def importPrimerPairs(fastafile):
primerfile = MultiFasta(fastafile)
primers = primerfile.createPrimers(
config['targeting']['bowtieindex']) # places in genome
for primer in primers:
primer.calcProperties() # calc Tm and GC
# find pairs
left_suffix, rite_suffix = ['F', 'f', 'L', 'l', '5',
'left'], ['R', 'r', '3', 'right']
pairs = []
for i in range(len(primers)):
primerI = [
'_'.join(primers[i].name.split('_')[:-1]),
primers[i].name.split('_')[-1]
]
for j in range(i, len(primers)):
if i == j: continue # skip same primer
primerJ = [
'_'.join(primers[j].name.split('_')[:-1]),
primers[j].name.split('_')[-1]
]
if primerI[0] == primerJ[0]:
if primerI[1] in left_suffix and primerJ[1] in rite_suffix:
pairs.append([primers[i], primers[j]])
elif primerJ[1] in left_suffix and primerI[1] in rite_suffix:
pairs.append([primers[j], primers[i]])
else:
raise Exception('saasd')
# return valid pairs
return pairs
def importPrimerPairs(fastafile):
primerfile = MultiFasta(fastafile)
primers = primerfile.createPrimers(config['targeting']['bowtieindex']) # places in genome
for primer in primers:
primer.calcProperties() # calc Tm and GC
# find pairs
left_suffix, rite_suffix = ['F','f','L','l','5','left'],['R','r','3','right']
pairs = []
for i in range(len(primers)):
primerI = [ '_'.join(primers[i].name.split('_')[:-1]), primers[i].name.split('_')[-1] ]
for j in range(i,len(primers)):
if i==j: continue # skip same primer
primerJ = [ '_'.join(primers[j].name.split('_')[:-1]), primers[j].name.split('_')[-1] ]
if primerI[0] == primerJ[0]:
if primerI[1] in left_suffix and primerJ[1] in rite_suffix:
pairs.append([primers[i], primers[j]])
elif primerJ[1] in left_suffix and primerI[1] in rite_suffix:
pairs.append([primers[j], primers[i]])
else:
raise Exception('saasd')
# return valid pairs
return pairs
def importPrimerPairs(inputfile, config, primer3=True):
# read table/fasta
primersets = defaultdict(list) # pair primersets
primertags = {} # primer tags from table
if not inputfile.split('.')[-1].startswith(
'fa'): # ignores duplicate sequence
primerseqs = {}
fastafile = 'import_' + fileMD5(inputfile)[:8] + '.fasta'
with open(fastafile, 'w') as outfh:
with open(inputfile) as infh:
for i, line in enumerate(infh):
if i == 0:
minimalHeader = set([
'primername', 'primerset', 'tag', 'sequence',
'vessel', 'well'
])
header = map(lambda x: x.lower(),
line.rstrip().split('\t'))
try:
assert not minimalHeader.difference(set(header))
except:
print >> sys.stderr, 'ERROR: Missing columns (%s)' % ','.join(
list(minimalHeader.difference(set(header))))
raise Exception('FileHeaderError')
else:
f = map(lambda x: x.strip('"'),
line.rstrip().split('\t'))
l = dict(zip(header, f))
# remove tag from sequence
if l['tag']:
try:
tagseqs = config['ordersheet']['sequencetags'][
l['tag']]['tags']
except:
pass
else:
for t in tagseqs:
if l['sequence'].startswith(t):
l['sequence'] = l['sequence'][len(t):]
break
# store metadata and write fasta
if l['primername'] in primerseqs.keys():
try:
assert l['sequence'] == primerseqs[
l['primername']]
assert l['tag'] == primertags[l['primername']]
except:
print >> sys.stderr, l['primername']
print >> sys.stderr, primerseqs[
l['primername']]
print >> sys.stderr, primertags[
l['primername']]
raise Exception('ImportFormattingError')
else:
print >> outfh, '>' + l['primername']
print >> outfh, l['sequence']
if l['primerset']:
primersets[l['primername']].append(l['primerset'])
primertags[l['primername']] = l['tag']
primerseqs[l['primername']] = l['sequence']
primerfile = MultiFasta(fastafile)
else:
primerfile = MultiFasta(inputfile)
# set default tags for import
for r in primerfile.references:
primertags[r] = config['import']['tag']
print >> sys.stderr, "Placing primers on genome..."
