def mauve_pw_align(ref, query, dirs):
"""Set up and perform a pairwise alignment with Mauve."""
# set outputs
mauve_outfile = dirs['mauve']+ref.name+"_"+query.name+".mauve"
segfile = dirs['aln_segs']+ref.name+"_"+query.name+"_segs.txt"
# check for existing alignment
if path.exists(segfile):
print "already done"
else:
# prep segments file
open(segfile, 'w').write('')
# purge any pre-existing sslist files
sslist_files = from_dir(dirs['seqfiles'], re.compile(r'.*\.sslist.*'))
for sslist in sslist_files:
try: os.remove(dirs['seqfiles']+sslist)
except Exception: raise
# do Mauve alignment
file_list = [ref.gbk, query.gbk]
align_mauve(file_list, mauve_outfile)
try:
# parse Mauve output (without initial clumping)
coords = mauver_load2_k0(mauve_outfile+".backbone", 0)
print "\nSegment results:", len(coords), '->',
# chop segments that are too long
chop_array = chop_rows(coords, max_size, chop_mode)
print len(chop_array), 'segments <', max_size, 'bp'
# make detailed pairwise alignments of the segments
print "Aligning segments ..."
ref_rec = load_genbank(ref.gbk)
query_rec = load_genbank(query.gbk)
id = iter_align(chop_array, ref_rec, query_rec,
dirs['aln_segs'], segfile)
print "Results:", id, "% id. overall"
except IOError:
print "\nERROR: Mauve alignment failed"
def mauve_pw_align(ref, query, r_root_dir, g_root_dir, dirs, run, max_size,
chop_mode, mauve_exec, mtype):
"""Set up and perform a pairwise alignment with Mauve."""
aln_dir = r_root_dir + run + dirs['aln_segs']
mauve_dir = r_root_dir + run + dirs['mauve']
# set outputs
mauve_outfile = mauve_dir + ref.name + "_" + query.name + ".mauve"
segfile = aln_dir + ref.name + "_" + query.name + "_segs.txt"
# check for existing alignment
if path.exists(segfile):
print "already done"
else:
# prep segments file
open(segfile, 'w').write('')
# purge any pre-existing sslist files
sslist_files = from_dir(g_root_dir, re.compile(r'.*\.sslist.*'))
for sslist in sslist_files:
try:
os.remove(g_root_dir + sslist)
except Exception:
raise
# do Mauve alignment
file_list = [ref.gbk, query.gbk]
align_mauve(file_list, mauve_outfile, mauve_exec)
try:
# parse Mauve output (without initial clumping)
coords = mauver_load2_k0(mauve_outfile + ".backbone", 0, mtype)
print "\nSegment results:", len(coords), '->',
# chop segments that are too long
chop_array = chop_rows(coords, max_size, chop_mode, mtype)
print len(chop_array), 'segments <', max_size, 'bp'
# make detailed pairwise alignments of the segments
print "Aligning segments ..."
ref_rec = load_genbank(ref.gbk)
query_rec = load_genbank(query.gbk)
id = iter_align(chop_array, ref_rec, query_rec, aln_dir, segfile)
print "Results:", id, "% id. overall"
except IOError:
print "\nERROR: Mauve alignment failed"
raise
def align_ctg2ref(run_ref, run_id, timestamp, r_root_dir, run_dirs, genomes,
mauve_exec, max_size, chop_mode, mtype):
"""Align contigs pairwise to the reference contig."""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir + run_id + "/"
ref_ctg_file = run_ref.file
mauve_root = run_root + run_dirs['mauve_out_dir'] + ref_n + "/contigs/"
segments_root = run_root + run_dirs['aln_seg_dir'] + ref_n + "/contigs/"
q_ctgs_root = run_root + run_dirs['match_out_dir'] + ref_n + "/"
ensure_dir([segments_root])
print " ", ref_n
# log
logstring = "".join(["\n\n# Align contigs to ref @", timestamp, "\n"])
run_ref.log(logstring)
# cycle through genomes
for genome in genomes:
# set inputs and outputs
g_name = genome['name']
ctgs_fas_dir = q_ctgs_root + g_name + "/"
mauve_dir = mauve_root + g_name + "/"
aln_segs_root = segments_root + g_name + "/"
ensure_dir([mauve_dir])
print "\t", g_name, "...",
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
# list genbank files in matches directory
dir_contents = listdir(ctgs_fas_dir)
for item in dir_contents:
pattern = re.compile(r'.*_(\d*)\.fas$')
match = pattern.match(item)
if match:
ctg_num = match.group(1)
print ctg_num,
logstring = "".join(["\t", ctg_num])
run_ref.log(logstring)
# set inputs and outputs
q_contig = ctgs_fas_dir + item
file_list = (ref_ctg_file, q_contig)
mauve_outfile = mauve_dir + ctg_num + ".mauve"
aln_segs_dir = aln_segs_root + ctg_num + "/"
