def calc_maxima_forward_reverse_all_peaks(self):
self.log_info(
"Calculating the forward and reverse maxima for each of the peak regions..."
)
self.peak_maxima = defaultdict(
lambda: defaultdict(lambda: defaultdict(list)))
for chrom, start, end in self.peaks:
peak_str = shared.peak_str(start, end)
self.log_debug("Calculating maxima for peak %s..." % peak_str)
max_depth = max_depth = max(
self.genome.forward_depth[chrom][start:end + 1] +
self.genome.reverse_depth[chrom][start:end + 1])
forward = self.get_maxima_single(self.genome.forward_depth, chrom,
start, end, max_depth)
reverse = self.get_maxima_single(self.genome.reverse_depth, chrom,
start, end, max_depth)
self.peak_maxima[chrom][peak_str]["FORWARD"] = forward
self.peak_maxima[chrom][peak_str]["REVERSE"] = reverse
self.log_debug("Forward Maxima: %s" % shared.pprint_list(forward))
self.log_debug("Reverse Maxima: %s" % shared.pprint_list(reverse))
peak_pairs = self.calc_peak_pairs_single(forward, reverse)
self.log_debug("Peak Pairs: %s" % shared.pprint_list(peak_pairs))
self.peak_pairs[chrom][peak_str] = peak_pairs
def find_sites(self):
self.log_info("Building the genome read depth model...")
self.genome = GenomeAlignment(self.bam_file, self.peaks, self.log)
self.genome.generateModel()
self.log_debug("Finished building the genome model...")
self.log_debug("Creating the PeakQC object...")
self.peakQC = PeakQC(self.genome, self.peaks, self.log)
for chrom, start_zeroi, end_zeroi in self.peaks:
peak_str = shared.peak_str(start_zeroi, end_zeroi)
if self.log: self.log.info("Processing the peak %s..." % peak_str)
self.peakQC.perform_peak_QC(chrom, start_zeroi, end_zeroi)
continue
max_depth = max(
self.forward_read_count[chrom][start_zeroi:end_zeroi + 1] +
self.reverse_read_count[chrom][start_zeroi:end_zeroi + 1])
forward_maxima = self.get_maxima(self.forward_read_count, chrom,
start_zeroi, end_zeroi, max_depth,
lower_peak_cutoff_perc)
reverse_maxima = self.get_maxima(self.reverse_read_count, chrom,
start_zeroi, end_zeroi, max_depth,
lower_peak_cutoff_perc)
print start_zeroi + 1
print end_zeroi + 1
print ', '.join([str(i) for i in forward_maxima])
print ', '.join([str(i) for i in reverse_maxima])
self.peakQC.print_QC_log()
sys.exit()
def find_troughs(self):
for chrom, start, stop in self.peaks:
peak_str = shared.peak_str(start, stop)
for first_maxima, second_maxima in self.peak_pairs[chrom][
peak_str]:
trough = self.find_trough(chrom, first_maxima, second_maxima)
self.trough_sites[chrom][peak_str].append(trough)
def print_QC_log(self):
print '\t'.join([
"CHROM", "PEAK", self.depth_tag, self.exist_maxima_tag,
self.paired_maxima_tag, 'TroughSites', 'TargetSites', 'TargetSeqs'
])
for chrom, start, end in self.peaks:
peak_str = shared.peak_str(start, end)
out_list = [chrom, peak_str]
out_list.append(self.QC_log[chrom][peak_str][self.depth_tag])
out_list.append(
self.QC_log[chrom][peak_str][self.exist_maxima_tag])
out_list.append(
str(self.QC_log[chrom][peak_str][self.paired_maxima_tag]))
if len(self.peak_analyzer.trough_sites[chrom][peak_str]) > 0:
out_list.append(';'.join([
str(i + 1)
for i in self.peak_analyzer.trough_sites[chrom][peak_str]
]))
else:
out_list.append("None")
if len(self.peak_analyzer.target_sites[chrom][peak_str]) > 0:
targets = self.peak_analyzer.target_sites[chrom][peak_str]
out_targets = []
for peak in targets:
out = []
for target in peak:
out.append('-'.join([str(i + 1) for i in target]))
out_targets.append('|'.join(out))
out_str = ';'.join(out_targets)
out_list.append(out_str)
else:
out_list.append("None")
if len(self.peak_analyzer.target_seqs[chrom][peak_str]) > 0:
targets = self.peak_analyzer.target_seqs[chrom][peak_str]
