def Run2(flist):
aps = AllPossStrings('ACGT', 7)
NF = len(flist) # the number of files
for i in range(NF):
print 'Working on ', flist[i]
# get the data
dna = fasta.Fasta('bacteria/' + flist[i])
# count the number of genes
NG = len(dna)
# if there are multiple genes then combine
if NG > 1:
t = []
for j in range(NG):
t.append(dna[j][1])
st = ''.join(t)
del t
else:
st = dna[0][1]
del dna
# for every million bases make a plot
NP = int(len(st) / 1000000)
#if NP > 3:
#NP = 3
for j in range(NP):
print '\tPortion', j, 'of', NP
ctr = Counter(aps, st[j * 1000000:j * 1000000 + 1000000])
akando.a2i(ctr).save('work/chaos' + str(i) + 'c' + str(j) + '.gif')
def design_probes(self,input):
# setup some log formating so we have a record of activity
time_str = strftime("%a, %d %b %Y %H:%M:%S", localtime())
print "Design probes request: %s" % time_str
inseq = fasta.Fasta(input.inseq,strflag=True)
# We'll put in the target Tm range in here manually (and the targetTm).
# If we want to make this an adjustable option later,
# we can do that here without having to refactor a bunch
# of downstream code.
targetRange = [-26.0, -20.0]
targetTm = -23.0
results = find_probes.designV4(inseq,
input.noligos,
input.oligo_length,
input.spacer_length,
input.maskingflag,
input.species,
targetTm,
targetRange)
alignment = probe_design.alignOutput(results['masked_seq'],
results['output'][-1][1],input.oligo_length)
# We have to replace ending spaces with non-whitespace characters
# so the XML serializing wont cut out all the work we did to
# nicely align the positions of the probes and labels against inseq
align = Alignment()
align.raw_sequence = '/' + inseq.one_line() + '/' # masked is alignment[0]
align.probe_oligos = '/' + alignment[1][0:-1] + '/'
align.labels = '/' + alignment[2][0:-1] + '/'
maxoligos = results['output'][-1] # scores, [1] matches [2] oligos
maxoligos = probe_design.probeNames(maxoligos[2],input.probeprefix)
olis = []
for i in range(0,len(maxoligos)): # for each designed oligo
oli = ProbeOligo()
oli.GC = maxoligos[i][0]
oli.outseq = maxoligos[i][1]
oli.label = maxoligos[i][2]
olis.append(oli)
output = FindProbesOutput()
output.alignment = align
output.oligos = olis
print "Successful design of %d oligos" % len(olis)
return output
def extract(filename):
global fcount, ftotal
fcount += 1
#print "\r%i/%i" % (fcount, ftotal),
print "%i/%i" % (fcount, ftotal),
exti = filename.rfind('.')
ext = filename[exti:]
path, query = os.path.split(filename[:exti])
fhits = fasta.read_from(filename)
print filename, exti, ext, path, query
if query in fhits:
fquery = fasta.Fasta()
fquery.add_seq(fhits[query])
fhits.remove(query)
fhits.save_to(os.path.join(path, query + '-hits' + ext))
fquery.save_to(os.path.join(path, query + '-query' + ext))
def recombine(self, sequences):
'''
Recombine the sequences with the provided recombinations
>>> str(vdj_recombination().recombine([fasta.Fasta('a', 'AATTAT'),\
fasta.Fasta('b', 'GGGACACAT'),\
fasta.Fasta('c', 'ATAGATATGA')]))
'>a 0//0 b 0//0 c\\nAATTAT\\nGGGACACAT\\nATAGATATGA\\n\\n'
>>> str(vdj_recombination(deletions=[(lambda: 2)]).recombine([fasta.Fasta('a', 'AATTAT'),\
fasta.Fasta('b', 'GGGACACAT'),\
fasta.Fasta('c', 'ATAGATATGA')]))
'>a 2//2 b 2//2 c\\nAATT\\nGACAC\\nAGATATGA\\n\\n'
'''
name = ''
seq = ''
insertions = self.insertions * (len(sequences) - 1)
deletions = self.deletions * (len(sequences) * 2 - 2)
process = self.processing * (len(sequences))
for i, sequence in enumerate(sequences):
nb_deletions_start = 0
nb_deletions_end = 0
N_insertions = ''
sequence.seq = sequence.seq.translate(None, '.')
