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 make_output_folders(data_folder, adaID, VERBOSE=0):
'''Make output folders for the script'''
from hivwholeseq.utils.generic import mkdirs
dirname = data_folder + foldername_adapter(adaID) + 'map_iter/'
mkdirs(dirname)
if VERBOSE:
print 'Folder created:', dirname
def make_output_folders(data_folder,
adapters_designed,
VERBOSE=0,
summary=True):
'''Make output folders for all adapters and unclassified (e.g. PhiX)'''
from hivwholeseq.utils.generic import mkdirs
# Make folders for the samples
for (adaID, s) in adapters_designed:
dirname = foldername_adapter(adaID)
mkdirs(data_folder + dirname)
if VERBOSE:
print 'Folder created:', dirname
# Make a default directory for unclassified reads
mkdirs(data_folder + 'unclassified_reads')
if VERBOSE:
print 'Folder created: unclassified reads'
if summary:
with open(get_demultiplex_summary_filename(data_folder), 'a') as f:
f.write('\n')
f.write(
'Folders created for samples and unclassified reads (including phix).'
)
f.write('\n')
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 make_output_folders(data_folder, adaID, VERBOSE=0):
'''Make output folders for the script'''
from hivwholeseq.utils.generic import mkdirs
dirname = data_folder+foldername_adapter(adaID)+'map_iter/'
mkdirs(dirname)
if VERBOSE:
print 'Folder created:', dirname
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_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 make_output_folders(data_folder, adaID, VERBOSE=0):
'''Make output folders'''
from hivwholeseq.utils.generic import mkdirs
output_filename = get_divided_filenames(data_folder, adaID, fragments=['F1'])[0]
dirname = os.path.dirname(output_filename)
mkdirs(dirname)
if VERBOSE:
print 'Folder created:', dirname
def copy_folder(patient, dst_fn, foldername):
'''Copy a whole folder'''
src_fn = patient.folder+foldername+os.sep
map_fn = dst_fn+foldername+os.sep
mkdirs(map_fn)
for fn_src in src_fn.listdir():
copy(src_fn+fn_src, map_fn+fn_src)
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 make_output_folders(data_folder, adaID, VERBOSE=0, summary=True):
'''Make output folders'''
from hivwholeseq.utils.generic import mkdirs
outfiles = [get_premapped_filename(data_folder, adaID)]
if summary:
outfiles.append(get_coverage_figure_filename(data_folder, adaID, 'premapped'))
for outfile in outfiles:
dirname = os.path.dirname(outfile)
mkdirs(dirname)
if VERBOSE:
print 'Folder created:', dirname
def make_output_folders(data_folder, adaIDs, VERBOSE=0):
'''Make output folders for symlinking'''
from hivwholeseq.utils.generic import mkdirs
mkdirs(data_folder)
if VERBOSE >= 1:
print 'Folder created:', data_folder
for adaID in adaIDs + [-1]:
mkdirs(data_folder+foldername_adapter(adaID))
if VERBOSE >= 1:
print 'Folder created:', data_folder+foldername_adapter(adaID)
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 make_output_folders(data_folder, adaID, VERBOSE=0, summary=True):
'''Make output folders'''
from hivwholeseq.utils.generic import mkdirs
outfiles = [get_premapped_filename(data_folder, adaID)]
if summary:
outfiles.append(
get_coverage_figure_filename(data_folder, adaID, 'premapped'))
for outfile in outfiles:
dirname = os.path.dirname(outfile)
mkdirs(dirname)
if VERBOSE:
print 'Folder created:', dirname
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 copy_initial_reference(patient, dst_fn):
'''Copy initial patient mapping reference'''
ref_fn = dst_fn+'reference/'
mkdirs(ref_fn)