# Align primers to genome
primers = primerfile.createPrimers(config['design']['bowtieindex'], \
delete=False,tags=primertags, \
tmThreshold=config['design']['mispriming']['minimaltm'], \
endMatch=config['design']['mispriming']['identity3prime']) # places in genome
# pair primers (by name or by primerset) MAKE COPIES!!!!
pairs = {}
for p in primers:
setnames = primersets[p.name] \
if p.name in primersets.keys() and len(primersets[p.name])>0 \
else [ parsePrimerName(p.name)[0] ]
for setname in setnames:
try:
pairs[setname]
except KeyError:
try:
pairs[setname] = PrimerPair([None, None], name=setname)
except:
print >> sys.stderr, '>>', primersets[
p.name], '|', p.name, '|', setnames, '<'
raise
except:
raise
# get primer orientation (might be wrong if guesses from name, will correct after)
## this basically just makes sure primers get paired (one fwd, one reverse)
reverse = p.targetposition.reverse if p.targetposition else parsePrimerName(
p.name)[1] < 0
try:
if reverse and pairs[setname][1] is None:
pairs[setname][1] = deepcopy(p)
else:
if pairs[setname][0] is None:
pairs[setname][0] = deepcopy(p)
else:
assert pairs[setname][1] is None
pairs[setname][1] = deepcopy(p)
except:
print >> sys.stderr, "ERROR: Primer pair strand conflict?"
print >> sys.stderr, "PRIMER0", pairs[setname][0]
print >> sys.stderr, "PRIMER1", pairs[setname][1]
print >> sys.stderr, "REVERSE", reverse
print >> sys.stderr, "SETNAME", setname
print >> sys.stderr, "PRIMER", p.name, parsePrimerName(p.name)
print >> sys.stderr, "PAIRS", pairs[setname]
raise
# check if any unpaired primers
for k, v in pairs.items():
if not all(v):
print >> sys.stderr, "WARNING: primer set %s is incomplete and skipped" % k
del pairs[k]
# prune ranks in primer3 mode (renames pair)
if primer3:
for p in pairs.values():
assert p[0].targetposition and p[
1].targetposition # make sure target postiions are set
p.pruneRanks()
validPairs = pairs.values()
else: # guess target if not set
validPairs = []
print >> sys.stderr, 'Identifying correct amplicons for unplaced primer pairs...'
