def add_info(
peaks,
rip_targets,
gtf_sep_cols,
gtf_sep_cols_dict,
peaks_fname,
sequences,
deseq=None,
cufflinks=None,
):
"""Adds sequences, fbes, RIP-chip and peak location.
Takes a dataframe.
Returns the altered dataframe."""
if (deseq is not None) and (cufflinks is not None):
# Compare with the list in Ortiz et al. of gonadal genes.
add_gonad_expression(peaks, deseq, cufflinks)
# Add genomic sequence in the peak.
peaks = subset_peaks_with_fbe.add_seqs(peaks, sequences)
# Get yes/no FBE, -1/-2 C, number of FBEs.
subset_peaks_with_fbe.score_binding_site(peaks)
# Overlaps with the top 1350 unique RIP-chip targets?
compare_with_ripchip.add_column_of_overlap(peaks, rip_targets)
# Locate each peak in the gene.
if re.search('fbf1', peaks_fname):
locate_in_gene(gtf_sep_cols_dict,
peaks,
use_this_column='fbf1_reads_pos_of_max_coverage')
elif re.search('fbf2', peaks_fname):
locate_in_gene(gtf_sep_cols_dict,
peaks,
use_this_column='fbf2_reads_pos_of_max_coverage')
else:
locate_in_gene(gtf_sep_cols_dict, peaks)
return peaks
def add_info(peaks, rip_targets, gtf_sep_cols,
gtf_sep_cols_dict, peaks_fname, sequences,
deseq=None, cufflinks=None,
):
"""Adds sequences, fbes, RIP-chip and peak location.
Takes a dataframe.
Returns the altered dataframe."""
if (deseq is not None) and (cufflinks is not None):
# Compare with the list in Ortiz et al. of gonadal genes.
add_gonad_expression(peaks, deseq, cufflinks)
# Add genomic sequence in the peak.
peaks = subset_peaks_with_fbe.add_seqs(peaks, sequences)
# Get yes/no FBE, -1/-2 C, number of FBEs.
subset_peaks_with_fbe.score_binding_site(peaks)
# Overlaps with the top 1350 unique RIP-chip targets?
compare_with_ripchip.add_column_of_overlap(
peaks, rip_targets)
# Locate each peak in the gene.
if re.search('fbf1', peaks_fname):
locate_in_gene(
gtf_sep_cols_dict, peaks, use_this_column='fbf1_reads_pos_of_max_coverage')
elif re.search('fbf2', peaks_fname):
locate_in_gene(
gtf_sep_cols_dict, peaks, use_this_column='fbf2_reads_pos_of_max_coverage')
else:
locate_in_gene(
gtf_sep_cols_dict, peaks)
return peaks
def add_minimal_info(peaks, rip_targets, peaks_fname, sequences):
# Compare with the list in Ortiz et al. of gonadal genes.
# Add genomic sequence in the peak.
print '1'
peaks = subset_peaks_with_fbe.add_seqs(peaks, sequences)
# Get yes/no FBE, -1/-2 C, number of FBEs.
print '2'
subset_peaks_with_fbe.score_binding_site(peaks)
# Overlaps with the top 1350 unique RIP-chip targets?
print '3'
compare_with_ripchip.add_column_of_overlap(peaks, rip_targets)
print 'done'
def add_info(peaks, filename, rip_targets, gtf_sep_cols, top_level_dir):
# Add genomic sequence in the peak.
subset_peaks_with_fbe.get_sequences(peaks)
# Get yes/no FBE, -1/-2 C, number of FBEs.
subset_peaks_with_fbe.score_binding_site(peaks)
# Overlaps with the top 1350 unique RIP-chip targets?
compare_with_ripchip.add_column_of_overlap(
peaks, rip_targets)
# Locate each peak in the gene.
locate_in_gene(gtf_sep_cols, peaks)
subset_peaks_with_fbe.write_subset_with_fbe(
peaks, top_level_dir, label=os.path.basename(filename))
write_subset_of_columns(
peaks, top_level_dir, label=os.path.basename(filename))
def add_minimal_info(peaks, rip_targets,
peaks_fname, sequences):
# Compare with the list in Ortiz et al. of gonadal genes.
# Add genomic sequence in the peak.
print '1'
peaks = subset_peaks_with_fbe.add_seqs(peaks, sequences)
# Get yes/no FBE, -1/-2 C, number of FBEs.
print '2'
subset_peaks_with_fbe.score_binding_site(peaks)
# Overlaps with the top 1350 unique RIP-chip targets?
print '3'
compare_with_ripchip.add_column_of_overlap(
peaks, rip_targets)
print 'done'
def add_seqs(combined, sequences):
seq = [seq_from_iv(tup[0], tup[1], tup[2], tup[3], sequences) \
for tup in \
zip(combined.chrm, combined.left, combined.right, combined.strand)]
combined['seq'] = seq
return combined
def write_subset_with_fbe(peaks, top_level_dir, label):
top_level_dir = os.path.dirname(top_level_dir)
peaks_w_fbe = peaks[peaks['has_fbe']==1]
if not os.path.exists('with_fbe_%s' % top_level_dir):
os.system('mkdir with_fbe_%s' % top_level_dir)
peaks_w_fbe.to_csv('with_fbe_%s/%s' % (top_level_dir, label), sep='\t')
if __name__ == '__main__':
top_level_dir = sys.argv[1]
combined = {}
rip_targets = compare_with_ripchip.get_ripchip_targets()
for filename in glob.glob(top_level_dir + '/combined*.txt'):
print filename
combined[filename] = pandas.read_csv(filename, sep='\t')
get_sequences(combined[filename])
score_binding_site(combined[filename])
compare_with_ripchip.add_column_of_overlap(
combined[filename], rip_targets)
write_subset_with_fbe(
combined[filename], top_level_dir, label=os.path.basename(filename))