ensure_dir([aln_segs_dir])
segfile = aln_segs_dir + ctg_num + "_" + ref_n + "_segs.txt"
open(segfile, 'w').write('')
# do Mauve alignment
try:
open(ref_ctg_file, 'r')
open(q_contig, 'r')
except IOError:
msg = "\nERROR: File missing, cannot align\n\t\t\t"
run_ref.log(msg)
print msg
else:
align_mauve(file_list, mauve_outfile, mauve_exec)
try:
# parse Mauve output (without initial clumping)
coords = mauver_load2_k0(mauve_outfile + ".backbone",
0, mtype)
# chop segments that are too long
chop_array = chop_rows(coords, max_size, chop_mode,
mtype)
# make detailed pairwise alignments of the segments
ref_rec = load_genbank(ref_ctg_file)
query_rec = load_fasta(q_contig)
iter_align(chop_array, ref_rec, query_rec,
aln_segs_dir, segfile)
except IOError:
msg = "\nERROR: Mauve alignment failed\n\t\t\t"
run_ref.log(msg)
print msg
except Exception:
msg = "\nERROR: Iteration failed\n\t\t\t"
run_ref.log(msg)
print msg
print ""
def align_cstrct2ref(run_ref, run_id, timestamp, r_root_dir, run_dirs, genomes,
max_size, chop_mode, mtype, mauve_exec):
"""Align constructs pairwise to the reference contig."""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir + run_id + "/"
ref_ctg_file = run_ref.file
mauve_root = run_root + run_dirs['mauve_out_dir'] + ref_n + "/constructs/"
segments_root = run_root + run_dirs['aln_seg_dir'] + ref_n + "/constructs/"
scaff_root = run_root + run_dirs['scaffolds_dir'] + ref_n + "/"
ensure_dir([segments_root])
print " ", ref_n
# log
logstring = "".join(
["\n\n# Align scaffold constructs to reference @", timestamp, "\n"])
run_ref.log(logstring)
# cycle through genomes
for genome in genomes:
# set inputs
g_name = genome['name']
scaff_gbk = scaff_root + g_name + "_" + ref_n + "_scaffold.gbk"
file_list = (ref_ctg_file, scaff_gbk)
print "\t", g_name, "...",
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
# set outputs
mauve_dir = mauve_root + g_name + "/"
aln_segs_dir = segments_root + g_name + "/"
ensure_dir([mauve_dir, aln_segs_dir])
mauve_outfile = mauve_dir + g_name + "_" + ref_n + ".mauve"
segfile = aln_segs_dir + g_name + "_" + ref_n + "_segs.txt"
# abort if the reference file is not found
try:
open(ref_ctg_file, 'r')
except IOError:
msg = "ERROR: Reference file not found"
print msg
run_ref.log(msg)
raise
# abort if there is no scaffold construct
try:
open(scaff_gbk, 'r')
except IOError:
msg = "WARNING: No scaffold construct to align"
print msg
run_ref.log(msg)
else:
# prep segments file
open(segfile, 'w').write('')
# purge any pre-existing sslist file
sslist_file = scaff_gbk + ".sslist"
if os.path.isfile(sslist_file):
try:
os.remove(sslist_file)
except Exception:
raise
# do Mauve alignment
align_mauve(file_list, mauve_outfile, mauve_exec)
try:
# parse Mauve output (without initial clumping)
coords = mauver_load2_k0(mauve_outfile + ".backbone", 0, mtype)
print len(coords), '->',
logstring = "".join(["\t", str(len(coords))])
run_ref.log(logstring)
# chop segments that are too long
chop_array = chop_rows(coords, max_size, chop_mode, mtype)
print len(chop_array), 'segments <', max_size, 'bp',
logstring = "".join(["\t", str(len(chop_array))])
run_ref.log(logstring)
# make detailed pairwise alignments of the segments
ref_rec = load_genbank(ref_ctg_file)
query_rec = load_genbank(scaff_gbk)
id = iter_align(chop_array, ref_rec, query_rec, aln_segs_dir,
segfile)
print "@", id, "% id. overall"
logstring = "".join(["\t", str(id)])
run_ref.log(logstring)
except IOError:
msg = "\nERROR: Mauve alignment failed"
run_ref.log(msg)
print msg
def align_ctg2ref(run_ref, run_id, timestamp, r_root_dir, run_dirs,
genomes, mauve_exec, max_size, chop_mode, mtype):
"""Align contigs pairwise to the reference contig."""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
ref_ctg_file = run_ref.file
mauve_root = run_root+run_dirs['mauve_out_dir']+ref_n+"/contigs/"
segments_root = run_root+run_dirs['aln_seg_dir']+ref_n+"/contigs/"
q_ctgs_root = run_root+run_dirs['match_out_dir']+ref_n+"/"
ensure_dir([segments_root])
print " ", ref_n
# log
logstring = "".join(["\n\n# Align contigs to ref @", timestamp, "\n"])
run_ref.log(logstring)
# cycle through genomes
for genome in genomes:
# set inputs and outputs
g_name = genome['name']