for p in pairs.values():
if not p[0].targetposition or not p[1].targetposition:
amplicons = p.amplicons(config['import']['ampliconsize'],
autoreverse=True)
if amplicons:
shortest = sorted(
amplicons,
key=lambda x: len(x[2]))[0] # sort amplicons by size
if len(amplicons) > 1:
print >> sys.stderr, 'WARNING: multiple amplicons for {}. Assuming shortest ({}bp) is correct.'.format(
p.name, str(len(shortest[2])))
p[0].targetposition = shortest[0] # m
p[1].targetposition = shortest[1] # n
validPairs.append(p)
elif not primer3:
# try to find amplicon by sequence matching if no amplicons from genome mapping with sufficient Tm
refGenome = Genome(config['design']['genome'])
# get new loci (one round)
newLoci = [[], []]
for mapped, query in [[0, 1], [1, 0]]:
for l in p[mapped].loci:
newLoci[query] += refGenome.primerMatch(
l, p[query].seq,
config['import']['ampliconsize'])
# add new loci
if not newLoci[0] and not newLoci[1]:
print >> sys.stderr, 'WARNING: {} is not specific and not imported ({},{})'.format(
p.name, len(p[0].loci), len(p[1].loci))
continue
else: # add new loci
for i, loc in enumerate(newLoci):
p[i].loci += loc
p[i].loci = list(set(
p[i].loci)) # remove redundancy
# store new amplicon
amplicons = p.amplicons(config['import']['ampliconsize'],
autoreverse=True)
if amplicons:
p[0].targetposition = amplicons[0][0] # m
p[1].targetposition = amplicons[0][1] # n
validPairs.append(p)
else:
print >> sys.stderr, '\n'.join(
["-> " + str(l) for l in p[0].loci])
print >> sys.stderr, '\n'.join(
["<- " + str(l) for l in p[1].loci])
print >> sys.stderr, 'WARNING: Primer set {} has no valid amplicons ({},{})'.format(
p.name, len(p[0].loci), len(p[1].loci))
else:
print >> sys.stderr, 'WARNING: Primer set {} does not produce a well-sized, unique amplicon ({},{})'.format(
p.name, len(p[0].loci), len(p[1].loci))
else:
validPairs.append(p)
return validPairs
def importPrimerPairs(inputfile, config, primer3=True):
# read table/fasta
primersets = defaultdict(list) # pair primersets
primertags = {} # primer tags from table
if not inputfile.split(".")[-1].startswith("fa"): # ignores duplicate sequence
primerseqs = {}
fastafile = "import_" + fileMD5(inputfile)[:8] + ".fasta"
with open(fastafile, "w") as outfh:
with open(inputfile) as infh:
for i, line in enumerate(infh):
if i == 0:
minimalHeader = set(["primername", "primerset", "tag", "sequence", "vessel", "well"])
header = map(lambda x: x.lower(), line.rstrip().split("\t"))
try:
assert not minimalHeader.difference(set(header))
except:
print >> sys.stderr, "ERROR: Missing columns (%s)" % ",".join(
list(minimalHeader.difference(set(header)))
)
raise Exception("FileHeaderError")
else:
f = map(lambda x: x.strip('"'), line.rstrip().split("\t"))
l = dict(zip(header, f))
# remove tag from sequence
if l["tag"]:
try:
tagseqs = config["ordersheet"]["sequencetags"][l["tag"]]["tags"]
except:
pass
else:
for t in tagseqs:
if l["sequence"].startswith(t):
l["sequence"] = l["sequence"][len(t) :]
break
# store metadata and write fasta
if l["primername"] in primerseqs.keys():
try:
assert l["sequence"] == primerseqs[l["primername"]]
assert l["tag"] == primertags[l["primername"]]
except:
print >> sys.stderr, l["primername"]
print >> sys.stderr, primerseqs[l["primername"]]
print >> sys.stderr, primertags[l["primername"]]
raise Exception("ImportFormattingError")
else:
print >> outfh, ">" + l["primername"]
print >> outfh, l["sequence"]
if l["primerset"]:
primersets[l["primername"]].append(l["primerset"])
primertags[l["primername"]] = l["tag"]
primerseqs[l["primername"]] = l["sequence"]
primerfile = MultiFasta(fastafile)
else:
primerfile = MultiFasta(inputfile)
# set default tags for import
for r in primerfile.references:
primertags[r] = config["import"]["tag"]
print >> sys.stderr, "Placing primers on genome..."
