def test_read_and_write_fasta_short(self):
with tempfile.TemporaryFile(mode='r+') as new_fasta_file:
fasta.write(fasta.read(self.fasta_file, ignore_comments=True),
new_fasta_file)
new_fasta_file.seek(0)
new_entries = list(fasta.read(new_fasta_file,
ignore_comments=True))
self.assertEqual(new_entries, self.fasta_entries_short)
def getAllProts(fastaDir):
fastaData = fas.read(fastaDir)
listOfProts = []
for prot in fastaData:
listOfProts.append(prot.sequence)
return listOfProts
def process_fasta(df, path_to_fasta, decoy_prefix, decoy_infix=False):
protsS = dict()
decoy_check_flag = False
for x in fasta.read(path_to_fasta):
dbname = x[0].split(' ')[0]
if not decoy_check_flag:
if (not decoy_infix and dbname.startswith(decoy_prefix)) or (
decoy_infix and decoy_infix in dbname):
decoy_check_flag = True
protsS[dbname] = x[1]
df['sequence'] = df['dbname'].apply(
lambda x: protsS.get(x, protsS.get(x.split(' ')[0], '')))
if not decoy_check_flag:
if not decoy_infix:
df['sequence'] = df.apply(
lambda x: x['sequence'] if x['sequence'] else protsS.get(
x['dbname'].replace(decoy_prefix, ''),
protsS.get(
x['dbname'].split(' ')[0].replace(decoy_prefix, ''), ''
)),
axis=1)
else:
df['sequence'] = df.apply(lambda x: x['sequence']
if x['sequence'] else protsS.get(
x['dbname'].replace(decoy_infix, ''),
protsS.get(
x['dbname'].split(' ')[0].
replace(decoy_infix, ''), '')),
axis=1)
return df
def peptide_db_graph(peps, db, id_regex=None):
''' search a set of peptides against a FASTA database '''
g = nx.Graph()
prot_dict = dict()
for header, seq, in fasta.read(db):
seq = seq.replace('I', 'L').upper() # convert DB sequence I -> L
prot_id = header.split()[0]
if id_regex is not None:
find_id = re.findall(id_regex, header)
if len(find_id) > 0:
prot_id = find_id[0]
prot_dict[prot_id] = seq
def _map_seq(p):
pairs = []
for prot_id, seq, in prot_dict.items():
if p in seq:
pairs.append([p, prot_id])
return pairs
for p in peps:
ppps = _map_seq(p)
if len(ppps):
g.add_edges_from(ppps)
return g
def ingest_fasta(input_filename):
"""Ingest an fasta file given its name and return a dataframe of the file
"""
with fasta.read(input_filename) as reader:
for entry in reader:
prot_list = [[item.description.split("|")[0]+":"+item.description.split("|")[1],
item.description.split("|")[3],item.description.split("|")[4],item.sequence] for item in reader]
df = pd.DataFrame(prot_list,columns=["GeneInfo ID","Accession","Description","Sequence"])
return df
def make_reverse_fasta(input_file, output_file):
'''
Takes as input fasta file, drops all _REVERSED proteins and creates a new _REVERSED decoy proteins.
