def match_fasta(self, fasta_text, do_align=False):
fasta_fileobj = fasta(fasta_text, 's').readentries()
queries = []
for entry in fasta_fileobj:
query = self.__qm.new_query(entry['sequence'])
query.label = entry['name']
queries.append(query)
return [self.__mm.match(q) for q in queries]
from oldowan.mtconvert import seq2sites, sites2seq, str2sites
from oldowan.fasta import fasta
from string import translate
import pandas as pd
import sys
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
ff = fasta('non_2011.fasta', 'r')
data = ff.readentries()
ff.close()
hids = []
seqs = []
sites = []
# four sequences are shorter than all the rest, will drop them
for e in data:
if len(e['sequence']) > 350:
hids.append(e['name'].split()[0])
seqs.append(e['sequence'])
sites.append(seq2sites(e['sequence']))
## Validate
passed_validation = True
from oldowan.mtconvert import seq2sites, sites2seq, str2sites
from oldowan.fasta import fasta
from string import translate
import pandas as pd
import sys
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
region = range2region(metadata.ix[0,'SeqRange'])
ff = fasta('tishkoff2007.fasta', 'rU')
data = ff.readentries()
ff.close()
drop = ['San_43', 'San_67', 'tzbg040', 'tzdt045', 'tzhz108', 'tzhz130', 'tzhz131']
hids = []
seqs = []
for e in data:
name_parts = e['name'].split()
if name_parts[0] not in drop:
hids.append(name_parts[0])
seqs.append(e['sequence'])
## Validate
passed_validation = True
from oldowan.mtconvert import seq2sites, sites2seq, str2sites
from oldowan.fasta import fasta
from string import translate
import pandas as pd
import sys
import re
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
region = range2region(metadata.ix[0, 'SeqRange'])
ff = fasta('krings1999.fasta', 'r')
data = ff.readentries()
ff.close()
hids = []
sites = []
for entry in data:
hids.append(entry['name'])
sites.append(seq2sites(entry['sequence']))
# some of the sequences are short. Many are just missing a base or two
# from the beginning or the end - will keep those
# a few are missing large chunks of the end, so will drop those
ok = [41, 42, 43, 44, 54, 55, 83, 85, 89, 136, 137, 157, 178]
skip = [8, 123, 124]
import sys
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
region = range2region(metadata.ix[0, 'SeqRange'])
## load sample info
sinfo = pd.read_csv('HGDP_info.csv', index_col=0)
newindices = ['HGDP' + str(x).zfill(5) for x in sinfo.index]
sinfo['hgdpid'] = newindices
sinfo = sinfo.set_index('hgdpid')
ff = fasta('hgdp_africa.fasta', 'r')
data = ff.readentries()
ff.close()
hids = []
sites = []
for entry in data:
words = entry['name'].split()
hids.append(words[4])
sites.append(seq2sites(entry['sequence']))
# three sequences have an 'N' at around 309 that breaks validation
# this will be treated as a heteroplasy of the T there and ignored
skip = [64, 67, 73]
def process_seq2sites(form):
"""Process data submitted in seq2sites form"""
# submission validation and error reporting
problems = []
valid = True
# first, just assume whatever is in the textarea is the submission
# even if that may be nothing
content = form.cleaned_data['query']
# then check to see if a file was supplied, and if so, replace the
# previously assumed content with the file data
if form.cleaned_data['file'] is not None:
if form.cleaned_data['file'].multiple_chunks():
pass
# error - return with error
content = form.cleaned_data['file'].read()
# clear off any trailing whitespace
content.strip()
# make sure something was submitted
if len(content) == 0:
valid = False
return HttpResponseRedirect(reverse('seq2sites'))
# determine format
format = None
if content.startswith('>'):
format = 'fasta'
else:
format = 'single_seq'
# pull names and sequence out of submitted content
names = []
seqs = []
if format == 'fasta':
try:
fnames = []
fseqs = []
for entry in fasta(content, 's'):
fnames.append(entry['name'])
fseqs.append(RE_NON_IUPAC.sub('', entry['sequence'].upper()))
names = fnames
seqs = fseqs
except:
valid = False
problems.append('There was an error in the FASTA format')
else:
names = ['']
seqs = [RE_NON_IUPAC.sub('', content.upper())]
# enforce limits for multisequence submissions
if format == 'fasta':
if len(seqs) > MAX_SEQS:
valid = False
problems.append('Too many sequences submitted; current maximum allowed is %d' % MAX_SEQS)
if valid:
result_lines = []
sites_by_line = []
for seq in seqs:
try:
sites = seq2sites(seq)
sites_by_line.append(sites)
result_lines.append(sites2str(sites))
except Exception, e:
result_lines.append('There was an error: %s' % e)
results = list(Result(x,y) for x,y in zip(names,result_lines))
c = Context({'results': results})
from string import translate
import pandas as pd
import re
import sys
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