out_targets = []
for peak in targets:
out = []
for target in peak:
out.append(target)
out_targets.append('|'.join(out))
out_str = ';'.join(out_targets)
out_list.append(out_str)
else:
out_list.append("None")
print "\t".join(out_list)
def perform_depth_QC(self, chrom, start, end):
peak_str = shared.peak_str(start, end)
max_depth = max(self.genome.forward_depth[chrom][start:end + 1] +
self.genome.reverse_depth[chrom][start:end + 1])
if max_depth < 20:
self.QC_log[chrom][peak_str][self.depth_tag] = 'LOW'
elif max_depth < 50:
self.QC_log[chrom][peak_str][self.depth_tag] = 'MED'
else:
self.QC_log[chrom][peak_str][self.depth_tag] = 'HIGH'
def print_QC_log(self):
print '\t'.join(["CHROM", "PEAK", self.depth_tag, self.exist_maxima_tag, self.paired_maxima_qc])
for chrom, start, end in self.peaks:
peak_str = shared.peak_str(start, end)
out_list = [chrom, peak_str]
out_list.append(self.QC_log[chrom][peak_str][self.depth_tag])
out_list.append(self.QC_log[chrom][peak_str][self.exist_maxima_tag])
out_list.append(self.QC_log[chrom][peak_str][self.paired_maxima_tag])
print "\t".join(out_list)
def perform_exist_maxima_QC(self, chrom, start, end):
peak_str = shared.peak_str(start, end)
if len(self.peak_analyzer.peak_maxima[chrom][peak_str]['FORWARD']) == 0 and \
len(self.peak_analyzer.peak_maxima[chrom][peak_str]['REVERSE']) == 0:
self.QC_log[chrom][peak_str][self.exist_maxima_tag] = 'FALSE'
elif len(self.peak_analyzer.peak_maxima[chrom][peak_str]['FORWARD']) == 0:
self.QC_log[chrom][peak_str][self.exist_maxima_tag] = 'MISSING_FORWARD'
elif len(self.peak_analyzer.peak_maxima[chrom][peak_str]['REVERSE']) == 0:
self.QC_log[chrom][peak_str][self.exist_maxima_tag] = 'MISSING_REVERSE'
else:
self.QC_log[chrom][peak_str][self.exist_maxima_tag] = 'TRUE'
def get_quality_peaks(self):
quality_peaks = []
for chrom, start, stop in self.peaks:
peak_str = shared.peak_str(start, stop)
if self.QC_log[chrom][peak_str][self.depth_tag] != 'LOW':
if self.QC_log[chrom][peak_str][
self.exist_maxima_tag] == "TRUE":
if self.QC_log[chrom][peak_str][
self.paired_maxima_tag] > 0:
quality_peaks.append([chrom, start, stop])
return quality_peaks
def find_targets(self, peaks):
for chrom, start, stop in peaks:
peak_str = shared.peak_str(start, stop)
for first_maxima, second_maxima in self.peak_pairs[chrom][
peak_str]:
targets = self.find_targets_single_peak(
chrom, first_maxima, second_maxima)
if len(targets) > 0:
target_seqs = [
self.genome.genome_seq[chrom][target[0]:target[1] + 1]
for target in targets
]
#print target_seqs
self.target_seqs[chrom][peak_str].append(target_seqs)
self.target_sites[chrom][peak_str].append(targets)
def find_sites(self):
self.log_info("Building the genome read depth model...")
self.genome = GenomeAlignment(self.genome_path, self.bam_file, self.peaks, self.log)
self.genome.generateModel()
self.log_debug("Finished building the genome model...")
self.log_debug("Creating the PeakQC object...")
self.peakQC = PeakQC(self.genome, self.peaks, self.log)
for chrom, start_zeroi, end_zeroi in self.peaks:
peak_str = shared.peak_str(start_zeroi, end_zeroi)
if self.log: self.log.info("Processing the peak %s..." % peak_str)
self.peakQC.perform_peak_QC(chrom, start_zeroi, end_zeroi)
self.peakQC.find_insertion_sites()
self.peakQC.print_QC_log()
sys.exit()
def perform_paired_maxima_QC(self, chrom, start, end):
peak_str = shared.peak_str(start, end)
self.peak_analyzer
def perform_paired_maxima_QC(self, chrom, start, end):
peak_str = shared.peak_str(start, end)
paired_peaks = len(self.peak_analyzer.peak_pairs[chrom][peak_str])
self.QC_log[chrom][peak_str][self.paired_maxima_tag] = paired_peaks