'''
prots = []
for prot_desc, prot_seq in fasta.read(input_file):
if not prot_desc.endswith('_REVERSED'):
prots.append((prot_desc, prot_seq))
prots.append((prot_desc + '_REVERSED', smart_reverse(prot_seq)))
fasta.write(prots, output_file, file_mode='w')
def _load_fasta(db, id_regex):
global prot_dict
prot_dict = dict()
for header, seq, in fasta.read(db):
seq = seq.replace('I', 'L').upper() # convert DB sequence I -> L
prot_id = header.split()[0]
if id_regex is not None:
find_id = re.findall(id_regex, header)
if len(find_id) > 0:
prot_id = find_id[0]
prot_dict[prot_id] = seq
def digetsProteinFromFASTA():
sequenceIter = fasta.read(source=options.fasta)
uniquePeptides = set()
for s in sequenceIter:
newPeptides = parser.cleave(s.sequence,
'trypsin',
missed_cleavages=options.missed,
min_length=options.minLength)
uniquePeptides.update(newPeptides)
uniquePeptides = list(uniquePeptides)
return [Peptide(x) for x in uniquePeptides]
def split_fasta_decoys(db, decoy_prefix, decoy_infix=None):
decoy_dbnames = set()
with fasta.read(db) as f:
for protein in f:
dbname = protein.description.split()[0]
if (decoy_infix and decoy_infix in dbname) or dbname.startswith(decoy_prefix):
decoy_dbnames.add(dbname)
decoy_dbnames = sorted(decoy_dbnames)
random.seed(SEED)
all_decoys_2 = set(random.sample(decoy_dbnames, len(decoy_dbnames) // 2))
logger.debug('Marking %s out of %s decoys as decoy2', len(all_decoys_2), len(decoy_dbnames))
return all_decoys_2
def test_write_decoy_db(self):
with tempfile.TemporaryFile(mode='r+') as decdb:
fasta.write_decoy_db(self.fasta_file,
decdb,
decoy_only=False,
prefix='PREFIX_')
decdb.seek(0)
all_entries = list(fasta.read(decdb, False))
self.assertEqual(
all_entries,
self.fasta_entries_long + [('PREFIX_' + a, b[::-1])
for a, b in self.fasta_entries_long])
def generate_db(fasta_file, bins):
peptides = set()
with fasta.read(fasta_file) as db:
for _, protein in db:
peptides |= generate_peptides(protein)
mzs = list()
for peptide in peptides:
mzs.append(compute_mass_spectrum(peptide))
intensity_matrix = lil_matrix((len(mzs), bins), dtype=numpy.int8)
for i, intensity in enumerate(mzs):
intensity_matrix[i, :] = bin_spectrum(mzs[i], bins=bins)
peptides_vector = numpy.array(list(peptides))
mzs_vector = numpy.array(mzs)
return peptides_vector, mzs_vector, intensity_matrix
def background_maker(args):
# print('Making background DB')
#хотим сделать background из идентифицированных белков
bg_fasta = dict()
# bg = defaultdict()
background = set()
with fasta.read(args.fasta) as f:
for name, sequence in f:
name_id = name.split('|')[1]
extended_seq = ''.join(['-' * args.interval_length, sequence, '-' * args.interval_length])
bg_fasta[name_id] = extended_seq
mod_aa_indexes = re.finditer(args.modification_site, extended_seq)
bg_intervals = [extended_seq[i.span()[0] - args.interval_length: i.span()[0] + args.interval_length + 1] for i in mod_aa_indexes]
# bg[name_id] = bg_intervals
background.update(bg_intervals)
logging.info(u'Set of %s background intervals is created', len(background))
logging.debug(u'Background DB is ready')
with open('bg.csv', 'w') as f:
f.write('\n'.join(background))
return pd.DataFrame([list(i) for i in background], columns=range(-args.interval_length, args.interval_length + 1)), bg_fasta
def read_fasta_sequences(fasta_file):
""" Read sequence records from a FASTA file. """
sequence_records = []
for description, sequence in fasta.read(fasta_file):
# Initialize sequence record with sequence string.
sequence_record = {'sequence': sequence}
# Get sequence info.
description_parts = description.split()
sequence_record['id'] = description_parts[0]
# Get the sequence's peptides.
sequence_record['peptides'] = parser.cleave(
sequence,
parser.expasy_rules['trypsin'],
missed_cleavages=1 #max no. of missed cleavages.
)
# Save the sequence record, keyed by the id.
sequence_records.append(sequence_record)
return sequence_records
def build(fasta_file: str) -> Database:
'''Create a Database namedtuple from a fasta file
:param fasta_file: the full path to a fasta database file
:type fasta_file: str
:returns: a Database object with the fasta file and protein fields filled in
:rtype: Database
'''
db = Database(fasta_file)
prots = defaultdict(list)
# pull the name out
get_name = lambda x: x.split('|')[-1].split()[0]
for entry in fasta.read(fasta_file):
p_name = get_name(entry.description)
prots[p_name].append(entry)
db = db._replace(proteins=prots)
return db
def describe(args):
"""Read database and produce a summary"""
logger.debug('describe called with: %s', args)
try:
dlist = [d for d, seq in fasta.read(args.file)]
except Exception as e:
logger.info('Not a valid FASTA file.')
