def plot_cumulative_histogram(data_folder, adaID, fragment, insert_sizes,
title=None,
ax=None,
show=False, savefig=False,
**kwargs):
'''Plot cumulative histogram of insert sizes'''
import matplotlib.pyplot as plt
if ax is None:
fig, ax = plt.subplots(1, 1)
ax.plot(insert_sizes, np.linspace(0, 1, len(insert_sizes)), **kwargs)
ax.set_xlabel('Insert size')
ax.set_ylabel('Cumulative fraction')
ax.set_xlim(-1, 1000)
ax.set_ylim(-0.02, 1.02)
if title is not None:
ax.set_title(title)
plt.tight_layout()
if show:
plt.ion()
plt.show()
if savefig:
output_filename = get_insert_size_distribution_cumulative_filename(data_folder,
adaID,
fragment)
from hivwholeseq.utils.generic import mkdirs
from hivwholeseq.sequencing.filenames import get_figure_folder
mkdirs(get_figure_folder(data_folder, adaID))
fig.savefig(output_filename)
def plot_quality_along_reads(data_folder, adaID, title, quality, VERBOSE=0, savefig=False):
"""Plot the results of the quality scores along reads"""
import matplotlib.pyplot as plt
from matplotlib import cm
fig, axs = plt.subplots(1, 2, figsize=(16, 9))
for i, (ax, qual) in enumerate(izip(axs, quality)):
for j, qpos in enumerate(qual):
x = qpos
y = np.linspace(0, 1, len(x))[::-1]
ax.plot(x, y, color=cm.jet(int(255.0 * j / len(qual))), alpha=0.5, lw=2)
ax.set_xlabel("Phred quality", fontsize=14)
ax.set_ylabel("Fraction of bases above quality x", fontsize=14)
ax.set_title("Read" + str(i + 1), fontsize=16)
ax.text(2, 0.03, "blue to red: 0 to " + str(len(qual)) + " base", fontsize=18)
fig.suptitle(title, fontsize=20)
if savefig:
from hivwholeseq.utils.generic import mkdirs
from hivwholeseq.sequencing.filenames import get_figure_folder, get_quality_along_reads_filename
fig_folder = get_figure_folder(data_folder, adaID)
fig_filename = get_quality_along_reads_filename(data_folder, adaID)
mkdirs(fig_folder)
fig.savefig(fig_filename)
else:
plt.tight_layout()
plt.ion()
plt.show()
def plot_histogram(data_folder, adaID, fragment, h,
title=None,
ax=None,
show=False, savefig=False,
**kwargs):
'''Plot histogram of insert sizes'''
import matplotlib.pyplot as plt
if ax is None:
fig, ax = plt.subplots(1, 1)
if title is not None:
ax.set_title(title)
x = 0.5 * (h[1][1:] + h[1][:-1])
y = h[0]
ax.plot(x, y, **kwargs)
ax.set_xlabel('Insert size')
ax.set_ylabel('Density')
plt.tight_layout()
if show:
plt.ion()
plt.show()
if savefig:
output_filename = get_insert_size_distribution_filename(data_folder, adaID,
fragment)
from hivwholeseq.utils.generic import mkdirs
from hivwholeseq.sequencing.filenames import get_figure_folder
mkdirs(get_figure_folder(data_folder, adaID))
plt.savefig(output_filename)
def plot_quality_along_reads(data_folder, adaID, title, quality, VERBOSE=0, savefig=False):
'''Plot the results of the quality scores along reads'''
import matplotlib.pyplot as plt
from matplotlib import cm
fig, axs = plt.subplots(1, 2, figsize=(16, 9))
for i, (ax, qual) in enumerate(izip(axs, quality)):
for j, qpos in enumerate(qual):
x = qpos
y = np.linspace(0, 1, len(x))[::-1]
ax.plot(x, y, color=cm.jet(int(255.0 * j / len(qual))),
alpha=0.5,
lw=2)
ax.set_xlabel('Phred quality', fontsize=14)
ax.set_ylabel('Fraction of bases above quality x', fontsize=14)
ax.set_title('Read'+str(i+1), fontsize=16)
ax.text(2, 0.03, 'blue to red: 0 to '+str(len(qual))+' base', fontsize=18)
fig.suptitle(title, fontsize=20)
if savefig:
from hivwholeseq.utils.generic import mkdirs
from hivwholeseq.sequencing.filenames import get_figure_folder, \
get_quality_along_reads_filename
fig_folder = get_figure_folder(data_folder, adaID)
fig_filename = get_quality_along_reads_filename(data_folder, adaID)
mkdirs(fig_folder)
fig.savefig(fig_filename)
else:
plt.tight_layout()
plt.ion()
plt.show()
def plot_cuts_quality_along_reads(data_folder,
adaID,
quality,
title='',
VERBOSE=0,
savefig=False):
