def bg_len_GC_bins(bg_file, bg_dir):
"""
Compute G+C content for all sequences in the background.
Compute and store the %GC information in a list. To each G+C percentage
bin, we associate the corresponding sequence names.
Return lists of GC contents, GC bins, and lengths distrib.
"""
with open_for_parsing(bg_file) as stream:
gc_bins = []
gc_list = []
lengths = []
dinuc = [0] * len(IUPAC) * len(IUPAC)
for _ in range(0, 101):
gc_bins.append({})
for record in SeqIO.parse(stream, "fasta"):
gc = GC(record.seq)
gc_list.append(gc)
if len(record) in gc_bins[gc]:
gc_bins[gc][len(record)].append(record)
else:
gc_bins[gc][len(record)] = [record]
lengths.append(len(record.seq))
dinuc = [x + y for x, y in zip(dinuc, dinuc_count(record.seq))]
print_in_bg_dir(gc_bins, bg_dir, True)
return gc_list, gc_bins, lengths, dinuc
def fg_len_GC_bins(fg_file):
"""
Compute G+C content for all sequences in the foreground.
Computes %GC contant and store the information in a list. To each G+C
percentage bin, we associate the number of sequences falling in the
corresponding bin.
Return lists of GC contents, GC bins, and lengths distrib.
"""
with open_for_parsing(fg_file) as stream:
gc_bins = []
for _ in range(0, 101):
gc_bins.append({})
gc_list = []
lengths = []
dinuc = [0] * len(IUPAC) * len(IUPAC)
for record in SeqIO.parse(stream, "fasta"):
gc = GC(record.seq)
gc_list.append(gc)
length = len(record)
dinuc = [x + y for x, y in zip(dinuc, dinuc_count(record.seq))]
lengths.append(length)
if length in gc_bins[gc]:
gc_bins[gc][length] += 1
else:
gc_bins[gc][length] = 1
return gc_list, gc_bins, lengths, dinuc
def bg_GC_bins(bg_file, bg_dir):
"""
Compute G+C content for all sequences in the background.
Compute and store the GC information in a list. To each G+C percentage bin,
we associate the corresponding sequence names.
Files representing the binning are stored in the "bg_dir" directory.
Return lists of GC contents, GC bins, and lengths distrib.
"""
with open_for_parsing(bg_file) as stream:
gc_bins = []
gc_list = []
lengths = []
dinuc = [0] * len(IUPAC) * len(IUPAC)
for _ in range(0, 101):
gc_bins.append([])
for record in SeqIO.parse(stream, "fasta"):
gc = GC(record.seq)
gc_list.append(gc)
gc = int(round(gc)) # python3 fix
gc_bins[gc].append(record)
lengths.append(len(record.seq))
dinuc = [x + y for x, y in zip(dinuc, dinuc_count(record.seq))]
print_in_bg_dir(gc_bins, bg_dir)
return gc_list, gc_bins, lengths, dinuc
def fg_GC_bins(fg, winlen, step):
"""
Get %GC info for foreground sequences.
Compute G+C content for all sequences in the foreground and store the
information in a list. To each G+C percentage bin, we associate the number
of sequences falling in the corresponding bin
Return the corresponding lists.
"""
with open_for_parsing(fg) as stream:
tmp_gc_bins = []
gc_list = []
lengths = []
dinuc = [0] * len(IUPAC) * len(IUPAC)
for _ in range(0, 101):
tmp_gc_bins.append([])
for record in SeqIO.parse(stream, "fasta"):
gc, min_gc, max_gc, sd_gc, cv_gc = GC_info(record.seq, winlen,
step)
gc_list.append(gc)
# python 3 fix
gc = int(round(gc))
tmp_gc_bins[gc].append((min_gc, max_gc, sd_gc, cv_gc))
lengths.append(len(record.seq))
dinuc = [x + y for x, y in zip(dinuc, dinuc_count(record.seq))]
return gc_list, avg_and_sd_gc_info(tmp_gc_bins), lengths, dinuc
def bg_len_GC_bins(bg, bg_dir):
"""
Get lengths info for background sequences.