for fragment in fragments:
fn_src = patient.get_reference_filename(fragment)
fn_dst = ref_fn+os.path.basename(fn_src)
shutil.copy(fn_src, fn_dst)
fn_src = patient.get_reference_filename('genomewide', format='gb')
fn_dst = ref_fn+os.path.basename(fn_src)
shutil.copy(fn_src, fn_dst)
def make_output_folders(pname, samplename, PCR=1, VERBOSE=0):
'''Make the output folders if necessary for hash and map'''
hash_foldername = os.path.dirname(get_initial_hash_filename(pname, 'F0'))
map_foldername = get_mapped_to_initial_foldername(pname, samplename, PCR=PCR)
if not os.path.isdir(hash_foldername):
mkdirs(hash_foldername)
if VERBOSE:
print 'Folder created:', hash_foldername
mkdirs(map_foldername)
if VERBOSE:
print 'Folder created:', map_foldername
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 make_output_folders(data_folder, adapters_designed, VERBOSE=0, summary=True):
'''Make output folders for all adapters and unclassified (e.g. PhiX)'''
from hivwholeseq.utils.generic import mkdirs
# Make folders for the samples
for (adaID, s) in adapters_designed:
dirname = foldername_adapter(adaID)
mkdirs(data_folder+dirname)
if VERBOSE:
print 'Folder created:', dirname
# Make a default directory for unclassified reads
mkdirs(data_folder+'unclassified_reads')
if VERBOSE:
print 'Folder created: unclassified reads'
if summary:
with open(get_demultiplex_summary_filename(data_folder), 'a') as f:
f.write('\n')
f.write('Folders created for samples and unclassified reads (including phix).')
f.write('\n')
def predict_RNA_structure(seq, label='seq', maxstructs=1, VERBOSE=0):
'''Predict RNA secondary structures using RNAstructure'''
import os
import subprocess as sp
from hivwholeseq.utils.generic import mkdirs
from Bio import SeqIO
from Bio.Seq import Seq
from Bio.SeqRecord import SeqRecord
from Bio.Alphabet.IUPAC import ambiguous_dna
rna_fold_bin = '/ebio/ag-neher/home/fzanini/programs/RNAstructure_cli/exe/Fold'
# Make tmp input file
tmp_file_in = '/ebio/ag-neher/home/fzanini/tmp/RNAfold/'+label+'.fasta'
tmp_file_out = '/ebio/ag-neher/home/fzanini/tmp/RNAfold/'+label+'.ct'
mkdirs(os.path.dirname(tmp_file_in))
seqrec = SeqRecord(Seq(seq, ambiguous_dna),
id=label,
name=label,
description='')
SeqIO.write(seqrec, tmp_file_in, 'fasta')
# Call RNAStructure with all the crap (env vars, etc)
rna_tables = '/ebio/ag-neher/home/fzanini/programs/RNAstructure_cli/data_tables/'
env = os.environ.copy()
env['DATAPATH'] = rna_tables
call_list = [rna_fold_bin, '-m', str(maxstructs), tmp_file_in, tmp_file_out]
if VERBOSE >= 2:
print ' '.join(call_list)
output = sp.check_output(call_list, shell=False)
if VERBOSE >= 3:
print output
if 'Writing output ct file...done.' in output:
structs = parse_ct_file_multiple(tmp_file_out)
else:
IOError('RNAstructure had problems predicting the structure')
return structs
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()
'--repnumber',
type=int,
default=0,
help='Index of the sequenced sample within that patient sample')
args = parser.parse_args()
pname = args.patient
fragments = args.fragments
VERBOSE = args.verbose
repn = args.repnumber
samplename = args.sample
patient = load_patient(pname)
patient.discard_nonsequenced_samples()
mkdirs(get_initial_reference_foldername(pname))
if not fragments:
fragments = ['F' + str(i) for i in xrange(1, 7)]
if VERBOSE >= 3:
print 'fragments', fragments
if samplename is None:
sample = SamplePat(patient.samples.iloc[samplen])
else:
sample = load_sample_sequenced(samplename)
for fragment in fragments:
sample_seq = SampleSeq(sample.samples_seq.iloc[repn])
seq_run = sample_seq['seq run']
parser.add_argument('--sample',
help='Use a specific sample (not the first time point) for the reference')
parser.add_argument('--repnumber', type=int, default=0,
help='Index of the sequenced sample within that patient sample')
args = parser.parse_args()
pname = args.patient
fragments = args.fragments
VERBOSE = args.verbose
repn = args.repnumber
samplename = args.sample
patient = load_patient(pname)
patient.discard_nonsequenced_samples()
mkdirs(get_initial_reference_foldername(pname))
if not fragments:
fragments = ['F'+str(i) for i in xrange(1, 7)]
if VERBOSE >= 3:
print 'fragments', fragments
if samplename is None:
sample = SamplePat(patient.samples.iloc[samplen])
else:
sample = load_sample_sequenced(samplename)
for fragment in fragments:
sample_seq = SampleSeq(sample.samples_seq.iloc[repn])
seq_run = sample_seq['seq run']
# Script
if __name__ == '__main__':
parser = argparse.ArgumentParser(description="make figure")
parser.add_argument('--redo', action='store_true', help='recalculate data')
params = parser.parse_args()
fragment = 'F1'
VERBOSE = 2
username = os.path.split(os.getenv('HOME'))[-1]
foldername = get_figure_folder(username, 'first')
fn_data = foldername+'data/'
mkdirs(fn_data)
fn_data = fn_data + 'minor_alleles_example.pickle'
if not os.path.isfile(fn_data) or params.redo:
samplename = 'NL4-3'
sample = lss(samplename)
counts = sample.get_allele_counts(fragment, merge_read_types=True)
data = compress_data(counts, samplename, fragment)
samplename = '27134'
sample = lssp(samplename)
counts = sample.get_allele_counts(fragment, merge_read_types=True)
data = compress_data(counts, samplename, fragment, data=data)
store_data(data, fn_data)
import sys
import os
from hivwholeseq.utils.generic import mkdirs
from hivwholeseq.patients.samples import itersample
from hivwholeseq.sequencing.samples import load_samples_sequenced as lss
from hivwholeseq.patients.samples import load_samples_sequenced as lssp
from hivwholeseq.sequencing.filenames import get_sample_foldername
# Script
if __name__ == '__main__':
samples_pat = lssp()
samples_seq = lss()
for samplename, sample in itersample(samples_pat):
root_foldername = sample.get_foldername()+'samples_sequencing/'
mkdirs(root_foldername)
for samplenameseq, sampleseq in samples_seq.iterrows():
if sampleseq['patient sample'] == samplename:
src_folder = get_sample_foldername(samplenameseq)
dst_folder = root_foldername+samplenameseq
if not os.path.islink(dst_folder):
os.symlink(src_folder, dst_folder)
print 'Symlink:', src_folder, dst_folder
else:
print 'Esists:', dst_folder
dist_hist = get_distance_histogram(data_folder,
adaID,
fragment,
VERBOSE=VERBOSE)
except IOError:
continue
dist_hists.append((samplename_seq, fragment, dist_hist))
dist_hists.sort(key=itemgetter(1))
fig, ax = plt.subplots()
for i, (samplename_seq, fragment, h) in enumerate(dist_hists):
plot_distance_histogram(h,
ax=ax,
color=cm.jet(1.0 * i / len(dist_hists)),
label=', '.join([samplename_seq,
fragment]))
ax.set_title(samplename)
ax.legend(loc=1, fontsize=10)
if use_save:
foldername = sample.get_foldername() + 'figures/'
mkdirs(foldername)
fn = foldername + 'distance_to_consensus_seqsamples.png'
fig.savefig(fn)
plt.close(fig)
if not use_save:
plt.ion()
plt.show()
if not len(datum['ind']):
win_start += gap
continue
datum['times'] = patient.times[datum['ind']]
datum['pcode'] = patient.code
datum['window'] = (win_start, win_end)