'''Plot some cuts of the quality along the read'''
from scipy.stats import percentileofscore as pof
import matplotlib.pyplot as plt
from matplotlib import cm
fig, axs = plt.subplots(1, 2, figsize=(14, 8))
qthreshs = [10, 20, 30, 35]
for i, (ax, qual) in enumerate(izip(axs, quality)):
for j, qthresh in enumerate(qthreshs):
x = np.arange(len(qual))
y = np.array(
[100 - pof(qual[k], qthresh) for k in xrange(len(qual))])
ax.plot(x,
y,
color=cm.jet(int(255.0 * j / len(qthreshs))),
alpha=0.8,
lw=2,
label='Q = ' + str(qthresh))
ax.set_xlabel('Position [bp]', fontsize=14)
ax.set_ylabel('Percentage of bases above quality x', fontsize=14)
ax.set_title('Read' + str(i + 1), fontsize=16)
ax.set_ylim(-1, 101)
ax.set_xlim(-1, len(qual) + 1)
ax.legend(loc='best')
if title:
fig.suptitle(title, fontsize=20)
if savefig:
from hivwholeseq.utils.generic import mkdirs
if savefig == True:
from hivwholeseq.sequencing.filenames import get_figure_folder, \
get_quality_along_reads_filename
fig_folder = get_figure_folder(data_folder, adaID)
fig_filename = get_quality_along_reads_filename(data_folder,
adaID,
simple=True)
elif isinstance(savefig, basestring):
import os
fig_filename = savefig
fig_folder = os.path.dirname(fig_filename)
else:
raise ValueError(
'savefig must be a bool or a figure filename (string)')
mkdirs(fig_folder)
fig.savefig(fig_filename)
else:
plt.tight_layout()
plt.ion()
plt.show()
def report_coverage(data_folder, adaID, VERBOSE=0, summary=True):
'''Produce a report on rough coverage on reference (ignore inserts)'''
ref_filename = get_reference_premap_filename(data_folder, adaID)
refseq = SeqIO.read(ref_filename, 'fasta')
# Prepare data structures
coverage = np.zeros(len(refseq), int)
# Parse the BAM file
unmapped = 0
mapped = 0
bamfilename = get_premapped_filename(data_folder, adaID, type='bam')
with pysam.Samfile(bamfilename, 'rb') as bamfile:
for read in bamfile:
if read.is_unmapped or (not read.is_proper_pair) or (not len(
read.cigar)):
unmapped += 1
continue
# Proceed along CIGARs
ref_pos = read.pos
for (bt, bl) in read.cigar:
if bt not in (0, 2):
continue
# Treat deletions as 'covered'
coverage[ref_pos:ref_pos + bl] += 1
ref_pos += bl
mapped += 1
# Save results
from hivwholeseq.sequencing.filenames import get_coverage_figure_filename
import matplotlib.pyplot as plt
fig, ax = plt.subplots(1, 1, figsize=(13, 6))
ax.plot(np.arange(len(refseq)), coverage + 1, lw=2, c='b')
ax.set_xlabel('Position')
ax.set_ylabel('Coverage')
ax.set_yscale('log')
ax.set_title('adaID ' + adaID + ', premapped', fontsize=18)
ax.set_xlim(-20, len(refseq) + 20)
plt.tight_layout()
from hivwholeseq.utils.generic import mkdirs
from hivwholeseq.sequencing.filenames import get_figure_folder
mkdirs(get_figure_folder(data_folder, adaID))
plt.savefig(get_coverage_figure_filename(data_folder, adaID, 'premapped'))
plt.close(fig)
if summary:
with open(get_premap_summary_filename(data_folder, adaID), 'a') as f:
f.write('\nPremapping results: '+\
str(mapped)+' read pairs mapped, '+str(unmapped)+' unmapped\n')
f.write('\nCoverage plotted: '+\
get_coverage_figure_filename(data_folder, adaID, 'premapped')+'\n')
def report_coverage(data_folder, adaID, VERBOSE=0, summary=True):
'''Produce a report on rough coverage on reference (ignore inserts)'''
ref_filename = get_reference_premap_filename(data_folder, adaID)
refseq = SeqIO.read(ref_filename, 'fasta')
# Prepare data structures
coverage = np.zeros(len(refseq), int)
# Parse the BAM file
unmapped = 0
mapped = 0
bamfilename = get_premapped_filename(data_folder, adaID, type='bam')
with pysam.Samfile(bamfilename, 'rb') as bamfile:
for read in bamfile:
if read.is_unmapped or (not read.is_proper_pair) or (not len(
read.cigar)):
unmapped += 1
continue
# Proceed along CIGARs
ref_pos = read.pos
for (bt, bl) in read.cigar:
if bt not in (0, 2):