ctgs_fas_dir = q_ctgs_root+g_name+"/"
mauve_dir = mauve_root+g_name+"/"
aln_segs_root = segments_root+g_name+"/"
ensure_dir([mauve_dir])
print "\t", g_name, "...",
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
# list genbank files in matches directory
dir_contents = listdir(ctgs_fas_dir)
for item in dir_contents:
pattern = re.compile(r'.*_(\d*)\.fas$')
match = pattern.match(item)
if match:
ctg_num = match.group(1)
print ctg_num,
logstring = "".join(["\t", ctg_num])
run_ref.log(logstring)
# set inputs and outputs
q_contig = ctgs_fas_dir+item
file_list = (ref_ctg_file, q_contig)
mauve_outfile = mauve_dir+ctg_num+".mauve"
aln_segs_dir = aln_segs_root+ctg_num+"/"
ensure_dir([aln_segs_dir])
segfile = aln_segs_dir+ctg_num+"_"+ref_n+"_segs.txt"
open(segfile, 'w').write('')
# do Mauve alignment
try:
open(ref_ctg_file, 'r')
open(q_contig, 'r')
except IOError:
msg = "\nERROR: File missing, cannot align\n\t\t\t"
run_ref.log(msg)
print msg
else:
align_mauve(file_list, mauve_outfile, mauve_exec)
try:
# parse Mauve output (without initial clumping)
coords = mauver_load2_k0(mauve_outfile+".backbone",
0, mtype)
# chop segments that are too long
chop_array = chop_rows(coords, max_size, chop_mode,
mtype)
# make detailed pairwise alignments of the segments
ref_rec = load_genbank(ref_ctg_file)
query_rec = load_fasta(q_contig)
iter_align(chop_array, ref_rec, query_rec,
aln_segs_dir, segfile)
except IOError:
msg = "\nERROR: Mauve alignment failed\n\t\t\t"
run_ref.log(msg)
print msg
except Exception:
msg = "\nERROR: Iteration failed\n\t\t\t"
run_ref.log(msg)
print msg
print ""
def align_cstrct2ref(run_ref, run_id, timestamp, r_root_dir, run_dirs,
genomes, max_size, chop_mode, mtype, mauve_exec):
"""Align constructs pairwise to the reference contig."""
# set inputs and outputs
ref_n = run_ref.name
run_root = r_root_dir+run_id+"/"
ref_ctg_file = run_ref.file
mauve_root = run_root+run_dirs['mauve_out_dir']+ref_n+"/constructs/"
segments_root = run_root+run_dirs['aln_seg_dir']+ref_n+"/constructs/"
scaff_root = run_root+run_dirs['scaffolds_dir']+ref_n+"/"
ensure_dir([segments_root])
print " ", ref_n
# log
logstring = "".join(["\n\n# Align scaffold constructs to reference @",
timestamp, "\n"])
run_ref.log(logstring)
# cycle through genomes
for genome in genomes:
# set inputs
g_name = genome['name']
scaff_gbk = scaff_root+g_name+"_"+ref_n+"_scaffold.gbk"
file_list = (ref_ctg_file, scaff_gbk)
print "\t", g_name, "...",
# log
logstring = "".join(["\n", g_name])
run_ref.log(logstring)
# set outputs
mauve_dir = mauve_root+g_name+"/"
aln_segs_dir = segments_root+g_name+"/"
ensure_dir([mauve_dir, aln_segs_dir])
mauve_outfile = mauve_dir+g_name+"_"+ref_n+".mauve"
segfile = aln_segs_dir+g_name+"_"+ref_n+"_segs.txt"
# abort if the reference file is not found
try: open(ref_ctg_file, 'r')
except IOError:
msg = "ERROR: Reference file not found"
print msg
run_ref.log(msg)
raise
# abort if there is no scaffold construct
try: open(scaff_gbk, 'r')
except IOError:
msg = "WARNING: No scaffold construct to align"
print msg
run_ref.log(msg)
else:
# prep segments file
open(segfile, 'w').write('')
# purge any pre-existing sslist file
sslist_file = scaff_gbk+".sslist"
if os.path.isfile(sslist_file):
try: os.remove(sslist_file)
except Exception: raise
# do Mauve alignment
align_mauve(file_list, mauve_outfile, mauve_exec)
try:
# parse Mauve output (without initial clumping)
coords = mauver_load2_k0(mauve_outfile+".backbone", 0, mtype)
print len(coords), '->',
logstring = "".join(["\t", str(len(coords))])
run_ref.log(logstring)
# chop segments that are too long
chop_array = chop_rows(coords, max_size, chop_mode, mtype)
print len(chop_array), 'segments <', max_size, 'bp',
logstring = "".join(["\t", str(len(chop_array))])
run_ref.log(logstring)
# make detailed pairwise alignments of the segments
ref_rec = load_genbank(ref_ctg_file)
query_rec = load_genbank(scaff_gbk)
id = iter_align(chop_array, ref_rec, query_rec,
aln_segs_dir, segfile)
print "@", id, "% id. overall"
logstring = "".join(["\t", str(id)])
run_ref.log(logstring)
except IOError:
msg = "\nERROR: Mauve alignment failed"
run_ref.log(msg)
print msg