logger.debug('Exception: %s', e)
else:
logger.info('Found %s FASTA entries.', len(dlist))
n = len(dlist)
if n:
logger.debug('First entry: %s', dlist[0])
if n > 2:
dlist.sort()
prefix_1 = os.path.commonprefix(dlist[:n // 2])
prefix_2 = os.path.commonprefix(dlist[n // 2 + 1:])
if prefix_1 != prefix_2:
logger.info('Common prefixes: %s, %s', prefix_1, prefix_2)
else:
logger.info('Common prefix: %s', prefix_1)
formats = []
for flavor in fasta.std_parsers:
try:
fasta.parse(dlist[0], flavor=flavor)
except Exception as e:
logger.debug('Header: %s; parsing exception: %s', dlist[0],
e)
else:
formats.append(flavor)
k = len(formats)
if not k:
logger.info('Unknown header format.')
elif k == 1:
logger.info('Suggested header format: %s', formats[0])
else:
logger.info('Possible header formats: %s', ', '.join(formats))
# -*- coding: utf-8 -*-
"""
Created on Fri Mar 22 11:28:43 2013
@author: ilya
"""
from pyteomics import fasta, mgf, parser
import pylab
fasta_file = '/home/ilya/src/pyteomics/RhoEcoli.fasta'
mgf_file = '/home/ilya/src/pyteomics/MultiConsensus.mgf'
peptides = set()
with open(fasta_file) as fi:
for description, sequence in fasta.read(fi):
new_peptides = parser.cleave(sequence, parser.expasy_rules['trypsin'])
peptides.update(new_peptides)
print "UNIQUE PEPTIDES"
print peptides
with open(mgf_file) as fi:
for spectrum in mgf.read(fi):
pylab.figure()
pylab.xlabel('m/z, Th')
pylab.ylabel('Intensity, rel.units')
pylab.bar(spectrum['m/z array'], spectrum['intensity array'], width=0.1, linewidth=2, edgecolor='black')
pylab.show()
inp = raw_input("Show more?")
if inp != "yes":
def prepare_decoy_db(args):
add_decoy = args['ad']
if add_decoy:
prefix = args['prefix']
db = args['d']
out1, out2 = os.path.splitext(db)
out_db = out1 + '_shuffled' + out2
logger.info('Creating decoy database: %s', out_db)
extra_check = False
if '{' in args['e']:
extra_check = True
if extra_check:
banned_pairs = set()
banned_aa = set()
for enzyme_local in args['e'].split(','):
if '{' in enzyme_local:
lpart, rpart = enzyme_local.split('|')
for aa_left, aa_right in itertools.product(
lpart[1:-1], rpart[1:-1]):
banned_aa.add(aa_left)
banned_aa.add(aa_right)
banned_pairs.add(aa_left + aa_right)
logger.debug(banned_aa)
logger.debug(banned_pairs)
enzyme = get_enzyme(args['e'])
cleave_rule_custom = enzyme + '|' + '([BXZUO])'
# cleave_rule_custom = '([RKBXZUO])'
logger.debug(cleave_rule_custom)
shuf_map = dict()
prots = []
for p in fasta.read(db):
if not p[0].startswith(prefix):
target_peptides = [
x[1] for x in parser.icleave(p[1], cleave_rule_custom, 0)
]
checked_peptides = set()
sample_list = []
for idx, pep in enumerate(target_peptides):
if len(pep) > 2:
pep_tmp = pep[1:-1]
if extra_check:
for bp in banned_pairs:
if bp in pep_tmp:
pep_tmp = pep_tmp.replace(bp, '')
checked_peptides.add(idx)
sample_list.extend(pep_tmp)
random.shuffle(sample_list)
idx_for_shuffle = 0
decoy_peptides = []
for idx, pep in enumerate(target_peptides):
if len(pep) > 2:
if pep in shuf_map:
tmp_seq = shuf_map[pep]
else:
if not extra_check or idx not in checked_peptides:
tmp_seq = pep[0]
for pep_aa in pep[1:-1]:
tmp_seq += sample_list[idx_for_shuffle]
idx_for_shuffle += 1
tmp_seq += pep[-1]
else:
max_l = len(pep)
tmp_seq = ''
ii = 0
while ii < max_l - 1:
# for ii in range(max_l-1):
if pep[ii] in banned_aa and pep[
ii +
1] in banned_aa and pep[ii] + pep[
ii + 1] in banned_pairs:
tmp_seq += pep[ii] + pep[ii + 1]
ii += 1
else:
if ii == 0:
tmp_seq += pep[ii]
else:
tmp_seq += sample_list[
idx_for_shuffle]
idx_for_shuffle += 1
ii += 1
tmp_seq += pep[max_l - 1]
shuf_map[pep] = tmp_seq
else:
tmp_seq = pep
decoy_peptides.append(tmp_seq)
assert len(target_peptides) == len(decoy_peptides)
prots.append((p[0], ''.join(target_peptides)))
prots.append(('DECOY_' + p[0], ''.join(decoy_peptides)))
fasta.write(prots, open(out_db, 'w')).close()
args['d'] = out_db
args['ad'] = 0
return args
sep='\t',
header=None,
names=['sb1', 'sb2', 'genome', 'start', 'end', 'maxlen']).drop(0)
dataset = os.listdir(args.orgs + org + '/All/')
for refsta in strains:
for strain in dataset:
if '.DS_Store' in strain:
continue
if refsta in strain[:-12]:
print('Analyzing of {} ...'.format(refsta))
edge_cake = ''
cake = ''
seq = ''
SBs = ''
coord_g = coord.loc[coord.genome == strain[:-12]]
fasta = read(args.orgs + org + '/All/' + strain)
for line in fasta:
seq += line.sequence
for i in coord_g.index:
if i != 0:
if int(coord_g.start[i]) > int(coord_g.end[i]):
seq_i = seq[int(coord_g.start[i]
):int(coord_g.end[i]):-1]
else:
seq_i = seq[int(coord_g.start[i]):int(coord_g.
end[i])]
if args.cut[-1] == '%':
if len(seq_i) < 50:
continue
left_edge = seq_i[:int(
len(seq_i) * float(args.cut[:-1]) * 0.01)]
def run():
parser = argparse.ArgumentParser(
description='run DirectMS1quant for ms1searchpy results',
epilog='''
Example usage
-------------
$ directms1quant -S1 sample1_1_proteins_full.tsv sample1_n_proteins_full.tsv -S2 sample2_1_proteins_full.tsv sample2_n_proteins_full.tsv
-------------
''',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('-S1',
nargs='+',
help='input files for S1 sample',
required=True)
parser.add_argument('-S2',
nargs='+',
help='input files for S2 sample',
required=True)
parser.add_argument('-out',
help='name of DirectMS1quant output file',
default='directms1quant_out')
parser.add_argument(
'-min_samples',
help=
'minimum number of samples for peptide usage. 0 means 50%% of input files',
default=0)
parser.add_argument('-fold_change',
help='FC threshold standard deviations',
default=3.0,
type=float)
parser.add_argument(
'-fold_change_abs',
help=
'Use absolute log2 scale FC threshold instead of standard deviations',
action='store_true')
parser.add_argument('-qval',
help='qvalue threshold',
default=0.05,
type=float)
parser.add_argument('-intensity_norm',
help='Intensity normalization: 0-none, 1-median',
default=1,
type=int)
parser.add_argument('-all_proteins',
help='use all proteins instead of FDR controlled',
action='store_true')
parser.add_argument('-all_pfms',
help='use all PFMs instead of ML controlled',
action='store_true')
parser.add_argument('-output_peptides',
help='Add output table with peptides',
action='store_true')
parser.add_argument('-d',
'-db',
help='path to uniprot fasta file for gene annotation')
args = vars(parser.parse_args())
logging.basicConfig(format='%(levelname)9s: %(asctime)s %(message)s',
datefmt='[%H:%M:%S]',
level=logging.INFO)
logger = logging.getLogger(__name__)
replace_label = '_proteins_full.tsv'
fold_change = float(args['fold_change'])
df_final = False
all_s_lbls = {}
allowed_prots = set()
allowed_prots_all = set()
allowed_peptides = set()
for i in range(1, 3, 1):
sample_num = 'S%d' % (i, )
if args[sample_num]:
for z in args[sample_num]:
if not args['all_proteins']:
df0 = pd.read_table(
z.replace('_proteins_full.tsv', '_proteins.tsv'))
allowed_prots.update(df0['dbname'])
allowed_prots.update(
['DECOY_' + z for z in df0['dbname'].values])
else:
df0 = pd.read_table(z)
allowed_prots.update(df0['dbname'])
df0 = pd.read_table(
z.replace('_proteins_full.tsv', '_PFMs_ML.tsv'))
if not args['all_pfms']:
df0 = df0[df0['qpreds'] <= 10]
allowed_peptides.update(df0['seqs'])
logger.info('Total number of TARGET protein GROUPS: %d',
len(allowed_prots) / 2)
for i in range(1, 3, 1):
sample_num = 'S%d' % (i, )
if args.get(sample_num, 0):
for z in args[sample_num]:
df3 = pd.read_table(z.replace(replace_label, '_PFMs.tsv'))
df3 = df3[df3['sequence'].apply(
lambda x: x in allowed_peptides)]
df3_tmp = df3[df3['proteins'].apply(
lambda x: any(z in allowed_prots for z in x.split(';')))]
for dbnames in set(df3_tmp['proteins'].values):
for dbname in dbnames.split(';'):
allowed_prots_all.add(dbname)
for i in range(1, 3, 1):
sample_num = 'S%d' % (i, )
if args.get(sample_num, 0):
all_s_lbls[sample_num] = []
for z in args[sample_num]:
label = z.replace(replace_label, '')
all_s_lbls[sample_num].append(label)
df3 = pd.read_table(z.replace(replace_label, '_PFMs.tsv'))
df3 = df3[df3['sequence'].apply(
lambda x: x in allowed_peptides)]
# allowed_prots2 = set()
# df3_tmp = df3[df3['proteins'].apply(lambda x: any(z in allowed_prots for z in x.split(';')))]
# for dbnames in set(df3_tmp['proteins'].values):
# for dbname in dbnames.split(';'):
# allowed_prots2.add(dbname)
# print('!', len(allowed_prots), len(allowed_prots2))
# df3 = df3[df3['proteins'].apply(lambda x: any(z in allowed_prots for z in x.split(';')))]
df3 = df3[df3['proteins'].apply(lambda x: any(
z in allowed_prots_all for z in x.split(';')))]
df3['proteins'] = df3['proteins'].apply(lambda x: ';'.join(
[z for z in x.split(';') if z in allowed_prots_all]))
### df3['proteins'] = df3['proteins'].apply(lambda x: ';'.join([z for z in x.split(';') if z in allowed_prots]))
df3['origseq'] = df3['sequence']
df3['sequence'] = df3['sequence'] + df3['charge'].astype(
int).astype(str) + df3['ion_mobility'].astype(str)
df3 = df3.sort_values(by='Intensity', ascending=False)
df3 = df3.drop_duplicates(subset='sequence')
df3[label] = df3['Intensity']
df3['protein'] = df3['proteins']
df3['peptide'] = df3['sequence']
df3 = df3[['origseq', 'peptide', 'protein', label]]
if df_final is False:
df_final = df3.reset_index(drop=True)
else:
df_final = df_final.reset_index(drop=True).merge(
df3.reset_index(drop=True), on='peptide', how='outer')
df_final.protein_x.fillna(value=df_final.protein_y,
inplace=True)
df_final.origseq_x.fillna(value=df_final.origseq_y,
inplace=True)
df_final['protein'] = df_final['protein_x']
df_final['origseq'] = df_final['origseq_x']
df_final = df_final.drop(
columns=['protein_x', 'protein_y'])
df_final = df_final.drop(
columns=['origseq_x', 'origseq_y'])
logger.info('Total number of peptide sequences used in quantitation: %d',
len(set(df_final['origseq'])))
# print('Total number of proteins used in quantitation %d' % (len(allowed_prots_all), ))
df_final = df_final.assign(protein=df_final['protein'].str.split(
';')).explode('protein').reset_index(drop=True)
df_final = df_final.set_index('peptide')
df_final['proteins'] = df_final['protein']
df_final = df_final.drop(columns=['protein'])
# cols = df_final.columns.tolist()
cols = [z for z in df_final.columns.tolist() if not z.startswith('mz_')]
cols.remove('proteins')
cols.insert(0, 'proteins')
df_final = df_final[cols]
all_lbls = all_s_lbls['S1'] + all_s_lbls['S2']
df_final_copy = df_final.copy()
custom_min_samples = int(args['min_samples'])
if custom_min_samples == 0:
custom_min_samples = int(len(all_lbls) / 2)
df_final = df_final_copy.copy()
max_missing = len(all_lbls) - custom_min_samples
logger.info('Allowed max number of missing values: %d', max_missing)
df_final['nummissing'] = df_final.isna().sum(axis=1)
df_final['nonmissing'] = df_final['nummissing'] <= max_missing
df_final = df_final[df_final['nonmissing']]
logger.info('Total number of PFMs passed missing values threshold: %d',
len(df_final))
df_final['S2_mean'] = df_final[all_s_lbls['S2']].mean(axis=1)
df_final['S1_mean'] = df_final[all_s_lbls['S1']].mean(axis=1)
df_final['FC_raw'] = np.log2(df_final['S2_mean'] / df_final['S1_mean'])
FC_max = df_final['FC_raw'].max()
FC_min = df_final['FC_raw'].min()
df_final.loc[(pd.isna(df_final['S2_mean'])) &
(~pd.isna(df_final['S1_mean'])), 'FC_raw'] = FC_min
df_final.loc[(~pd.isna(df_final['S2_mean'])) &
(pd.isna(df_final['S1_mean'])), 'FC_raw'] = FC_max
if args['intensity_norm'] == 1:
for cc in all_lbls:
# print(cc, df_final[cc].median())
df_final[cc] = df_final[cc] / df_final[cc].median()
df_final['S2_mean'] = df_final[all_s_lbls['S2']].mean(axis=1)
df_final['S1_mean'] = df_final[all_s_lbls['S1']].mean(axis=1)
for cc in all_lbls:
df_final[cc] = df_final[cc].fillna(df_final[cc].min())
df_final['p-value'] = list(
ttest_ind(df_final[all_s_lbls['S1']].values.astype(float),
df_final[all_s_lbls['S2']].values.astype(float),
axis=1,
nan_policy='omit',
equal_var=True)[1])
df_final['p-value'] = df_final['p-value'].astype(float)
df_final['p-value'] = df_final['p-value'].fillna(1.0)
p_val_threshold = 0.05
df_final['sign'] = df_final['p-value'] <= p_val_threshold
df_final['intensity_median'] = df_final[all_s_lbls['S1'] +
all_s_lbls['S2']].median(axis=1)
df_final['FC'] = np.log2(df_final['S2_mean'] / df_final['S1_mean'])
df_final_for_calib = df_final.copy()
df_final_for_calib = df_final_for_calib[
~pd.isna(df_final_for_calib['S1_mean'])]
df_final_for_calib = df_final_for_calib[
~pd.isna(df_final_for_calib['S2_mean'])]
df_final_for_calib = df_final_for_calib[~df_final_for_calib['sign']]
FC_max = df_final['FC'].max()
FC_min = df_final['FC'].min()
df_final.loc[(pd.isna(df_final['S2_mean'])) &
(~pd.isna(df_final['S1_mean'])), 'FC'] = FC_min
df_final.loc[(~pd.isna(df_final['S2_mean'])) &
(pd.isna(df_final['S1_mean'])), 'FC'] = FC_max
df_final['decoy'] = df_final['proteins'].apply(
lambda x: all(z.startswith('DECOY_') for z in x.split(';')))
from scipy.stats import scoreatpercentile
from scipy.optimize import curve_fit
from scipy import exp
def noisygaus(x, a, x0, sigma, b):
return a * exp(-(x - x0)**2 / (2 * sigma**2)) + b
def calibrate_mass(bwidth, mass_left, mass_right, true_md):
bbins = np.arange(-mass_left, mass_right, bwidth)
H1, b1 = np.histogram(true_md, bins=bbins)
b1 = b1 + bwidth
b1 = b1[:-1]
popt, pcov = curve_fit(noisygaus,
b1,
H1,
p0=[1, np.median(true_md), 1, 1])
mass_shift, mass_sigma = popt[1], abs(popt[2])
return mass_shift, mass_sigma, pcov[0][0]
try:
FC_mean, FC_std, covvalue_cor = calibrate_mass(
0.05, -df_final_for_calib['FC'].min(),
