def rmdup(prog, opts, wd, hitDict, organelle, nucleus):
dupDict = {}
dupCounts = 0
nuclearDups = set()
# command line format strings
lastdb = opts.last + "/lastdb %s %s %s"
lastal = opts.last + "/lastal -e%s -j4 -f0 %s %s | grep -v '#'"
lastex = opts.last + "/lastex -E%s %s.prj %s.prj | sed -n 4p - | cut -f1"
# flanking sequence file names
flankSeq = wd + "/flankSeq.fa"
flankRev = wd + "/flankRev.fa"
# index file name
flankIndex = wd + "/flankIndex"
# compare flanking sequence similarity
if opts.verbose:
print prog + ": filter nuclear duplications"
print prog + ": compare flanking sequence similarities"
wrote = extractFlankSeq(flankSeq, flankRev,
opts.flank_len, hitDict, nucleus)
if not wrote:
hitCounts = sum(map(len, hitDict.values()))
return hitCounts, dupDict, dupCounts
commands.getoutput(lastdb % ("-c", flankIndex, flankSeq))
score = commands.getoutput(lastex % ('1e-25', flankIndex, flankIndex))
score, evalue = evalueSimulation(lastex, lastal, flankIndex,
flankIndex, flankRev, score, '1e-25')
flankResult = commands.getoutput(lastal % (score, flankIndex, flankSeq))
nuclearDups = formatResult(flankResult, opts.dup_coverage)
# crosscheck with segmental duplication database
if opts.segdup_db:
if opts.verbose:
print prog + ": crosscheck with segmental duplication database"
segmentalDups = checkSegDup(opts.segdup_db, hitDict, opts.dup_coverage)
nuclearDups = nuclearDups.union(segmentalDups)
# remove duplicate hits
dupDict = filterDups(nuclearDups, hitDict)
hitCounts = sum(map(len, hitDict.values()))
dupCounts = sum(map(len, dupDict.values()))
return hitCounts, dupDict, dupCounts
def rmdup(prog, opts, wd, hitDict, organelle, nucleus):
dupDict = {}
dupCounts = 0
nuclearDups = set()
# command line format strings
lastdb = opts.last + "/lastdb %s %s %s"
lastal = opts.last + "/lastal -e%s -j4 -f0 %s %s | grep -v '#'"
lastex = opts.last + "/lastex -E%s %s.prj %s.prj | sed -n 4p - | cut -f1"
# flanking sequence file names
flankSeq = wd + "/flankSeq.fa"
flankRev = wd + "/flankRev.fa"
# index file name
flankIndex = wd + "/flankIndex"
# compare flanking sequence similarity
if opts.verbose:
print prog + ": filter nuclear duplications"
print prog + ": compare flanking sequence similarities"
wrote = extractFlankSeq(flankSeq, flankRev, opts.flank_len, hitDict,
nucleus)
if not wrote:
hitCounts = sum(map(len, hitDict.values()))
return hitCounts, dupDict, dupCounts
commands.getoutput(lastdb % ("-c", flankIndex, flankSeq))
score = commands.getoutput(lastex % ('1e-25', flankIndex, flankIndex))
score, evalue = evalueSimulation(lastex, lastal, flankIndex, flankIndex,
flankRev, score, '1e-25')
flankResult = commands.getoutput(lastal % (score, flankIndex, flankSeq))
nuclearDups = formatResult(flankResult, opts.dup_coverage)
# crosscheck with segmental duplication database
if opts.segdup_db:
if opts.verbose:
print prog + ": crosscheck with segmental duplication database"
segmentalDups = checkSegDup(opts.segdup_db, hitDict, opts.dup_coverage)
nuclearDups = nuclearDups.union(segmentalDups)
# remove duplicate hits
dupDict = filterDups(nuclearDups, hitDict)
hitCounts = sum(map(len, hitDict.values()))
dupCounts = sum(map(len, dupDict.values()))
return hitCounts, dupDict, dupCounts
def checkGenomes(prog, coord, ref, labels, gens, spes, spesV, opts, wd):
# command line format strings
lastdb = opts.last + "/lastdb %s %s %s"
lastal = opts.last + "/lastal -e%s -j4 -f0 %s %s | grep -v '#'"
lastex = opts.last + "/lastex -E%s %s.prj %s.prj | sed -n 4p - | cut -f1"
# index file name
bBaseFreq = wd + "/bBaseFreq"
# hit and its flanking sequences
baitSeq = wd + "/baitSeq.fa"
baitRev = wd + "/baitRev.fa"
i = [k for k in spes.keys() if spes[k] == ref][0]
extractBaitSeq(baitSeq, baitRev, gens[i], coord)
commands.getoutput(lastdb % ("-x", bBaseFreq, baitSeq))
spes.pop(i)
L = spes.keys()
L.sort()
homolog = {}
for chrom in coord.keys():
for obj in coord[chrom]:
key = ":".join([chrom, obj[0], obj[1]])
beg = map(int, obj[0].split(","))
end = map(int, obj[1].split(","))
l = sum([end[i] - beg[i] for i in range(len(beg))])
homolog.setdefault(key, [{} for j in L])
homolog[key].append(l + 400)
for j in range(len(L)):
if opts.verbose:
print prog + ": ...... surveying: " + spesV[L[j]]
gIndex = wd + "/" + spes[L[j]] + "Index"
commands.getoutput(lastdb % ("-c", gIndex, gens[L[j]]))
score = commands.getoutput(lastex % ('1e-10', gIndex, bBaseFreq))
score, e = evalueSimulation(lastdb, lastal, gIndex, bBaseFreq, baitRev,
score, '1e-10')
alignResult = commands.getoutput(lastal % (score, gIndex, baitSeq))
formatResult(j, alignResult, homolog)
return homolog
def checkGenomes(prog, coord, ref, labels, gens, spes, spesV, opts, wd):
# command line format strings
lastdb = opts.last + "/lastdb %s %s %s"
lastal = opts.last + "/lastal -e%s -j4 -f0 %s %s | grep -v '#'"
lastex = opts.last + "/lastex -E%s %s.prj %s.prj | sed -n 4p - | cut -f1"
# index file name
bBaseFreq = wd + "/bBaseFreq"
# hit and its flanking sequences
baitSeq = wd + "/baitSeq.fa"
baitRev = wd + "/baitRev.fa"
i = [k for k in spes.keys() if spes[k] == ref][0]
extractBaitSeq(baitSeq, baitRev, gens[i], coord)
commands.getoutput(lastdb % ("-x", bBaseFreq, baitSeq))
spes.pop(i)
L = spes.keys()
L.sort()
homolog = {}
for chrom in coord.keys():
for obj in coord[chrom]:
key = ":".join([chrom, obj[0], obj[1]])
beg = map(int, obj[0].split(","))
end = map(int, obj[1].split(","))
l = sum([end[i] - beg[i] for i in range(len(beg))])
homolog.setdefault(key, [{} for j in L])
homolog[key].append(l + 400)
for j in range(len(L)):
if opts.verbose:
print prog + ": ...... surveying: " + spesV[L[j]]
gIndex = wd + "/" + spes[L[j]] + "Index"
commands.getoutput(lastdb % ("-c", gIndex, gens[L[j]]))
score = commands.getoutput(lastex % ("1e-10", gIndex, bBaseFreq))
score, e = evalueSimulation(lastdb, lastal, gIndex, bBaseFreq, baitRev, score, "1e-10")
alignResult = commands.getoutput(lastal % (score, gIndex, baitSeq))
formatResult(j, alignResult, homolog)
return homolog