data.append(datum)
if use_save:
if VERBOSE >= 2:
print 'Save to file'
rname = 'scan_' + str(win_start) + '-' + str(win_end)
fn_out = patient.get_haplotype_count_trajectory_filename(rname)
mkdirs(os.path.dirname(fn_out))
np.savez_compressed(
fn_out,
hct=datum['hct'],
ind=datum['ind'],
times=datum['times'],
seqs=datum['seqs'],
ali=datum['alim'],
)
if VERBOSE >= 2:
print 'Build tree'
times = datum['times']
alim = datum['alim']
hct = datum['hct']
hft = 1.0 * hct / hct.sum(axis=0)
adaID = sample_seq.adapter
for fragment in fragments:
try:
dist_hist = get_distance_histogram(data_folder, adaID, fragment,
VERBOSE=VERBOSE)
except IOError:
continue
dist_hists.append((samplename_seq, fragment, dist_hist))
dist_hists.sort(key=itemgetter(1))
fig, ax = plt.subplots()
for i, (samplename_seq, fragment, h) in enumerate(dist_hists):
plot_distance_histogram(h, ax=ax,
color=cm.jet(1.0 * i / len(dist_hists)),
label=', '.join([samplename_seq, fragment]))
ax.set_title(samplename)
ax.legend(loc=1, fontsize=10)
if use_save:
foldername = sample.get_foldername()+'figures/'
mkdirs(foldername)
fn = foldername+'distance_to_consensus_seqsamples.png'
fig.savefig(fn)
plt.close(fig)
if not use_save:
plt.ion()
plt.show()
parser = argparse.ArgumentParser(description='Copy data folder',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('destination',
help='Destination folder')
parser.add_argument('--strip-PCR1', action='store_true',
help='Strip the ../PCR1/ part of the file tree')
args = parser.parse_args()
dst_fn = args.destination.lstrip(os.sep)+os.sep
stirp_PCR1 = args.strip_PCR1
patients_fn = dst_fn+'patients/'
ref_fn = dst_fn+'reference/'
print 'Make root folders'
mkdirs(patients_fn)
mkdirs(ref_fn)
print 'Reference sequences'
copy_reference(ref_fn)
patients = load_patients()
for pname, patient in patients.iterrows():
print pname
patient = Patient(patient)
print 'Make folder'
pat_fn = patients_fn+pname+os.sep
mkdirs(pat_fn)
if not len(datum['ind']):
win_start += gap
continue
datum['times'] = patient.times[datum['ind']]
datum['pcode'] = patient.code
datum['window'] = (win_start, win_end)
data.append(datum)
if use_save:
if VERBOSE >= 2:
print 'Save to file'
rname = 'scan_'+str(win_start)+'-'+str(win_end)
fn_out = patient.get_haplotype_count_trajectory_filename(rname)
mkdirs(os.path.dirname(fn_out))
np.savez_compressed(fn_out,
hct=datum['hct'],
ind=datum['ind'],
times=datum['times'],
seqs=datum['seqs'],
ali=datum['alim'],
)
if VERBOSE >= 2:
print 'Build tree'
times = datum['times']
alim = datum['alim']
hct = datum['hct']
hft = 1.0 * hct / hct.sum(axis=0)
ali = expand_annotate_alignment(alim, hft, hct, times,
# Script
if __name__ == '__main__':
parser = argparse.ArgumentParser(description="make figure")
parser.add_argument('--redo', action='store_true', help='recalculate data')
params = parser.parse_args()
fragment = 'F1'
VERBOSE = 2
username = os.path.split(os.getenv('HOME'))[-1]
foldername = get_figure_folder(username, 'first')
fn_data = foldername + 'data/'
mkdirs(fn_data)
fn_data = fn_data + 'minor_alleles_example.pickle'
if not os.path.isfile(fn_data) or params.redo:
samplename = 'NL4-3'
sample = lss(samplename)
counts = sample.get_allele_counts(fragment, merge_read_types=True)
data = compress_data(counts, samplename, fragment)
samplename = '27134'
sample = lssp(samplename)
counts = sample.get_allele_counts(fragment, merge_read_types=True)
data = compress_data(counts, samplename, fragment, data=data)
store_data(data, fn_data)
else: