def get_gc(qsorted_bam_file, reference_fasta, prefix, java_heap=None):
'''
Uses picard tools (CollectGcBiasMetrics). Note that the reference
MUST be the same fasta file that generated the bowtie indices.
Assumes picard was already loaded into space (module add picard-tools)
'''
# remove redundant (or malformed) info (read group) from bam
logging.info('Getting GC bias...')
output_file = '{0}.gc.txt'.format(prefix)
plot_file = '{0}.gcPlot.pdf'.format(prefix)
summary_file = '{0}.gcSummary.txt'.format(prefix)
if java_heap is None:
java_heap_param = '-Xmx10G'
else:
java_heap_param = '-Xmx{}'.format(java_heap)
get_gc_metrics = ('java {6} -XX:ParallelGCThreads=1 -jar '
'{5} '
'CollectGcBiasMetrics R={0} I={1} O={2} '
'USE_JDK_DEFLATER=TRUE USE_JDK_INFLATER=TRUE '
'VERBOSITY=ERROR QUIET=TRUE '
'ASSUME_SORTED=FALSE '
'CHART={3} S={4}').format(reference_fasta,
qsorted_bam_file,
output_file,
plot_file,
summary_file,
locate_picard(),
java_heap_param)
logging.info(get_gc_metrics)
os.system(get_gc_metrics)
return output_file, plot_file, summary_file
def run_preseq(bam_w_dups, prefix):
'''
Runs preseq. Look at preseq data output to get PBC/NRF.
'''
# First sort because this file no longer exists...
sort_bam = 'samtools sort -o {1}.sorted.bam -T {1} -@ 2 {0}'.format(
bam_w_dups, prefix)
os.system(sort_bam)
logging.info('Running preseq...')
preseq_data = '{0}.preseq.dat'.format(prefix)
preseq_log = '{0}.preseq.log'.format(prefix)
preseq = ('preseq lc_extrap '
'-P -B -o {0} {1}.sorted.bam -seed 1 -v 2> {2}').format(
preseq_data, prefix, preseq_log)
logging.info(preseq)
os.system(preseq)
os.system('rm {0}.sorted.bam'.format(prefix))
return preseq_data, preseq_log
def run_preseq(bam_w_dups, prefix, nth=1, mem_gb=None):
'''
Runs preseq. Look at preseq data output to get PBC/NRF.
'''
# First sort because this file no longer exists...
sort_bam = samtools_sort(bam_w_dups, nth, mem_gb)
logging.info('Running preseq...')
preseq_data = '{0}.preseq.dat'.format(prefix)
preseq_log = '{0}.preseq.log'.format(prefix)
run_shell_cmd('preseq lc_extrap -P -B -o {preseq_data} {sort_bam} '
'-seed 1 -v 2> {preseq_log}'.format(
preseq_data=preseq_data,
sort_bam=sort_bam,
preseq_log=preseq_log,
))
rm_f(sort_bam)
return preseq_data, preseq_log
def make_tss_plot(bam_file,
tss,
prefix,
chromsizes,
read_len,
bins=400,
bp_edge=2000,
processes=8,
greenleaf_norm=True):
'''
Take bootstraps, generate tss plots, and get a mean and
standard deviation on the plot. Produces 2 plots. One is the
aggregation plot alone, while the other also shows the signal
at each TSS ordered by strength.
'''
logging.info('Generating tss plot...')
tss_plot_file = '{0}.tss_enrich.png'.format(prefix)
tss_plot_large_file = '{0}.large_tss_enrich.png'.format(prefix)
tss_log_file = '{0}.tss_enrich.qc'.format(prefix)
# Load the TSS file
tss = pybedtools.BedTool(tss)
tss_ext = tss.slop(b=bp_edge, g=chromsizes)
# Load the bam file
# Need to shift reads and just get ends, just load bed file?
bam = metaseq.genomic_signal(bam_file, 'bam')
# Shift to center the read on the cut site
bam_array = bam.array(tss_ext,
bins=bins,
shift_width=-read_len / 2,
processes=processes,
stranded=True)
# Normalization (Greenleaf style): Find the avg height
# at the end bins and take fold change over that
if greenleaf_norm:
# Use enough bins to cover 100 bp on either end
num_edge_bins = int(100 / (2 * bp_edge / bins))
bin_means = bam_array.mean(axis=0)
avg_noise = (sum(bin_means[:num_edge_bins]) +
sum(bin_means[-num_edge_bins:])) / (2 * num_edge_bins)
bam_array /= avg_noise
else:
bam_array /= bam.mapped_read_count() / 1e6
# Generate a line plot
fig = plt.figure()
ax = fig.add_subplot(111)
x = np.linspace(-bp_edge, bp_edge, bins)
ax.plot(x, bam_array.mean(axis=0), color='r', label='Mean')
ax.axvline(0, linestyle=':', color='k')
# Note the middle high point (TSS)
tss_point_val = max(bam_array.mean(axis=0))
# write tss_point_val to file
with open(tss_log_file, 'w') as fp:
fp.write(str(tss_point_val))
ax.set_xlabel('Distance from TSS (bp)')
if greenleaf_norm:
ax.set_ylabel('TSS Enrichment')
else:
ax.set_ylabel('Average read coverage (per million mapped reads)')
ax.legend(loc='best')
fig.savefig(tss_plot_file)
# Print a more complicated plot with lots of info
# Find a safe upper percentile - we can't use X if the Xth percentile is 0
upper_prct = 99
if mlab.prctile(bam_array.ravel(), upper_prct) == 0.0:
upper_prct = 100.0
plt.rcParams['font.size'] = 8
fig = metaseq.plotutils.imshow(bam_array,
x=x,
figsize=(5, 10),
vmin=5,
vmax=upper_prct,
percentile=True,
line_kwargs=dict(color='k', label='All'),
fill_kwargs=dict(color='k', alpha=0.3),
sort_by=bam_array.mean(axis=1))
# And save the file
fig.savefig(tss_plot_large_file)
return tss_plot_file, tss_plot_large_file, tss_log_file