def merge(args):
"""
%prog merge folder1 ...
Consolidate split contents in the folders. The folders can be generated by
the split() process and several samples may be in separate fastq files. This
program merges them.
"""
p = OptionParser(merge.__doc__)
p.add_option("--outdir",
default="outdir",
help="Output final reads in [default: %default]")
opts, args = p.parse_args(args)
if len(args) < 1:
sys.exit(not p.print_help())
folders = args
outdir = opts.outdir
mkdir(outdir)
files = flatten(glob("{0}/*.*.fastq".format(x)) for x in folders)
files = list(files)
key = lambda x: op.basename(x).split(".")[0]
files.sort(key=key)
for id, fns in groupby(files, key=key):
fns = list(fns)
outfile = op.join(outdir, "{0}.fastq".format(id))
FileMerger(fns, outfile=outfile).merge(checkexists=True)
def merge(args):
"""
%prog merge folder1 ...
Consolidate split contents in the folders. The folders can be generated by
the split() process and several samples may be in separate fastq files. This
program merges them.
"""
p = OptionParser(merge.__doc__)
p.set_outdir(outdir="outdir")
opts, args = p.parse_args(args)
if len(args) < 1:
sys.exit(not p.print_help())
folders = args
outdir = opts.outdir
mkdir(outdir)
files = flatten(glob("{0}/*.*.fastq".format(x)) for x in folders)
files = list(files)
key = lambda x: op.basename(x).split(".")[0]
files.sort(key=key)
for id, fns in groupby(files, key=key):
fns = list(fns)
outfile = op.join(outdir, "{0}.fastq".format(id))
FileMerger(fns, outfile=outfile).merge(checkexists=True)
def notify(args):
"""
%prog notify "Message to be sent"
Send a message via email/push notification.
Email notify: Recipient email address is constructed by joining the login `username`
and `dnsdomainname` of the server
Push notify: Uses available API
"""
from jcvi.utils.iter import flatten
valid_notif_methods.extend(available_push_api.keys())
fromaddr, toaddr = get_email_address()
p = OptionParser(notify.__doc__)
p.add_option("--method",
default="email",
choices=valid_notif_methods,
help="Specify the mode of notification [default: %default]")
p.add_option("--subject",
default="JCVI: job monitor",
help="Specify the subject of the notification message")
p.set_email()
g1 = OptionGroup(p, "Optional `push` parameters")
g1.add_option("--api", default="pushover", \
choices=list(flatten(available_push_api.values())),
help="Specify API used to send the push notification")
g1.add_option("--priority",
default=0,
type="int",
help="Message priority (-1 <= p <= 2) [default: %default]")
g1.add_option("--timestamp", default=None, type="int", \
dest="timestamp", \
help="Message timestamp in unix format [default: %default]")
p.add_option_group(g1)
opts, args = p.parse_args(args)
if len(args) == 0:
logging.error("Please provide a brief message to be sent")
sys.exit(not p.print_help())
subject = opts.subject
message = " ".join(args).strip()
if opts.method == "email":
if not is_valid_email(opts.toaddr):
logging.debug("Email address `{0}` is not valid!".format(
opts.toaddr))
sys.exit()
toaddr = [opts.toaddr] # TO address should be in a list
send_email(fromaddr, toaddr, subject, message)
else:
pushnotify(subject, message, api=opts.api, priority=opts.priority, \
timestamp=opts.timestamp)
def notify(args):
"""
%prog notify "Message to be sent"
Send a message via email/push notification.
Email notify: Recipient email address is constructed by joining the login `username`
and `dnsdomainname` of the server
Push notify: Uses available API
"""
from jcvi.utils.iter import flatten
debug()
valid_notif_methods.extend(available_push_api.keys())
fromaddr, toaddr = get_email_address()
p = OptionParser(notify.__doc__)
p.add_option("--method", default="email", choices=valid_notif_methods,
help="Specify the mode of notification [default: %default]")
p.add_option("--subject", default="JCVI: job monitor",
help="Specify the subject of the notification message")
p.set_email()
g1 = OptionGroup(p, "Optional `push` parameters")
g1.add_option("--api", default="pushover", \
choices=list(flatten(available_push_api.values())),
help="Specify API used to send the push notification" + \
" [default: %default]")
g1.add_option("--priority", default=0, type="int",
help="Message priority (-1 <= p <= 2) [default: %default]")
g1.add_option("--timestamp", default=None, type="int", \
dest="timestamp", \
help="Message timestamp in unix format [default: %default]")
p.add_option_group(g1)
opts, args = p.parse_args(args)
if len(args) == 0:
logging.error("Please provide a brief message to be sent")
sys.exit(not p.print_help())
subject = opts.subject
message = " ".join(args).strip()
if opts.method == "email":
if not is_valid_email(opts.toaddr):
logging.debug("Email address `{0}` is not valid!".format(opts.toaddr))
sys.exit()
toaddr = [opts.toaddr] # TO address should be in a list
send_email(fromaddr, toaddr, subject, message)
else:
pushnotify(subject, message, api=opts.api, priority=opts.priority, \
timestamp=opts.timestamp)
def join(args):
"""
%prog join file1.txt file2.txt ..
