'
]
#select different SNP
if arguments.selected_snp != '-1':
ind = sp.where(identifiers == arguments.selected_snp)[0]
if ind.shape[0] == 0:
print "\nSNP " + arguments.selected_snp + " not found in dataset!"
print "Please select a different SNP identifier!\n"
quit()
else:
chrom_list = sp.array([arguments.selected_snp.split("_")[0]])
__pos = sp.array([int(arguments.selected_snp.split("_")[1])])
else:
chrom_list = sp.unique(chromosomes)
for i, chrom in enumerate(chrom_list):
idx = chromosomes == chrom
_pos = positions[idx]
_chrs = chromosomes[idx]
_pv = pv[idx]
if arguments.selected_snp == '-1':
idx = sp.where(_pv <= bf_threshold)[0]
__pos = _pos[idx]
__chrs = _chrs[idx]
__pv = _pv[idx]
__pv = -sp.log10(__pv)
else:
idx = sp.where(__pos[0] == _pos)[0]
__pv = _pv[idx]
__pv = -sp.log10(__pv)
#plot LD
cnorm = mcol.Normalize(vmin=0.0, vmax=1)
cpick = cm.ScalarMappable(norm=cnorm, cmap=pl.get_cmap("jet"))
cpick.set_array([])
ind = sp.argsort(_pos)
_pos = _pos[ind]
_pv = _pv[ind]
for k, pp in enumerate(__pos):
fig = pl.figure(figsize=(12, 4))
if sqlite == None:
ax1 = pl.subplot2grid((5, 5), (0, 0), colspan=5, rowspan=4)
else:
ax1 = pl.subplot2grid((6, 5), (0, 0), colspan=5, rowspan=4)
ax1.axhline(-sp.log10(bf_threshold),
color='#F68E55',
linestyle='--')
marker = '.'
if str(chrom) + "_" + str(pp) in pathogenicity_map:
if pathogenicity_map[str(chrom) + "_" +
str(pp)].prediction == "DELETERIOUS":
marker = '^'
else:
marker = 'v'
ax1.plot(
pp,
__pv[k],
marker,
color="#b000ff",
alpha=0.9,
markersize=10,
)
if pp - snp_distance > 0 and pp + snp_distance <= _pos.max():
ranges = sp.where((_pos >= pp - snp_distance)
& (_pos <= pp + snp_distance))[0]
elif pp - snp_distance < 0 and pp + snp_distance <= _pos.max():
ranges = sp.where(_pos <= pp + snp_distance)[0]
elif pp - snp_distance > 0 and pp + snp_distance > _pos.max():
ranges = sp.where(_pos >= pp - snp_distance)[0]
pp_idx = sp.where(pp == _pos)[0][0]
vline_map = {}
rr_color_list = []
maf_list = []
xlabels = [_pos[ranges][0], __pos[0], _pos[ranges][-1]]
idx = sp.where(str(chrom) + "_" + str(pp) == identifiers)[0]
for l, sra in enumerate(ranges):
sind = sp.where(
str(chrom) + "_" + str(_pos[sra]) == identifiers)[0]
maf_list.append(maf[sind])
if r2_measure == "excoffier_slatkin":
rr = ld.esem_r(
sp.array(encoded[:, sind].flatten(), dtype="int"),
sp.array(encoded[:, idx].flatten(), dtype="int"))**2
elif r2_measure == "roger_huff":
rr = ld.get_r(
sp.array(encoded[:, sind].flatten(), dtype="int"),
sp.array(encoded[:, idx].flatten(), dtype="int"))**2
elif r2_measure == "pearson_r2":
rr = stats.pearsonr(encoded[:, sind].flatten(),
encoded[:, idx].flatten())[0]**2
rr_color_list.append(cpick.to_rgba(rr))
if _pos[sra] == pp:
ax1.vlines(pp,
0,
__pv[k],
color='#b000ff',
linestyle='--',
alpha=0.8,
linewidth=0.5)
vline_map[pp] = "#b000ff"
continue
marker = '.'