Compute G+C content for all sequences in the background and store the
information in a list. To each G+C percentage bin, we associate the
corresponding sequence names with information about GC composition within
sliding windows.
Return info in lists.
"""
with open_for_parsing(bg) as stream:
gc_bins = []
gc_list = []
lengths = []
dinuc = [0] * len(IUPAC) * len(IUPAC)
for _ in range(0, 101):
gc_bins.append({})
for record in SeqIO.parse(stream, "fasta"):
gc = GC(record.seq)
gc_list.append(gc)
if len(record) in gc_bins[gc]:
gc_bins[gc][len(record)].append(record)
else:
gc_bins[gc][len(record)] = [record]
lengths.append(len(record.seq))
dinuc = [x + y for x, y in zip(dinuc, dinuc_count(record.seq))]
print_in_bg_dir(gc_bins, bg_dir, True)
return gc_list, gc_bins, lengths, dinuc
def fg_len_GC_bins(fg, winlen, step):
"""
Get needed lengths info for foreground sequences.
Compute G+C content for all sequences in the foreground and store the
information in a list. To each G+C percentage bin, we associate the number
of sequences falling in the corresponding bin.
Return the corresponding info in lists.
"""
with open_for_parsing(fg) as stream:
tmp_gc_bins = []
gc_list = []
lengths = []
dinuc = [0] * len(IUPAC) * len(IUPAC)
l_dic = []
for _ in range(0, 101):
tmp_gc_bins.append([])
l_dic.append({})
for record in SeqIO.parse(stream, "fasta"):
gc, min_gc, max_gc, sd_gc, cv_gc = GC_info(record.seq, winlen,
step)
gc_list.append(gc)
tmp_gc_bins[gc].append((min_gc, max_gc, sd_gc, cv_gc))
length = len(record)
if length in l_dic[gc]:
l_dic[gc][length] += 1
else:
l_dic[gc][length] = 1
lengths.append(length)
dinuc = [x + y for x, y in zip(dinuc, dinuc_count(record.seq))]
return gc_list, avg_and_sd_len_gc_info(l_dic, tmp_gc_bins), lengths, dinuc
def get_bins_from_bg_dir(bg_dir, percent):
""" Return the sequences from the corresponding bin file. """
filename = "{0}/bg_bin_{1:d}.txt".format(bg_dir, percent)
with open_for_parsing(filename) as stream:
bin_seq = []
for record in SeqIO.parse(stream, "fasta"):
bin_seq.append(record)
return bin_seq
def get_bins_len_from_bg_dir(bg_dir, percent):
""" Return the sequences from the corresponding bin file. """
filename = "{0}/bg_bin_{1:d}.txt".format(bg_dir, percent)
with open_for_parsing(filename) as stream:
bin_seq = {}
for record in SeqIO.parse(stream, "fasta"):
length = len(record)
if length in bin_seq:
bin_seq[length].append(record)
else:
bin_seq[length] = [record]
return bin_seq
def fg_GC_bins(fg_file):
"""
Compute G+C content for all sequences in the foreground.
It computes the %GC content and store the information in a list. To each
G+C percentage bin, we associate the number of sequences falling in the
corresponding bin.
Return lists of GC contents, GC bins, lengths distrib, and dinuc compo.
"""
with open_for_parsing(fg_file) as stream:
gc_bins = [0] * 101
gc_list = []
lengths = []
dinuc = [0] * len(IUPAC) * len(IUPAC)
for record in SeqIO.parse(stream, "fasta"):
gc = GC(record.seq)
gc_list.append(gc)
# python 3 fix
gc = int(round(gc))
gc_bins[gc] += 1
lengths.append(len(record.seq))
dinuc = [x + y for x, y in zip(dinuc, dinuc_count(record.seq))]
return gc_list, gc_bins, lengths, dinuc