continue
# Treat deletions as 'covered'
coverage[ref_pos:ref_pos + bl] += 1
ref_pos += bl
mapped += 1
# Save results
from hivwholeseq.sequencing.filenames import get_coverage_figure_filename
import matplotlib.pyplot as plt
fig, ax = plt.subplots(1, 1, figsize=(13, 6))
ax.plot(np.arange(len(refseq)), coverage + 1, lw=2, c='b')
ax.set_xlabel('Position')
ax.set_ylabel('Coverage')
ax.set_yscale('log')
ax.set_title('adaID ' + adaID + ', premapped', fontsize=18)
ax.set_xlim(-20, len(refseq) + 20)
plt.tight_layout()
from hivwholeseq.utils.generic import mkdirs
from hivwholeseq.sequencing.filenames import get_figure_folder
mkdirs(get_figure_folder(data_folder, adaID))
plt.savefig(get_coverage_figure_filename(data_folder, adaID, 'premapped'))
plt.close(fig)
if summary:
with open(get_premap_summary_filename(data_folder, adaID), 'a') as f:
f.write('\nPremapping results: '+\
str(mapped)+' read pairs mapped, '+str(unmapped)+' unmapped\n')
f.write('\nCoverage plotted: '+\
get_coverage_figure_filename(data_folder, adaID, 'premapped')+'\n')
def plot_cuts_quality_along_reads(data_folder, adaID, quality, title="", VERBOSE=0, savefig=False):
"""Plot some cuts of the quality along the read"""
from scipy.stats import percentileofscore as pof
import matplotlib.pyplot as plt
from matplotlib import cm
fig, axs = plt.subplots(1, 2, figsize=(14, 8))
qthreshs = [10, 20, 30, 35]
for i, (ax, qual) in enumerate(izip(axs, quality)):
for j, qthresh in enumerate(qthreshs):
x = np.arange(len(qual))
y = np.array([100 - pof(qual[k], qthresh) for k in xrange(len(qual))])
ax.plot(x, y, color=cm.jet(int(255.0 * j / len(qthreshs))), alpha=0.8, lw=2, label="Q = " + str(qthresh))
ax.set_xlabel("Position [bp]", fontsize=14)
ax.set_ylabel("Percentage of bases above quality x", fontsize=14)
ax.set_title("Read" + str(i + 1), fontsize=16)
ax.set_ylim(-1, 101)
ax.set_xlim(-1, len(qual) + 1)
ax.legend(loc="best")
if title:
fig.suptitle(title, fontsize=20)
if savefig:
from hivwholeseq.utils.generic import mkdirs
if savefig == True:
from hivwholeseq.sequencing.filenames import get_figure_folder, get_quality_along_reads_filename
fig_folder = get_figure_folder(data_folder, adaID)
fig_filename = get_quality_along_reads_filename(data_folder, adaID, simple=True)
elif isinstance(savefig, basestring):
import os
fig_filename = savefig
fig_folder = os.path.dirname(fig_filename)
else:
raise ValueError("savefig must be a bool or a figure filename (string)")
mkdirs(fig_folder)
fig.savefig(fig_filename)
else:
plt.tight_layout()
plt.ion()
plt.show()
def plot_cuts_quality_along_reads(data_folder, adaID, title, quality, VERBOSE=0, savefig=False):
'''Plot some cuts of the quality along the read'''
from scipy.stats import percentileofscore as pof
import matplotlib.pyplot as plt
from matplotlib import cm
fig, axs = plt.subplots(1, 2, figsize=(14, 8))
qthreshs = [10, 20, 30, 35]
for i, (ax, qual) in enumerate(izip(axs, quality)):
for j, qthresh in enumerate(qthreshs):
x = np.arange(len(qual))
y = np.array([100 - pof(qual[k], qthresh) for k in xrange(len(qual))])
ax.plot(x, y, color=cm.jet(int(255.0 * j / len(qthreshs))),
alpha=0.8,
lw=2,
label='Q = '+str(qthresh))
ax.set_xlabel('Position [bp]', fontsize=14)
ax.set_ylabel('Percentage of bases above quality x', fontsize=14)
ax.set_title('Read'+str(i+1), fontsize=16)
ax.set_ylim(-1, 101)
ax.set_xlim(-1, len(qual) + 1)
ax.legend(loc='best')
fig.suptitle(title, fontsize=20)
if savefig:
from hivwholeseq.utils.generic import mkdirs
from hivwholeseq.sequencing.filenames import get_figure_folder, \
get_quality_along_reads_filename
fig_folder = get_figure_folder(data_folder, adaID)
fig_filename = get_quality_along_reads_filename(data_folder, adaID, simple=True)
mkdirs(fig_folder)
fig.savefig(fig_filename)
else:
plt.tight_layout()
plt.ion()
plt.show()