if str(chrom) + "_" + str(_pos[sra]) in pathogenicity_map:
if pathogenicity_map[str(chrom) + "_" + str(
_pos[sra])].prediction == "DELETERIOUS":
marker = '^'
else:
marker = 'v'
if rr >= 0.7:
ax1.plot(_pos[sra],
-sp.log10(_pv[sra]),
marker,
color=cpick.to_rgba(rr),
alpha=0.9,
markeredgecolor='#DDDDDD',
markersize=10) #8
elif rr >= 0.5:
ax1.plot(_pos[sra],
-sp.log10(_pv[sra]),
marker,
color=cpick.to_rgba(rr),
alpha=0.9,
markeredgecolor='#DDDDDD',
markersize=9) #7
elif rr >= 0.3:
ax1.plot(_pos[sra],
-sp.log10(_pv[sra]),
marker,
color=cpick.to_rgba(rr),
alpha=0.9,
markeredgecolor='#DDDDDD',
markersize=8) #6
else:
ax1.plot(_pos[sra],
-sp.log10(_pv[sra]),
marker,
color=cpick.to_rgba(rr),
alpha=0.4,
markeredgecolor='#DDDDDD',
markersize=7) #4
if _pv[sra] <= bf_threshold:
ax1.vlines(_pos[sra],
0,
-sp.log10(_pv[sra]),
color=cpick.to_rgba(rr),
linestyle='--',
alpha=0.8,
linewidth=0.5)
vline_map[_pos[sra]] = cpick.to_rgba(rr)
remove_border()
ax1.set_ylabel("$-log10(p-value)$")
ax1.set_ylim(
0,
sp.maximum(
-sp.log10(bf_threshold) + 1,
sp.maximum((-sp.log10(_pv[ranges])).max() + 1,
-sp.log10(_pv[pp_idx]) + 1)))
if arguments.title:
ax1.set_title("Phenotype: %s, SNP: %d" % (fname, pp))
ax1.set_xticks([])
ax1.set_yticks(
sp.arange(
1,
sp.ceil(
sp.maximum(
-sp.log10(bf_threshold) + 1,
sp.maximum((-sp.log10(_pv[ranges])).max() + 1,
-sp.log10(_pv[pp_idx]) + 1))), 3))
ax1.set_xlim(_pos[ranges[0]] - snp_distance * 0.01,
_pos[ranges[-1]] + snp_distance * 0.01)
ax1.yaxis.grid()
if len(pathogenicity_map) > 0:
deleterious = pl.Line2D(range(1),
range(1),
marker='^',
color="white",
linestyle="None")
benign = pl.Line2D(range(1),
range(1),
marker='v',
color="white",
linestyle="None")
leg = pl.legend([benign, deleterious], [
'Benign Missense Mutation', 'Deleterious Missense Mutation'
],
frameon=True,
scatterpoints=1,
prop={'size': 12},
fancybox=True,
bbox_to_anchor=(1, 1.2),
numpoints=1,
ncol=2)
leg.get_frame().set_alpha(0.5)
leg.get_frame().set_linewidth(0.2)
leg.get_frame().set_facecolor("#DDDDDD")
if sqlite == None:
ax3 = pl.subplot2grid((5, 5), (4, 0), colspan=5)
ax3.set_xlabel("Genomic positions on chromosome: " +
str(chrom))
else:
ax3 = pl.subplot2grid((6, 5), (4, 0), colspan=5)
ax3.set_ylim(0, 0.6)
ax3.set_yticks([0.25, 0.5])
ax3.yaxis.grid()
ax3.set_ylabel("MAF")
ax3.set_xticks([])
remove_border(top=True)
ax3.set_xlim(_pos[ranges[0]] - snp_distance * 0.01,
_pos[ranges[-1]] + snp_distance * 0.01)
for key in vline_map:
ax3.axvline(key,
color=vline_map[key],
linestyle='--',
alpha=0.8,
linewidth=0.5)
for i in xrange(ranges.shape[0]):
ax3.plot(_pos[ranges[i]],
maf_list[i],
'x',
color=rr_color_list[i])
if sqlite != None:
ax2 = pl.subplot2grid((6, 5), (5, 0), colspan=5)
remove_border(top=True)
ax2.set_xlabel("Genomic positions on chromosome: " +
str(chrom))
ax2.set_yticks([])
ax2.set_xlim(_pos[ranges[0]] - snp_distance * 0.01,
_pos[ranges[-1]] + snp_distance * 0.01)
for key in vline_map:
ax2.axvline(key,
color=vline_map[key],
linestyle='--',
alpha=0.8,
linewidth=0.5)
sqlite_cursor.execute(
"SELECT * FROM geneannotation WHERE chromosome_id=? AND annotation_end >=? AND annotation_start<=?",
(str(chrom), int(_pos[ranges[0]]), int(_pos[ranges[-1]])))
annotations = sqlite_cursor.fetchall()
for g, annotation in enumerate(annotations):
alpha = 0.6
start = int(annotation[3])
length = int(annotation[4]) - int(annotation[3])
head_width = 0.25
head_length = length * 0.15
width = 0.1
name = annotation[7].replace("Contig20-", "")
if annotation[5] == "+":
shape = "right"
arrow_params = {
'length_includes_head': True,
'shape': shape,
'head_starts_at_zero': True
}
if g % 2 == 0:
y_pos = 0.6
else:
y_pos = 0.5
y_text = y_pos + 2 * width
ax2.arrow(start,
y_pos,
length,
0,
head_width=head_width,
head_length=head_length,
fc=color_list[0],
ec=color_list[0],
alpha=alpha,
width=width,
**arrow_params)
else:
shape = "left"
arrow_params = {
'length_includes_head': True,
'shape': shape,
'head_starts_at_zero': True
}
if g % 2 == 0:
y_pos = 0.15
else:
y_pos = 0.35
y_text = y_pos - 2 * width
ax2.arrow(start + length,
y_pos,
-length,
0,
head_width=head_width,
head_length=head_length,
fc=color_list[3],
ec=color_list[3],
alpha=alpha,
width=width,
**arrow_params)
ax2.text(start + length / 2.0,
y_text,
name,
size=7,
ha='center',
va='center',
color="k")
pl.xticks(xlabels, xlabels)
#pl.gca().get_xaxis().get_major_formatter().set_useOffset(False)
if not sqlite == None:
cax = fig.add_axes([0.93, 0.4, 0.01, 0.5])
pl.colorbar(cpick, label="SNP r^2", cax=cax)
pl.subplots_adjust(left=0.07,
bottom=0.15,
right=0.92,
top=0.9,
wspace=0,
hspace=0)
else:
pl.subplots_adjust(left=0.07,
bottom=0.15,
right=0.99,
top=0.9,
wspace=0,
hspace=0)
if not arguments.title:
pl.subplots_adjust(top=0.9)
pl.savefig(
os.path.join(
arguments.out, 'ld_plot_chrom_' + str(chrom) + "_pos_" +
str(pp) + "_" + fname + '.' + arguments.iformat))
pl.close()
def QQPlot(arguments, pv, unique_pv, fname):
font_size = 18
mpl.rcParams['font.family'] = "sans-serif"
mpl.rcParams['font.sans-serif'] = "Arial"
mpl.rcParams['font.size'] = font_size
#mpl.rcParams['figure.dpi'] = 300
mpl.rcParams['font.weight'] = 'medium'
mpl.rcParams['figure.facecolor'] = 'white'
mpl.rcParams['lines.linewidth'] = 1
mpl.rcParams['axes.facecolor'] = 'white'
mpl.rcParams['patch.edgecolor'] = 'white'
mpl.rcParams['grid.linestyle'] = '-'
mpl.rcParams['grid.color'] = 'LightGray'
if arguments.ignore != None:
if arguments.ignore in fname:
return
if arguments.distinct:
pv = unique_pv
pl.figure(figsize=(5, 5))
pv_uni = (sp.arange(1.0 / float(pv.shape[0]), 1,
1.0 / float(pv.shape[0] + 1)))
pl.plot(-sp.log10(pv_uni), -sp.log10(sp.sort(pv_uni)), 'b--')
pl.ylim(0, (-sp.log10(pv[:])).max() + 1)
pl.plot(-sp.log10(pv_uni),
-sp.log10(sp.sort(pv[:], axis=0)),
'.',
color="#F68E55",
markersize=12,
markeredgewidth=0,
alpha=1)
#plot theoretical expectations
if arguments.estpv:
datapoints = 10**(sp.arange(sp.log10(0.5),
sp.log10(pv.shape[0] - 0.5) + 0.1, 0.1))
beta_alpha = sp.zeros(datapoints.shape[0])
beta_nalpha = sp.zeros(datapoints.shape[0])
beta_tmp = sp.zeros(datapoints.shape[0])
for n in xrange(datapoints.shape[0]):
m = datapoints[n]
beta_tmp[n] = stats.beta.ppf(0.5, m, pv.shape[0] - m)
beta_alpha[n] = stats.beta.ppf(0.05, m, pv.shape[0] - m)
beta_nalpha[n] = stats.beta.ppf(1 - 0.05, m, pv.shape[0] - m)
estimated_pvals = datapoints / pv.shape[0]
lower_bound = -sp.log10(estimated_pvals - (beta_tmp - beta_alpha))
upper_bound = -sp.log10(estimated_pvals + (beta_nalpha - beta_tmp))
pl.fill_between(-sp.log10(estimated_pvals),
lower_bound,
upper_bound,
color='#00BFF3',
alpha=0.4,
linewidth=0)
if arguments.title:
pl.title("Phenotype: %s" % (fname))
pl.xlabel('Expected $-log10(p-value)$')
pl.ylabel('Observed $-log10(p-value)$')
if arguments.gc:
gc = sp.median(stats.chi2.isf(pv, 1)) / 0.456
pl.text(4, 1, "$\hat \lambda=%.2f$" % (gc))
remove_border()
pl.subplots_adjust(left=0.14,
bottom=0.13,
right=0.97,
top=0.95,
wspace=0.45)
pl.savefig(
os.path.join(arguments.out,
'qqplot_' + fname + '.' + arguments.iformat))
pl.close()
def ManhattanPlot(arguments, pv, positions, chromosomes, hashs, unique_pv,
fname):
font_size = 14
mpl.rcParams['font.family'] = "sans-serif"
mpl.rcParams['font.sans-serif'] = "Arial"
mpl.rcParams['font.size'] = font_size
#mpl.rcParams['figure.dpi'] = 300
mpl.rcParams['font.weight'] = 'medium'
mpl.rcParams['figure.facecolor'] = 'white'
mpl.rcParams['lines.linewidth'] = 1
mpl.rcParams['axes.facecolor'] = 'white'
mpl.rcParams['patch.edgecolor'] = 'white'
mpl.rcParams['grid.linestyle'] = '-'
mpl.rcParams['grid.color'] = 'LightGray'
if arguments.ignore != None:
if arguments.ignore in fname:
return
chrom_list = sp.unique(chromosomes)
unsnps = arguments.distinct
pl.ion()
pl.figure(figsize=(20, 3))
if arguments.nr_hypothesis == -1:
if unsnps:
bf_threshold = 0.05 / unique_pv.shape[0]
else:
bf_threshold = 0.05 / pv.shape[0]
else:
bf_threshold = 0.05 / arguments.nr_hypothesis
current_pos = 0
max_y = (-sp.log10(pv[:])).max()
if max_y < -sp.log10(bf_threshold):
max_y = -sp.log10(bf_threshold)
max_y += 1
split_list = []
xtick_list = []
pl.axhline(-sp.log10(bf_threshold), color='#F68E55', linestyle='--')
if not arguments.nr_hypothesis2 == -1:
pl.axhline(-sp.log10(0.05 / arguments.nr_hypothesis2),
color='#00BFF3',
linestyle='--')
factor = sp.around(
float(arguments.nr_hypothesis2) / float(arguments.nr_hypothesis))
leg = pl.legend(['Bonferroni', 'Bonferroni ' + str(factor) + 'x'],
ncol=2,
fancybox=True,
prop={'size': 12},
bbox_to_anchor=(1, 1.05))
leg.get_frame().set_alpha(0.8)
leg.get_frame().set_linewidth(0.2)
for i, chrom in enumerate(chrom_list):
idx = chromosomes == chrom
_pos = positions[idx]
_chrs = chromosomes[idx]
_pv = pv[idx]
idx = sp.where(_pv > bf_threshold)[0]
__pos = _pos[idx]
__chrs = _chrs[idx]
__pv = _pv[idx]
__pv = -sp.log10(__pv)
pl.plot(
__pos + current_pos,
__pv,
'.',
alpha=0.3,
color="#A0A0A0",
markersize=5,
)
idx = sp.where(_pv <= bf_threshold)[0]
__pos = _pos[idx]
__chrs = _chrs[idx]
__pv = _pv[idx]
__pv = -sp.log10(__pv)
pl.plot(
__pos + current_pos,
__pv,
'.',
color="#b000ff",
alpha=0.9,
markersize=8,
)
xtick_list.append(float(current_pos) + float(_pos.max()) / 2.0)
current_pos = _pos.max() + current_pos
split_list.append(current_pos)
pl.xlim(0, current_pos)
pl.ylabel("$-log10(p-value)$")
pl.ylim(0, max_y)
for i in xrange(len(split_list)):
if i % 2 == 0:
try:
pl.fill_between([split_list[i], split_list[i + 1]],
0,
max_y,
color="#D7D7D7",
linewidth=0,
alpha=0.5)
except:
pass
s_list = split_list[:-1]
for split in s_list:
pl.axvline(split, color='k', linestyle='--')
pl.xticks(xtick_list, chrom_list)
pl.subplots_adjust(left=0.05,
bottom=0.09,
right=0.99,
top=0.9,
wspace=0.45)
remove_border()
if unsnps:
pv = unique_pv
if arguments.title:
if arguments.gc:
gc = sp.median(stats.chi2.isf(pv, 1)) / 0.456
pl.title("Phenotype: %s, genomic control $\hat \lambda=%0.2f$" %
(fname, gc))
else:
pl.title("Phenotype: %s" % (fname))
pl.savefig(
os.path.join(arguments.out,
'manhattan_' + fname + '.' + arguments.iformat))
pl.close()
def LDPlot(arguments,identifiers,encoded,maf,pv,unique_pv,positions,chromosomes,fname,pathogenicity_map):
font_size = 16
mpl.rcParams['font.family']="sans-serif"
mpl.rcParams['font.sans-serif']="Arial"
mpl.rcParams['font.size']=font_size
#mpl.rcParams['figure.dpi'] = 150
mpl.rcParams['font.weight']='medium'
mpl.rcParams['figure.facecolor'] = 'white'
mpl.rcParams['lines.linewidth'] = 1
mpl.rcParams['axes.facecolor'] = 'white'
mpl.rcParams['patch.edgecolor'] = 'white'
mpl.rcParams['grid.linestyle'] = '-'
mpl.rcParams['grid.color'] = 'LightGray'
if arguments.ignore!=None:
if arguments.ignore in fname:
return
if not arguments.sql_gene==None:
sqlite = sqlite3.connect(arguments.sql_gene)
sqlite_cursor = sqlite.cursor()
else:
sqlite = None
if arguments.nr_hypothesis==-1:
if arguments.distinct:
bf_threshold = 0.05/unique_pv.shape[0]
else:
bf_threshold = 0.05/pv.shape[0]
else:
bf_threshold = 0.05/arguments.nr_hypothesis
snp_distance = arguments.distance
r2_measure = arguments.r2_measure
pl.ion()
pl.figure(figsize=(20,4))
color_list = ['#F26C4F','#F68E55','#7CC576','#00BFF3','#605CA8','#F06EA9','#F26C4F','#F68E55']
#select different SNP
if arguments.selected_snp!='-1':
ind = sp.where(identifiers==arguments.selected_snp)[0]
if ind.shape[0]==0:
print "\nSNP " + arguments.selected_snp + " not found in dataset!"
print "Please select a different SNP identifier!\n"
quit()
else:
chrom_list = sp.array([arguments.selected_snp.split("_")[0]])
__pos = sp.array([int(arguments.selected_snp.split("_")[1])])
else:
chrom_list = sp.unique(chromosomes)
for i,chrom in enumerate(chrom_list):
idx = chromosomes==chrom
_pos = positions[idx]
_chrs = chromosomes[idx]
_pv = pv[idx]
if arguments.selected_snp=='-1':
idx = sp.where(_pv<=bf_threshold)[0]
__pos = _pos[idx]
__chrs = _chrs[idx]
__pv = _pv[idx]
__pv = -sp.log10(__pv)
else:
idx = sp.where(__pos[0]==_pos)[0]
__pv = _pv[idx]
__pv = -sp.log10(__pv)
#plot LD
cnorm = mcol.Normalize(vmin=0.0,vmax=1)
cpick = cm.ScalarMappable(norm=cnorm,cmap=pl.get_cmap("jet"))
cpick.set_array([])
ind = sp.argsort(_pos)
_pos = _pos[ind]
_pv = _pv[ind]
for k,pp in enumerate(__pos):
fig = pl.figure(figsize=(12,4))
if sqlite==None:
ax1 = pl.subplot2grid((5,5),(0,0),colspan=5,rowspan=4)
else:
ax1 = pl.subplot2grid((6,5),(0,0),colspan=5,rowspan=4)
ax1.axhline(-sp.log10(bf_threshold),color='#F68E55',linestyle='--')
marker = '.'
if str(chrom) + "_" + str(pp) in pathogenicity_map:
if pathogenicity_map[str(chrom) + "_" + str(pp)].prediction == "DELETERIOUS":
marker = '^'
else:
marker = 'v'
ax1.plot(pp,__pv[k],marker,
color="#b000ff",
alpha=0.9,
markersize=10,
)
if pp-snp_distance > 0 and pp+snp_distance<=_pos.max():
ranges = sp.where((_pos>=pp-snp_distance) & (_pos<=pp+snp_distance))[0]
elif pp-snp_distance < 0 and pp+snp_distance<=_pos.max():
ranges = sp.where(_pos<=pp+snp_distance)[0]
elif pp-snp_distance > 0 and pp+snp_distance>_pos.max():
ranges = sp.where(_pos>=pp-snp_distance)[0]
pp_idx = sp.where(pp==_pos)[0][0]
vline_map = {}
rr_color_list = []
maf_list = []
xlabels = [_pos[ranges][0],__pos[0],_pos[ranges][-1]]
idx = sp.where(str(chrom) + "_" + str(pp)==identifiers)[0]
for l,sra in enumerate(ranges):
sind = sp.where(str(chrom) + "_" + str(_pos[sra])==identifiers)[0]
maf_list.append(maf[sind])
if r2_measure=="excoffier_slatkin":
rr = ld.esem_r(sp.array(encoded[:,sind].flatten(),dtype="int"),sp.array(encoded[:,idx].flatten(),dtype="int"))**2
elif r2_measure=="roger_huff":
rr = ld.get_r(sp.array(encoded[:,sind].flatten(),dtype="int"),sp.array(encoded[:,idx].flatten(),dtype="int"))**2
elif r2_measure=="pearson_r2":
rr = stats.pearsonr(encoded[:,sind].flatten(),encoded[:,idx].flatten())[0]**2
rr_color_list.append(cpick.to_rgba(rr))
if _pos[sra]==pp:
ax1.vlines(pp, 0, __pv[k],color='#b000ff',linestyle='--',alpha=0.8,linewidth=0.5)
vline_map[pp] = "#b000ff"
continue
marker = '.'
if str(chrom) + "_" + str(_pos[sra]) in pathogenicity_map:
if pathogenicity_map[str(chrom) + "_" + str(_pos[sra])].prediction == "DELETERIOUS":
marker = '^'
else:
marker = 'v'
if rr>=0.7:
ax1.plot(_pos[sra],-sp.log10(_pv[sra]),marker,color=cpick.to_rgba(rr),alpha=0.9,markeredgecolor='#DDDDDD',markersize=10)#8
elif rr>=0.5:
ax1.plot(_pos[sra],-sp.log10(_pv[sra]),marker,color=cpick.to_rgba(rr),alpha=0.9,markeredgecolor='#DDDDDD',markersize=9)#7
elif rr>=0.3:
ax1.plot(_pos[sra],-sp.log10(_pv[sra]),marker,color=cpick.to_rgba(rr),alpha=0.9,markeredgecolor='#DDDDDD',markersize=8)#6
else:
ax1.plot(_pos[sra],-sp.log10(_pv[sra]),marker,color=cpick.to_rgba(rr),alpha=0.4,markeredgecolor='#DDDDDD',markersize=7)#4
if _pv[sra]<=bf_threshold:
ax1.vlines(_pos[sra], 0, -sp.log10(_pv[sra]),color=cpick.to_rgba(rr),linestyle='--',alpha=0.8,linewidth=0.5)
vline_map[_pos[sra]] = cpick.to_rgba(rr)
remove_border()
ax1.set_ylabel("$-log10(p-value)$")
ax1.set_ylim(0,sp.maximum(-sp.log10(bf_threshold)+1,sp.maximum((-sp.log10(_pv[ranges])).max()+1,-sp.log10(_pv[pp_idx])+1)))
if arguments.title:
ax1.set_title("Phenotype: %s, SNP: %d"%(fname,pp))
ax1.set_xticks([])
ax1.set_yticks(sp.arange(1,sp.ceil(sp.maximum(-sp.log10(bf_threshold)+1,sp.maximum((-sp.log10(_pv[ranges])).max()+1,-sp.log10(_pv[pp_idx])+1))),3))
ax1.set_xlim(_pos[ranges[0]]-snp_distance*0.01,_pos[ranges[-1]]+snp_distance*0.01)
ax1.yaxis.grid()
if len(pathogenicity_map)>0:
deleterious = pl.Line2D(range(1), range(1), marker='^', color="white",linestyle="None")
