def save_genbank_m4(gb_m3, gb_m4):
"""Generate modified-IV GenBank file (reduced version of the genome).
Args:
gb_m3 (Bio.SeqRecord.SeqRecord): modified-III GenBank annotation.
gb_m4 (str): modified-IV GenBank file path.
"""
deletions = []
for feature in gb_m3.features:
if is_deletion(feature):
deletions.append(feature.location)
deletions = CompoundLocation(deletions)
non_deletions = complementary_compoundloc(0, len(gb_m3), deletions)
reduced_annot = SeqRecord(seq=non_deletions.extract(gb_m3.seq),
id=gb_m3.id,
name=gb_m3.name,
description=gb_m3.description,
dbxrefs=gb_m3.dbxrefs,
annotations=gb_m3.annotations)
# Shift features' positions according to deletions
end = 0
for nondel in non_deletions.parts:
offset = nondel.start - end
for feature in gb_m3.features:
if (feature.location.start in nondel
and feature.location.end in nondel):
feature.location = feature.location + (-offset)
reduced_annot.features.append(feature)
end = nondel.end - offset + 2
SeqIO.write(reduced_annot, gb_m4, "genbank")
def best_deletion_order(gb_m3, ori, ter, log):
"""Calculate optimal deletion order for minimising replichore
imbalance at each step.
Args:
gb_m3 (Bio.SeqRecord.SeqRecord): modified-III GenBank annotation.
ori (int): position of origin of replication.
ter (int): position of terminus of replication.
log (str): report file path.
"""
# Replichore imbalance = len(genome)/2 - len(0-180º replichore)
if ter > ori:
len_1 = ter - ori
else:
len_1 = len(gb_m3) - (ori - ter)
imbalance = len(gb_m3) / 2 - len_1
# Calculate each deletion's contribution to the imbalance
deletions = []
for feature in gb_m3.features:
if is_deletion(feature):
name = feature.qualifiers["note"][0].split()[-1]
midpoint = (feature.location.end + feature.location.start) / 2
if midpoint > ori or midpoint < ter: # 0-180º replichore
contrib = len(feature)
else: # 180-360º replichore
contrib = -len(feature)
deletions.append((name, contrib))
# Best next deletion is the one that minimises the imbalance
with open(log, "a") as f:
f.write("Initial replichore imbalance: %d\n" % imbalance)
f.write("Best deletion order for minimising imbalance at each step:\n")
n = 0
while deletions:
n += 1
best = deletions.pop(
np.argmin(
[abs(imbalance + deletion[1]) for deletion in deletions]))
imbalance += best[1]
f.write(" %d. %s -> imbalance = %d\n" % (n, best[0], imbalance))
f.write(sep)
def get_stats(gb_m3):
"""Calculate essentiality prediction and deletion statistics.
Args:
gb_m3 (Bio.SeqRecord.SeqRecord): modified-III GenBank annotation.
Returns:
stats (dict): results.
"""
l = 0
e = 0
deletions = []
for feature in gb_m3.features:
if is_deletion(feature):
deletions.append(feature.location)
if not (l and e):
l = int(feature.qualifiers["note"][1].split("=")[-1])
e = float(feature.qualifiers["note"][2].split("=")[-1])
deletions = CompoundLocation(deletions)
stats = {
"org": gb_m3.annotations["organism"],
"size": len(gb_m3),
"l": l,
"e": e,
"del_n": 0,
"del_kb": 0,
"del_perc": 0,
"del_genes_pseudo": 0,
"del_genes_hypot": 0,
"del_genes_nonhypot": 0,
"ess_genes_rna": 0,
"ess_genes_hypot": 0,
"ess_genes_nonhypot": 0,
"ness_genes_pseudo": 0,
"ness_genes_hypot": 0,
"ness_genes_nonhypot": 0
}
for feature in gb_m3.features:
if is_essential(feature, e): # essential
if feature.type in ("tRNA", "rRNA", "tmRNA", "ncRNA"):
stats["ess_genes_rna"] += 1
elif feature.type == "CDS" and "locus_tag" in feature.qualifiers:
if ("product" in feature.qualifiers and "hypothetical"
in feature.qualifiers["product"][0]):
stats["ess_genes_hypot"] += 1
else:
stats["ess_genes_nonhypot"] += 1
elif feature.type == "CDS" and "locus_tag" in feature.qualifiers: # non-essential
if "pseudo" in feature.qualifiers:
stats["ness_genes_pseudo"] += 1
elif ("product" in feature.qualifiers
and "hypothetical" in feature.qualifiers["product"][0]):
stats["ness_genes_hypot"] += 1
else:
stats["ness_genes_nonhypot"] += 1
if (feature.location.start in deletions
or feature.location.end in deletions): # in a deletion
if "pseudo" in feature.qualifiers:
stats["del_genes_pseudo"] += 1
elif ("product" in feature.qualifiers
and "hypothetical" in feature.qualifiers["product"][0]):
stats["del_genes_hypot"] += 1
else:
stats["del_genes_nonhypot"] += 1
elif is_deletion(feature):
stats["del_n"] += 1
stats["del_kb"] += len(feature) / 1000
stats["del_perc"] += len(feature) / len(gb_m3) * 100
return stats
os.makedirs(OUT_DIR, exist_ok=True)
genbank_id = os.path.splitext(os.path.basename(GENBANK_M3))[0]
GENBANK_M4 = os.path.join(OUT_DIR, genbank_id + ".gbm4")
OUT_IMG = os.path.join(OUT_DIR, "genome_reduction")
REPORT_LOG = os.path.join(OUT_DIR, "reduction_stats.txt")
ORI = args.ORI
TER = args.TER
# Check input
try:
genbank_m3 = SeqIO.read(GENBANK_M3, "genbank")
except (FileNotFoundError, ValueError):
raise SystemExit("\n\terror: could not read .gbm3 file\n")
contains_dels = False
for feature in genbank_m3.features:
if is_deletion(feature):
contains_dels = True
break
if not contains_dels:
raise SystemExit("\n\terror: invalid GenBank file (must be .gbm3)\n")
if ORI and TER:
if not all(
[coord in range(1,
len(genbank_m3) + 1) for coord in (ORI, TER)]):
raise SystemExit("\n\terror: invalid ori/ter coordinates\n")
# Get ori + ter coordinates
if not (ORI and TER):
ORI, TER = find_ori_ter(genbank_m3)
# Draw circular genome plot