Join tabular files based on common column. --column specifies the column
index to pivot on. Use comma to separate multiple values if the pivot column
is different in each file. Maintain the order in the first file.
"""
from jcvi.utils.iter import flatten
p = OptionParser(join.__doc__)
p.add_option("--column", default="0",
help="0-based column id, multiple values allowed [default: %default]")
p.add_option("--noheader", default=False, action="store_true",
help="Do not print header [default: %default]")
set_outfile(p)
opts, args = p.parse_args(args)
nargs = len(args)
if len(args) < 2:
sys.exit(not p.print_help())
c = opts.column
if "," in c:
cc = [int(x) for x in c.split(",")]
else:
cc = [int(c)] * nargs
assert len(cc) == nargs, "Column index number != File number"
# Maintain the first file line order, and combine other files into it
pivotfile = args[0]
files = [DictFile(f, keypos=c, valuepos=None, delimiter="\t") \
for f, c in zip(args, cc)]
otherfiles = files[1:]
header = "\t".join(flatten([op.basename(x.filename)] * x.ncols \
for x in files))
fp = open(pivotfile)
fw = must_open(opts.outfile, "w")
if not opts.noheader:
print >> fw, header
for row in fp:
row = row.rstrip()
atoms = row.split("\t")
newrow = atoms
key = atoms[cc[0]]
for d in otherfiles:
drow = d.get(key, ["na"] * d.ncols)
newrow += drow
print >> fw, "\t".join(newrow)
def join(args):
"""
%prog join file1.txt file2.txt ..
Join tabular files based on common column. --column specifies the column
index to pivot on. Use comma to separate multiple values if the pivot column
is different in each file. Maintain the order in the first file.
"""
from jcvi.utils.iter import flatten
p = OptionParser(join.__doc__)
p.add_option("--column", default="0",
help="0-based column id, multiple values allowed [default: %default]")
p.add_option("--noheader", default=False, action="store_true",
help="Do not print header [default: %default]")
set_outfile(p)
opts, args = p.parse_args(args)
nargs = len(args)
if len(args) < 2:
sys.exit(not p.print_help())
c = opts.column
if "," in c:
cc = [int(x) for x in c.split(",")]
else:
cc = [int(c)] * nargs
assert len(cc) == nargs, "Column index number != File number"
# Maintain the first file line order, and combine other files into it
pivotfile = args[0]
files = [DictFile(f, keypos=c, valuepos=None, delimiter="\t") \
for f, c in zip(args, cc)]
otherfiles = files[1:]
header = "\t".join(flatten([op.basename(x.filename)] * x.ncols \
for x in files))
fp = open(pivotfile)
fw = must_open(opts.outfile, "w")
if not opts.noheader:
print >> fw, header
for row in fp:
row = row.rstrip()
atoms = row.split("\t")
newrow = atoms
key = atoms[cc[0]]
for d in otherfiles:
drow = d.get(key, ["na"] * d.ncols)
newrow += drow
print >> fw, "\t".join(newrow)
def fix_orientation(self, tour):
"""
Test each scaffold if flipping will increass longest monotonic chain
length.
"""
orientations = dict(tour) # old configuration here
scaffold_oo = defaultdict(list)
scaffolds, oos = zip(*tour)
for mlg in self.linkage_groups:
lg = mlg.lg
mapname = mlg.mapname
for s, o in tour:
i = scaffolds.index(s)
L = [self.get_series(lg, x, xo) for x, xo in tour[:i]]
U = [self.get_series(lg, x, xo) for x, xo in tour[i + 1:]]
L, U = list(flatten(L)), list(flatten(U))
M = self.get_series(lg, s)
plus = lms(L + M + U)
minus = lms(L + M[::-1] + U)
d = plus[0] - minus[0]
if not d:
continue
scaffold_oo[s].append((d, mapname)) # reset orientation
fixed = 0
for s, v in scaffold_oo.items():
d = self.weighted_mean(v)
old_d = orientations[s]
new_d = np.sign(d)
if new_d != old_d:
orientations[s] = new_d
fixed += 1
tour = [(x, orientations[x]) for x in scaffolds]
logging.debug("Fixed orientations for {0} scaffolds.".format(fixed))
return tour
def fix_orientation(self, tour):
"""
Test each scaffold if flipping will increass longest monotonic chain
length.
"""
orientations = dict(tour) # old configuration here
scaffold_oo = defaultdict(list)
scaffolds, oos = zip(*tour)
for mlg in self.linkage_groups:
lg = mlg.lg
mapname = mlg.mapname
for s, o in tour:
i = scaffolds.index(s)
L = [self.get_series(lg, x, xo) for x, xo in tour[:i]]
U = [self.get_series(lg, x, xo) for x, xo in tour[i + 1:]]
L, U = list(flatten(L)), list(flatten(U))
M = self.get_series(lg, s)
plus = lms(L + M + U)
minus = lms(L + M[::-1] + U)
d = plus[0] - minus[0]
if not d:
continue
scaffold_oo[s].append((d, mapname)) # reset orientation
fixed = 0
for s, v in scaffold_oo.items():
d = self.weighted_mean(v)
old_d = orientations[s]
new_d = np.sign(d)
if new_d != old_d:
orientations[s] = new_d
fixed += 1
tour = [(x, orientations[x]) for x in scaffolds]
logging.debug("Fixed orientations for {0} scaffolds.".format(fixed))
return tour
def make_attributes(s, gff3=True):
"""
In GFF3, the last column is typically:
ID=cds00002;Parent=mRNA00002;
In GFF2, the last column is typically:
Gene 22240.t000374; Note "Carbonic anhydrase"
"""
if gff3:
d = parse_qs(s)
else:
attributes = s.split("; ")
d = DefaultOrderedDict(list)
for a in attributes:
key, val = a.strip().split(' ', 1)
val = val.replace('"', '')
d[key].append(val)
for key, val in d.items():
d[key] = list(flatten([v.split(",") for v in val]))
return d
def make_attributes(s, gff3=True):
"""
In GFF3, the last column is typically:
ID=cds00002;Parent=mRNA00002;
In GFF2, the last column is typically:
Gene 22240.t000374; Note "Carbonic anhydrase"
"""
if gff3:
d = parse_qs(s)
else:
attributes = s.split("; ")
d = DefaultOrderedDict(list)
for a in attributes:
key, val = a.strip().split(' ', 1)
val = val.replace('"', '')
d[key].append(val)
for key, val in d.items():
d[key] = list(flatten([v.split(",") for v in val]))
return d
def waitpid(args):
"""
%prog waitpid PID ::: "./command_to_run param1 param2 ...."
Given a PID, this script will wait for the PID to finish running and
then perform a desired action (notify user and/or execute a new command)
Specify "--notify=METHOD` to send the user a notification after waiting for PID
Specify `--grid` option to send the new process to the grid after waiting for PID
"""
import shlex
from jcvi.utils.iter import flatten
valid_notif_methods.extend(list(flatten(available_push_api.values())))
p = OptionParser(waitpid.__doc__)
p.add_option("--notify",
default="email",
choices=valid_notif_methods,
help="Specify type of notification to be sent after waiting")
p.add_option("--interval",
default=120,
type="int",
help="Specify PID polling interval in seconds")
p.add_option("--message",
help="Specify notification message [default: %default]")
p.set_email()
p.set_grid()
opts, args = p.parse_args(args)
if len(args) == 0:
sys.exit(not p.print_help())
if not opts.message:
"""
If notification message not specified by user, just get
the name of the running command and use it as the message
"""
from subprocess import check_output
sep = ":::"
cmd = None
if sep in args:
sepidx = args.index(sep)
cmd = " ".join(args[sepidx + 1:]).strip()
args = args[:sepidx]
pid = int(" ".join(args).strip())
status = pid_exists(pid)
if status:
if opts.message:
msg = opts.message
else:
get_origcmd = "ps -p {0} -o cmd h".format(pid)
msg = check_output(shlex.split(get_origcmd)).strip()
_waitpid(pid, interval=opts.interval)
else:
logging.debug("Process with PID {0} does not exist".format(pid))
sys.exit()
if opts.notify:
notifycmd = ["[{0}] `{1}`".format(gethostname(), msg)]
if opts.notify != "email":
notifycmd.append("--method={0}".format("push"))
notifycmd.append("--api={0}".format(opts.notify))
else:
notifycmd.append('--email={0}'.format(opts.email))
notify(notifycmd)
if cmd is not None:
bg = False if opts.grid else True
sh(cmd, grid=opts.grid, background=bg)
def join(args):
"""
%prog join file1.txt(pivotfile) file2.txt ..
Join tabular-like files based on common column.
--column specifies the column index to pivot on.
Use comma to separate multiple values if the pivot column is different
in each file. Maintain the order in the first file.
--sep specifies the column separators, default to tab.
Use comma to separate multiple values if the column separator is different
in each file.
"""
from jcvi.utils.iter import flatten
p = OptionParser(join.__doc__)
p.add_option("--column", default="0",
help="0-based column id, multiple values allowed [default: %default]")
p.set_sep(multiple=True)
p.add_option("--noheader", default=False, action="store_true",
help="Do not print header [default: %default]")
p.add_option("--na", default="na",
help="Value for unjoined data [default: %default]")
p.add_option("--keysep", default=",",
help="specify separator joining multiple elements in the key column"
+ " of the pivot file [default: %default]")
p.set_outfile()
opts, args = p.parse_args(args)
nargs = len(args)
keysep = opts.keysep
if len(args) < 2:
sys.exit(not p.print_help())
na = opts.na
c = opts.column
if "," in c:
cc = [int(x) for x in c.split(",")]
else:
cc = [int(c)] * nargs
assert len(cc) == nargs, "Column index number != File number"
s = opts.sep
if "," in s:
ss = [x for x in s.split(",")]
else:
ss = [s] * nargs
assert len(ss) == nargs, "column separator number != File number"
# Maintain the first file line order, and combine other files into it
pivotfile = args[0]
files = [DictFile(f, keypos=c, valuepos=None, delimiter=s) \
for f, c, s in zip(args, cc, ss)]
otherfiles = files[1:]
header = "\t".join(flatten([op.basename(x.filename)] * x.ncols \
for x in files))
fp = must_open(pivotfile)
fw = must_open(opts.outfile, "w")
if not opts.noheader:
print >> fw, header
for row in fp:
row = row.rstrip()
atoms = row.split(ss[0])
newrow = atoms
key = atoms[cc[0]]
keys = key.split(keysep) if keysep in key else [key]
for d in otherfiles:
drows = list()
for key in keys:
drows.append(d.get(key, [na] * d.ncols))
drow = [keysep.join(x) for x in list(zip(*drows))]
newrow += drow
print >> fw, "\t".join(newrow)
def waitpid(args):
"""
%prog waitpid PID ::: "./command_to_run param1 param2 ...."
Given a PID, this script will wait for the PID to finish running and
then perform a desired action (notify user and/or execute a new command)
Specify "--notify=METHOD` to send the user a notification after waiting for PID
Specify `--grid` option to send the new process to the grid after waiting for PID
"""
import shlex
from time import sleep
from jcvi.utils.iter import flatten
valid_notif_methods.extend(list(flatten(available_push_api.values())))
p = OptionParser(waitpid.__doc__)
p.add_option("--notify", default=None, choices=valid_notif_methods,
help="Specify type of notification to be sent after waiting")
p.add_option("--interval", default=120, type="int",
help="Specify PID polling interval in seconds")
p.add_option("--message",
help="Specify notification message [default: %default]")
p.set_email()
p.set_grid()
opts, args = p.parse_args(args)
if len(args) == 0:
sys.exit(not p.print_help())
if not opts.message:
"""
If notification message not specified by user, just get
the name of the running command and use it as the message
"""
from subprocess import check_output
sep = ":::"
cmd = None
if sep in args:
sepidx = args.index(sep)
cmd = " ".join(args[sepidx + 1:]).strip()
args = args[:sepidx]
pid = int(" ".join(args).strip())
status = is_running(pid)
if status:
if opts.message:
msg = opts.message
else:
get_origcmd = "ps -p {0} -o cmd h".format(pid)
msg = check_output(shlex.split(get_origcmd)).strip()
while is_running(pid):
sleep(opts.interval)
else:
logging.debug("Process with PID {0} does not exist".format(pid))
sys.exit()
if opts.notify:
notifycmd = ["[completed] {0}: `{1}`".format(gethostname(), msg)]
if opts.notify != "email":
notifycmd.append("--method={0}".format("push"))
notifycmd.append("--api={0}".format(opts.notify))
else:
notifycmd.append('--email="{0}"'.format(opts.email))
notify(notifycmd)
if cmd is not None:
bg = False if opts.grid else True
sh(cmd, grid=opts.grid, background=bg)
def mask(args):
"""
%prog mask agpfile bedfile
Mask given ranges in componets to gaps.
"""
p = OptionParser(mask.__doc__)
p.add_option("--split", default=False, action="store_true",
help="Split object and create new names [default: %default]")
p.add_option("--log", default=False, action="store_true",
help="Write verbose logs to .masklog file [default: %default]")
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(p.print_help())
agpfile, bedfile = args
agp = AGP(agpfile)
bed = Bed(bedfile)
simple_agp = agp.order
# agp lines to replace original ones, keyed by the component
agp_fixes = defaultdict(list)
newagpfile = agpfile.replace(".agp", ".masked.agp")
logfile = bedfile.replace(".bed", ".masklog")
fw = open(newagpfile, "w")
if opts.log:
fwlog = open(logfile, "w")
for component, intervals in bed.sub_beds():
if opts.log:
print >> fwlog, "\n".join(str(x) for x in intervals)
i, a = simple_agp[component]
object = a.object
component_span = a.component_span
orientation = a.orientation
if opts.log:
print >> fwlog, a
assert a.component_beg, a.component_end
arange = a.component_beg, a.component_end
# Make sure `ivs` contain DISJOINT ranges, and located within `arange`
ivs = []
for i in intervals:
iv = range_intersect(arange, (i.start, i.end))
if iv is not None:
ivs.append(iv)
# Sort the ends of `ivs` as well as the arange
arange = a.component_beg - 1, a.component_end + 1
endpoints = sorted(flatten(ivs + [arange]))
# reverse if component on negative strand
if orientation == '-':
endpoints.reverse()
sum_of_spans = 0
# assign complements as sequence components
for i, (a, b) in enumerate(pairwise(endpoints)):
if orientation == '-':
a, b = b, a
if orientation not in ('+', '-'):
orientation = '+'
oid = object + "_{0}".format(i / 2) if opts.split else object
aline = [oid, 0, 0, 0]
if i % 2 == 0:
cspan = b - a - 1
aline += ['D', component, a + 1, b - 1, orientation]
is_gap = False
else:
cspan = b - a + 1
aline += ["N", cspan, "fragment", "yes"]
is_gap = True
if cspan <= 0:
continue
sum_of_spans += cspan
aline = "\t".join(str(x) for x in aline)
if not (opts.split and is_gap):
agp_fixes[component].append(aline)
if opts.log:
print >> fwlog, aline
assert component_span == sum_of_spans
if opts.log:
print >> fwlog
# Finally write the masked agp
for a in agp:
if not a.is_gap and a.component_id in agp_fixes:
print >> fw, "\n".join(agp_fixes[a.component_id])
else:
print >> fw, a
fw.close()
# Reindex
idxagpfile = reindex([newagpfile])
shutil.move(idxagpfile, newagpfile)
return newagpfile
def join(args):
"""
%prog join file1.txt(pivotfile) file2.txt ..
Join tabular-like files based on common column.
--column specifies the column index to pivot on.
Use comma to separate multiple values if the pivot column is different
in each file. Maintain the order in the first file.
--sep specifies the column separators, default to tab.
Use comma to separate multiple values if the column separator is different
in each file.
"""
from jcvi.utils.iter import flatten
p = OptionParser(join.__doc__)
p.add_option(
"--column",
default="0",
help="0-based column id, multiple values allowed [default: %default]")
p.set_sep(multiple=True)
p.add_option("--noheader",
default=False,
action="store_true",
help="Do not print header [default: %default]")
p.add_option("--na",
default="na",
help="Value for unjoined data [default: %default]")
p.add_option(
"--keysep",
default=",",
help="specify separator joining multiple elements in the key column" +
" of the pivot file [default: %default]")
p.set_outfile()
opts, args = p.parse_args(args)
nargs = len(args)
keysep = opts.keysep
if len(args) < 2:
sys.exit(not p.print_help())
na = opts.na
c = opts.column
if "," in c:
cc = [int(x) for x in c.split(",")]
else:
cc = [int(c)] * nargs
assert len(cc) == nargs, "Column index number != File number"
s = opts.sep
if "," in s:
ss = [x for x in s.split(",")]
else:
ss = [s] * nargs
assert len(ss) == nargs, "column separator number != File number"
# Maintain the first file line order, and combine other files into it
pivotfile = args[0]
files = [DictFile(f, keypos=c, valuepos=None, delimiter=s) \
for f, c, s in zip(args, cc, ss)]
otherfiles = files[1:]
header = "\t".join(flatten([op.basename(x.filename)] * x.ncols \
for x in files))
fp = must_open(pivotfile)
fw = must_open(opts.outfile, "w")
if not opts.noheader:
print >> fw, header
for row in fp:
row = row.rstrip()
atoms = row.split(ss[0])
newrow = atoms
key = atoms[cc[0]]
keys = key.split(keysep) if keysep in key else [key]
for d in otherfiles:
drows = list()
for key in keys:
drows.append(d.get(key, [na] * d.ncols))
drow = [keysep.join(x) for x in list(zip(*drows))]
newrow += drow
print >> fw, "\t".join(newrow)
def mask(args):
"""
%prog mask agpfile bedfile
Mask given ranges in components to gaps.
"""
p = OptionParser(mask.__doc__)
p.add_option("--split",
default=False,
action="store_true",
help="Split object and create new names [default: %default]")
p.add_option(
"--log",
default=False,
action="store_true",
help="Write verbose logs to .masklog file [default: %default]")
opts, args = p.parse_args(args)
if len(args) != 2:
sys.exit(p.print_help())
agpfile, bedfile = args
agp = AGP(agpfile)
bed = Bed(bedfile)
simple_agp = agp.order
# agp lines to replace original ones, keyed by the component
agp_fixes = defaultdict(list)
newagpfile = agpfile.replace(".agp", ".masked.agp")
logfile = bedfile.replace(".bed", ".masklog")
fw = open(newagpfile, "w")
if opts.log:
fwlog = open(logfile, "w")
for component, intervals in bed.sub_beds():
if opts.log:
print >> fwlog, "\n".join(str(x) for x in intervals)
i, a = simple_agp[component]
object = a.object
component_span = a.component_span
orientation = a.orientation
if opts.log:
print >> fwlog, a
assert a.component_beg, a.component_end
arange = a.component_beg, a.component_end
# Make sure `ivs` contain DISJOINT ranges, and located within `arange`
ivs = []
for i in intervals:
iv = range_intersect(arange, (i.start, i.end))
if iv is not None:
ivs.append(iv)
# Sort the ends of `ivs` as well as the arange
arange = a.component_beg - 1, a.component_end + 1
endpoints = sorted(flatten(ivs + [arange]))
# reverse if component on negative strand
if orientation == '-':
endpoints.reverse()
sum_of_spans = 0
# assign complements as sequence components
for i, (a, b) in enumerate(pairwise(endpoints)):
if orientation == '-':
a, b = b, a
if orientation not in ('+', '-'):
orientation = '+'
oid = object + "_{0}".format(i / 2) if opts.split else object
aline = [oid, 0, 0, 0]
if i % 2 == 0:
cspan = b - a - 1
aline += ['D', component, a + 1, b - 1, orientation]
is_gap = False
else:
cspan = b - a + 1
aline += ["N", cspan, "fragment", "yes"]
is_gap = True
if cspan <= 0:
continue
sum_of_spans += cspan
aline = "\t".join(str(x) for x in aline)
if not (opts.split and is_gap):
agp_fixes[component].append(aline)
if opts.log:
print >> fwlog, aline
#assert component_span == sum_of_spans
if opts.log:
print >> fwlog
# Finally write the masked agp
for a in agp:
if not a.is_gap and a.component_id in agp_fixes:
print >> fw, "\n".join(agp_fixes[a.component_id])
else:
print >> fw, a
fw.close()
# Reindex
idxagpfile = reindex([newagpfile])
shutil.move(idxagpfile, newagpfile)
return newagpfile