# Align primers to genome
primers = primerfile.createPrimers(
config["design"]["bowtieindex"],
delete=False,
tags=primertags,
tmThreshold=config["design"]["mispriming"]["minimaltm"],
endMatch=config["design"]["mispriming"]["identity3prime"],
) # places in genome
# pair primers (by name or by primerset) MAKE COPIES!!!!
pairs = {}
for p in primers:
setnames = (
primersets[p.name]
if p.name in primersets.keys() and len(primersets[p.name]) > 0
else [parsePrimerName(p.name)[0]]
)
for setname in setnames:
try:
pairs[setname]
except KeyError:
try:
pairs[setname] = PrimerPair([None, None], name=setname)
except:
print >> sys.stderr, ">>", primersets[p.name], "|", p.name, "|", setnames, "<"
raise
except:
raise
# get primer orientation (might be wrong if guesses from name, will correct after)
## this basically just makes sure primers get paired (one fwd, one reverse)
reverse = p.targetposition.reverse if p.targetposition else parsePrimerName(p.name)[1] < 0
try:
if reverse and pairs[setname][1] is None:
pairs[setname][1] = deepcopy(p)
else:
if pairs[setname][0] is None:
pairs[setname][0] = deepcopy(p)
else:
assert pairs[setname][1] is None
pairs[setname][1] = deepcopy(p)
except:
print >> sys.stderr, "ERROR: Primer pair strand conflict?"
print >> sys.stderr, "PRIMER0", pairs[setname][0]
print >> sys.stderr, "PRIMER1", pairs[setname][1]
print >> sys.stderr, "REVERSE", reverse
print >> sys.stderr, "SETNAME", setname
print >> sys.stderr, "PRIMER", p.name, parsePrimerName(p.name)
print >> sys.stderr, "PAIRS", pairs[setname]
raise
# check if any unpaired primers
for k, v in pairs.items():
if not all(v):
print >> sys.stderr, "WARNING: primer set %s is incomplete and skipped" % k
del pairs[k]
# prune ranks in primer3 mode (renames pair)
if primer3:
for p in pairs.values():
assert p[0].targetposition and p[1].targetposition # make sure target postiions are set
p.pruneRanks()
validPairs = pairs.values()
else: # guess target if not set
validPairs = []
print >> sys.stderr, "Identifying correct amplicons for unplaced primer pairs..."
for p in pairs.values():
if not p[0].targetposition or not p[1].targetposition:
amplicons = p.amplicons(config["import"]["ampliconsize"], autoreverse=True)
if amplicons:
shortest = sorted(amplicons, key=lambda x: len(x[2]))[0] # sort amplicons by size
if len(amplicons) > 1:
print >> sys.stderr, "WARNING: multiple amplicons for {}. Assuming shortest ({}bp) is correct.".format(
p.name, str(len(shortest[2]))
)
p[0].targetposition = shortest[0] # m
p[1].targetposition = shortest[1] # n
validPairs.append(p)
elif not primer3:
# try to find amplicon by sequence matching if no amplicons from genome mapping with sufficient Tm
refGenome = Genome(config["design"]["genome"])
# get new loci (one round)
newLoci = [[], []]
for mapped, query in [[0, 1], [1, 0]]:
for l in p[mapped].loci:
newLoci[query] += refGenome.primerMatch(l, p[query].seq, config["import"]["ampliconsize"])
# add new loci
if not newLoci[0] and not newLoci[1]:
print >> sys.stderr, "WARNING: {} is not specific and not imported ({},{})".format(
p.name, len(p[0].loci), len(p[1].loci)
)
continue
else: # add new loci
for i, loc in enumerate(newLoci):
p[i].loci += loc
p[i].loci = list(set(p[i].loci)) # remove redundancy
# store new amplicon
amplicons = p.amplicons(config["import"]["ampliconsize"], autoreverse=True)
if amplicons:
p[0].targetposition = amplicons[0][0] # m
p[1].targetposition = amplicons[0][1] # n
validPairs.append(p)
else:
print >> sys.stderr, "\n".join(["-> " + str(l) for l in p[0].loci])
print >> sys.stderr, "\n".join(["<- " + str(l) for l in p[1].loci])
print >> sys.stderr, "WARNING: Primer set {} has no valid amplicons ({},{})".format(
p.name, len(p[0].loci), len(p[1].loci)
)
else:
print >> sys.stderr, "WARNING: Primer set {} does not produce a well-sized, unique amplicon ({},{})".format(
p.name, len(p[0].loci), len(p[1].loci)
)
else:
validPairs.append(p)
return validPairs