if i > 0:
# Start deletion
nb_deletions_start = deletions[2 * i - 1]()
name += '/%d ' % nb_deletions_start
name += sequence.name
if i < len(sequences) - 1:
# End deletion
nb_deletions_end = deletions[2 * i]()
N_insertions = random_sequence(['A', 'C', 'G', 'T'],\
insertions[i]())
name += ' %d/%s' % (nb_deletions_end, N_insertions)
nb_deletions_end = -nb_deletions_end if nb_deletions_end > 0 else None
seq += process[i](sequence.seq[nb_deletions_start:nb_deletions_end]
) + "\n" + N_insertions + "\n"
return fasta.Fasta(name, seq)
#!/usr/bin/env python
import sys
import os
import fasta
splitext = os.path.splitext
if __name__ == '__main__':
for filename in sys.argv[1:]:
f = fasta.Fasta()
f.read_from(filename)
for s in f:
s.sequence.replace('NNN', '')
basename, ext = splitext(filename)
f.save_to(basename + '_N_removed' + ext)
import fasta
import bowtie_search as bts
import time
hot = fasta.Fasta("HOTAIR.txt")
tstart = time.time()
hits = bts.align(hot, 16, 'humanPseudo')
print "%f seconds" % (time.time() - tstart)
import bowtie_search
import fasta
import seq
import re
import math
import probe_design
lgr = fasta.Fasta('Lgr5.txt')
shark_tank = fasta.Fasta('shark_tank.fa')
for i in range(5,20):
hts = bowtie_search.align_for_hits(shark_tank,i,'humanReference')
sm1 = sum(hts)*1.0/len(hts)
hts = bowtie_search.align_for_hits(lgr,i,'humanReference')
sm2 = sum(hts)*1.0/len(hts)
print(i,sm1,sm2,sm1/sm2)
for i in range(5,20):
hts = bowtie_search.align_for_hits(shark_tank,i,'humanPseudo')
sm1 = sum(hts)*1.0/len(hts)
hts = bowtie_search.align_for_hits(lgr,i,'humanPseudo')
sm2 = sum(hts)*1.0/len(hts)
print(i,sm1,sm2,sm1/sm2)
for i in range(5,20):
hts = bowtie_search.align_for_hits(shark_tank,i,'humanMito')
sm1 = sum(hts)*1.0/len(hts)
hts = bowtie_search.align_for_hits(lgr,i,'humanMito')
sm2 = sum(hts)*1.0/len(hts)
# Load fasta file w/ contigs sequences. Typically, 454LargeContigs.fna
contigs = dict([(x.name, x) for x in fasta.load(open(sys.argv[1]))])
# get sorted list of contigs by length
lengths = [(contigs[x].length, x) for x in contigs]
lengths.sort(reverse=True)
cn = sys.argv[2]
# output BED tracks
cf = open(outdir + '/contigs.bed', 'w')
gf = open(outdir + '/gaps.bed', 'w')
cstart = 0
gn = 1
for length, name in lengths:
print >> cf, cn, cstart, cstart + length, name, 1000, '+'
cstart = cstart + length
print >> gf, cn, cstart, cstart + 100, 'gap%s' % gn
cstart = cstart + len(gap)
gn += 1
cf.close()
gf.close()
chromseq = gap.join([contigs[name].seq for length, name in lengths])
build = fasta.Fasta(cn, None, chromseq)
print build
sys.exit(1)
if o.gff2 is not None and o.gff3 is not None:
print "Specify either gff2 or gff3 but not both."
sys.exit(1)
if o.fas is None:
print "Specify the fasta database file."
sys.exit(1)
if o.output is None:
o.output = sys.stdout
else:
o.output = file(o.output, "w")
fas = fasta.Fasta()
fas.read_from(o.fas)
if o.gff2:
gff = parse_gff2.parse_gff2(o.gff2)
else:
gff = parse_gff3.parse_gff3(o.gff3)
try:
l = [
fasta.Sequence(
g.reference_sequence + ' ' + g.start + ' ' + g.stop + ' ' +
g.strand, fas[g.reference_sequence].sequence[int(g.start) -
1:int(g.stop)])
for g in gff
]