benign = pl.Line2D(range(1), range(1), marker='v', color="white",linestyle="None")
leg = pl.legend([benign,deleterious],['Benign Missense Mutation','Deleterious Missense Mutation'],frameon=True,scatterpoints=1,prop={'size':12},
fancybox=True,bbox_to_anchor=(1,1.2),numpoints=1,ncol=2)
leg.get_frame().set_alpha(0.5)
leg.get_frame().set_linewidth(0.2)
leg.get_frame().set_facecolor("#DDDDDD")
if sqlite==None:
ax3 = pl.subplot2grid((5,5),(4,0),colspan=5)
ax3.set_xlabel("Genomic positions on chromosome: " + str(chrom))
else:
ax3 = pl.subplot2grid((6,5),(4,0),colspan=5)
ax3.set_ylim(0,0.6)
ax3.set_yticks([0.25,0.5])
ax3.yaxis.grid()
ax3.set_ylabel("MAF")
ax3.set_xticks([])
remove_border(top=True)
ax3.set_xlim(_pos[ranges[0]]-snp_distance*0.01,_pos[ranges[-1]]+snp_distance*0.01)
for key in vline_map:
ax3.axvline(key,color=vline_map[key],linestyle='--',alpha=0.8,linewidth=0.5)
for i in xrange(ranges.shape[0]):
ax3.plot(_pos[ranges[i]],maf_list[i],'x',color=rr_color_list[i])
if sqlite!=None:
ax2 = pl.subplot2grid((6,5),(5,0),colspan=5)
remove_border(top=True)
ax2.set_xlabel("Genomic positions on chromosome: " + str(chrom))
ax2.set_yticks([])
ax2.set_xlim(_pos[ranges[0]]-snp_distance*0.01,_pos[ranges[-1]]+snp_distance*0.01)
for key in vline_map:
ax2.axvline(key,color=vline_map[key],linestyle='--',alpha=0.8,linewidth=0.5)
sqlite_cursor.execute("SELECT * FROM geneannotation WHERE chromosome_id=? AND annotation_end >=? AND annotation_start<=?",(str(chrom),int(_pos[ranges[0]]),int(_pos[ranges[-1]])))
annotations = sqlite_cursor.fetchall()
for g,annotation in enumerate(annotations):
alpha = 0.6
start = int(annotation[3])
length = int(annotation[4]) - int(annotation[3])
head_width=0.25
head_length=length*0.15
width=0.1
name = annotation[7].replace("Contig20-","")
if annotation[5]=="+":
shape = "right"
arrow_params={'length_includes_head':True, 'shape':shape,'head_starts_at_zero':True}
if g%2==0:
y_pos = 0.6
else:
y_pos = 0.5
y_text = y_pos+2*width
ax2.arrow(start,y_pos,length,0,head_width=head_width,head_length=head_length,fc=color_list[0],ec=color_list[0],alpha=alpha,width=width,**arrow_params)
else:
shape = "left"
arrow_params={'length_includes_head':True, 'shape':shape,'head_starts_at_zero':True}
if g%2==0:
y_pos = 0.15
else:
y_pos = 0.35
y_text = y_pos-2*width
ax2.arrow(start+length,y_pos,-length,0,head_width=head_width,head_length=head_length,fc=color_list[3],ec=color_list[3],alpha=alpha,width=width,**arrow_params)
ax2.text(start+length/2.0, y_text,name, size=7, ha='center', va='center', color="k")
pl.xticks(xlabels,xlabels)
#pl.gca().get_xaxis().get_major_formatter().set_useOffset(False)
if not sqlite==None:
cax = fig.add_axes([0.93, 0.4, 0.01, 0.5])
pl.colorbar(cpick,label="SNP r^2",cax=cax)
pl.subplots_adjust(left=0.07,bottom=0.15,right=0.92,top=0.9,wspace=0,hspace=0)
else:
pl.subplots_adjust(left=0.07,bottom=0.15,right=0.99,top=0.9,wspace=0,hspace=0)
if not arguments.title:
pl.subplots_adjust(top=0.9)