regionstr = metadata.ix[0,'SeqRange']
regionparts = regionstr.split(';')
region1 = range2region(regionparts[0])
region2 = range2region(regionparts[1])
ff = fasta('knight2003.fasta', 'r')
data = ff.readentries()
ff.close()
hids = []
seqs = []
sites = []
for e in data:
hids.append(e['name'].split()[0])
seqs.append(e['sequence'])
m = re.search(r'GATCACA', e['sequence'])
cut = m.start()
seq1 = e['sequence'][:cut]
seq2 = e['sequence'][cut:]
sites.append(seq2sites(seq1) + seq2sites(seq2))
from oldowan.mtconvert import seq2sites, sites2seq, str2sites
from oldowan.fasta import fasta
from string import translate
import pandas as pd
import sys
import re
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
region = range2region(metadata.ix[0, 'SeqRange'])
ff = fasta('barbieri2012.fasta', 'r')
data = ff.readentries()
ff.close()
hids = []
groups = []
sites = []
sequences = []
for entry in data:
words = entry['name'].split()
if entry['sequence'].count('N') > 10:
print "Too many Ns in %s (%s), skipping" % (words[0], words[1])
else:
hids.append(words[0])
groups.append(words[1])
sites.append(seq2sites(entry['sequence']))
from oldowan.mtconvert import seq2sites, sites2seq, str2sites
from oldowan.fasta import fasta
from string import translate
import pandas as pd
import sys
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
region = range2region(metadata.ix[0, 'SeqRange'])
ff = fasta('coia2005.fasta', 'r')
data = ff.readentries()
ff.close()
hids = []
sites = []
for entry in data:
words = entry['name'].split()
hids.append(words[4])
sites.append(seq2sites(entry['sequence']))
# validate
passed_validation = True
for i in range(len(sites)):
seq1 = data[i]['sequence']
if not seq1 == translate(sites2seq(sites[i], region), None, '-'):
from string import translate
import pandas as pd
import sys
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
regionstr = metadata.ix[0,'SeqRange']
regionparts = regionstr.split(';')
region1 = range2region(regionparts[0])
region2 = range2region(regionparts[1]) + range2region(regionparts[2])
ff = fasta('Poloni2009HVR1.fasta', 'r')
data1 = ff.readentries()
ff.close()
ff = fasta('Poloni2009HVR2.fasta', 'r')
data2 = ff.readentries()
ff.close()
hvr1 = {}
hvr2 = {}
for e in data1:
k = e['name'].split()[0]
hvr1[k] = e['sequence']
for e in data2:
sys.path.append("../../scripts")
from utils import *
## load metadata
metadata = pd.read_csv("metadata.csv", index_col=0)
region = range2region(metadata.ix[0, "SeqRange"])
groups = pd.read_csv("barbieri2014_groups.csv", index_col=0)
newindex = [x.upper() for x in groups.index]
groups.index = pd.Index(newindex)
hids = []
seqs = []
for filename in ["barbieri2013a.fasta", "barbieri2013b.fasta", "barbieri2013c.fasta", "barbieri2014.fasta"]:
ff = fasta(filename, "r")
data = ff.readentries()
ff.close()
for e in data:
if e["sequence"].count("N") < 20:
hid = e["name"].split()[4].upper()
if hid in groups.index:
hids.append(hid)
seqs.append(e["sequence"])
sites = []
for s in seqs:
sites.append(seq2sites(s, ambig_cutoff=20))
## Validate
from oldowan.mtconvert import seq2sites, sites2seq, str2sites
from oldowan.fasta import fasta
from string import translate
import pandas as pd
import re
import sys
sys.path.append('../../scripts')
from utils import *
## load metadata
metadata = pd.read_csv('metadata.csv', index_col=0)
region = range2region(metadata.ix['Nai', 'SeqRange'])
ff = fasta('brandstaetter2004.fasta', 'r')
data = ff.readentries()
ff.close()
## Validate
passed_validation = True
for entry in data:
seq1 = entry['sequence']
# Brandstatter et al. put an N at the end of an unstable poly-C run at the end
# of HVR3 in 5 samples. This spurious N messes with my conversion utility,
# so I strip it out.
if seq1.endswith('NACA'):
seq1 = seq1[:-4] + 'ACA'
mysites = seq2sites(seq1)
if not seq1 == translate(sites2seq(mysites, region), None, '-'):
passed_validation = False
from oldowan.fasta import fasta
DLOOPplus = range(15800,16570) + range(1,1500)
dloops_fn = os.path.join(here, 'dloops.fasta')
ofn = os.path.join(here, 'fail_dloop.txt')
fail_fn = os.path.join(here, 'dloop_expect_to_fail.txt')
expect_to_fail = []
for line in open(fail_fn, 'U'):
expect_to_fail.append(int(line.strip()[:-1]))
of = open(ofn, 'w')
count = 0
for entry in fasta(dloops_fn):
count += 1
if count not in expect_to_fail:
try:
sites = seq2sites(entry["sequence"])
seq = sites2seq(sites, region=DLOOPplus)
seq = seq.replace('-', '')
if entry["sequence"] in seq:
print count, entry["name"], sites2str(sites)
else:
print 'FAILED', count, entry["name"]
of.write('%d, %s\n' % (count, entry["name"]))
except Exception, e:
print 'FAILED',e, count, entry["name"]
of.write('%d, %s\n' % (count, entry["name"]))
of.flush()