pl.savefig(os.path.join(arguments.out,'ld_plot_chrom_' + str(chrom) + "_pos_" + str(pp) + "_" + fname +'.' + arguments.iformat) )
pl.close()
def ManhattanPlot(arguments,pv,positions,chromosomes,hashs,unique_pv,fname):
font_size = 14
mpl.rcParams['font.family']="sans-serif"
mpl.rcParams['font.sans-serif']="Arial"
mpl.rcParams['font.size']=font_size
#mpl.rcParams['figure.dpi'] = 300
mpl.rcParams['font.weight']='medium'
mpl.rcParams['figure.facecolor'] = 'white'
mpl.rcParams['lines.linewidth'] = 1
mpl.rcParams['axes.facecolor'] = 'white'
mpl.rcParams['patch.edgecolor'] = 'white'
mpl.rcParams['grid.linestyle'] = '-'
mpl.rcParams['grid.color'] = 'LightGray'
if arguments.ignore!=None:
if arguments.ignore in fname:
return
chrom_list = sp.unique(chromosomes)
unsnps = arguments.distinct
pl.ion()
pl.figure(figsize=(20,3))
if arguments.nr_hypothesis==-1:
if unsnps:
bf_threshold = 0.05/unique_pv.shape[0]
else:
bf_threshold = 0.05/pv.shape[0]
else:
bf_threshold = 0.05/arguments.nr_hypothesis
current_pos = 0
max_y = (-sp.log10(pv[:])).max()
if max_y<-sp.log10(bf_threshold):
max_y = -sp.log10(bf_threshold)
max_y += 1
split_list = []
xtick_list = []
pl.axhline(-sp.log10(bf_threshold),color='#F68E55',linestyle='--')
if not arguments.nr_hypothesis2==-1:
pl.axhline(-sp.log10(0.05/arguments.nr_hypothesis2),color='#00BFF3',linestyle='--')
factor = sp.around(float(arguments.nr_hypothesis2)/float(arguments.nr_hypothesis))
leg = pl.legend(['Bonferroni','Bonferroni ' + str(factor) + 'x'],ncol=2,fancybox=True,prop={'size':12},bbox_to_anchor=(1,1.05))
leg.get_frame().set_alpha(0.8)
leg.get_frame().set_linewidth(0.2)
for i,chrom in enumerate(chrom_list):
idx = chromosomes==chrom
_pos = positions[idx]
_chrs = chromosomes[idx]
_pv = pv[idx]
idx = sp.where(_pv>bf_threshold)[0]
__pos = _pos[idx]
__chrs = _chrs[idx]
__pv = _pv[idx]
__pv = -sp.log10(__pv)
pl.plot(__pos+current_pos,__pv,'.',
alpha=0.3,
color="#A0A0A0",
markersize=5,
)
idx = sp.where(_pv<=bf_threshold)[0]
__pos = _pos[idx]
__chrs = _chrs[idx]
__pv = _pv[idx]
__pv = -sp.log10(__pv)
pl.plot(__pos+current_pos,__pv,'.',
color="#b000ff",
alpha=0.9,
markersize=8,
)
xtick_list.append(float(current_pos) + float(_pos.max())/2.0)
current_pos = _pos.max() + current_pos
split_list.append(current_pos)
pl.xlim(0,current_pos)
pl.ylabel("$-log10(p-value)$")
pl.ylim(0,max_y)
for i in xrange(len(split_list)):
if i%2==0:
try:
pl.fill_between([split_list[i],split_list[i+1]],0,max_y,color="#D7D7D7",linewidth=0,alpha=0.5)
except:
pass
s_list = split_list[:-1]
for split in s_list:
pl.axvline(split,color='k',linestyle='--')
pl.xticks(xtick_list,chrom_list)
pl.subplots_adjust(left=0.05,bottom=0.09,right=0.99,top=0.9,wspace=0.45)
remove_border()
if unsnps:
pv = unique_pv
if arguments.title:
if arguments.gc:
gc = sp.median(stats.chi2.isf(pv,1))/0.456
pl.title("Phenotype: %s, genomic control $\hat \lambda=%0.2f$"%(fname,gc))
else:
pl.title("Phenotype: %s"%(fname))
pl.savefig(os.path.join(arguments.out,'manhattan_' + fname + '.' + arguments.iformat) )
pl.close()