df_final_for_calib['FC'].max(), df_final_for_calib['FC'])
except:
FC_mean, FC_std, covvalue_cor = calibrate_mass(
0.1, -df_final_for_calib['FC'].min(),
df_final_for_calib['FC'].max(), df_final_for_calib['FC'])
# print('df_final_FC', FC_mean, FC_std)
# FC_l = FC_mean-fold_change
# FC_r = FC_mean+fold_change
if not args['fold_change_abs']:
fold_change = FC_std * fold_change
logger.info('Absolute FC threshold = %.2f', fold_change)
FC_l = -fold_change
FC_r = fold_change
df_final['up'] = df_final['sign'] * (df_final['FC'] >= FC_r)
df_final['down'] = df_final['sign'] * (df_final['FC'] <= FC_l)
df_final = df_final.sort_values(by=['nummissing', 'intensity_median'],
ascending=(True, False))
df_final = df_final.drop_duplicates(subset=('origseq', 'proteins'))
up_dict = df_final.groupby('proteins')['up'].sum().to_dict()
down_dict = df_final.groupby('proteins')['down'].sum().to_dict()
####### !!!!!!! #######
df_final['up'] = df_final.apply(lambda x: x['up'] if up_dict.get(
x['proteins'], 0) >= down_dict.get(x['proteins'], 0) else x['down'],
axis=1)
protsN = df_final.groupby('proteins')['up'].count().to_dict()
prots_up = df_final.groupby('proteins')['up'].sum()
N_decoy_total = df_final['decoy'].sum()
upreg_decoy_total = df_final[df_final['decoy']]['up'].sum()
p_up = upreg_decoy_total / N_decoy_total
names_arr = np.array(list(protsN.keys()))
logger.info('Total number of proteins used in quantitation: %d',
sum(not z.startswith('DECOY_') for z in names_arr))
logger.info('Total number of peptides: %d', len(df_final))
logger.info('Total number of decoy peptides: %d', N_decoy_total)
logger.info('Total number of significantly changed decoy peptides: %d',
upreg_decoy_total)
logger.info(
'Probability of random peptide to be significantly changed: %.3f',
p_up)
# print(N_decoy_total, upreg_decoy_total, p_up)
if args['output_peptides']:
df_final.to_csv(path_or_buf=args['out'] + '_quant_peptides.tsv',
sep='\t',
index=False)
v_arr = np.array(list(prots_up.get(k, 0) for k in names_arr))
n_arr = np.array(list(protsN.get(k, 0) for k in names_arr))
all_pvals = calc_sf_all(v_arr, n_arr, p_up)
total_set = set()
total_set_genes = set()
FC_up_dict_basic = df_final.groupby('proteins')['FC'].median().to_dict()
FC_up_dict_raw_basic = df_final.groupby(
'proteins')['FC_raw'].median().to_dict()
df_final = df_final[df_final['up'] > 0]
df_final['bestmissing'] = df_final.groupby(
'proteins')['nummissing'].transform('min')
FC_up_dict = df_final[df_final['bestmissing'] ==
df_final['nummissing']].groupby(
'proteins')['FC'].median().to_dict()
FC_up_dict_raw = df_final[df_final['bestmissing'] ==
df_final['nummissing']].groupby(
'proteins')['FC_raw'].median().to_dict()
# FC_up_dict = df_final.groupby('proteins')['FC'].median().to_dict()
df_out = pd.DataFrame()
df_out['score'] = all_pvals
df_out['dbname'] = names_arr
df_out['FC'] = df_out['dbname'].apply(lambda x: FC_up_dict.get(x))
df_out['FC_raw'] = df_out['dbname'].apply(lambda x: FC_up_dict_raw.get(x))
df_out.loc[pd.isna(df_out['FC']),
'FC'] = df_out.loc[pd.isna(df_out['FC']), 'dbname'].apply(
lambda x: FC_up_dict_basic.get(x))
df_out.loc[pd.isna(df_out['FC_raw']), 'FC_raw'] = df_out.loc[
pd.isna(df_out['FC_raw']),
'dbname'].apply(lambda x: FC_up_dict_raw_basic.get(x))
df_out['v_arr'] = v_arr
df_out['n_arr'] = n_arr
df_out['decoy'] = df_out['dbname'].str.startswith('DECOY_')
df_out = df_out[~df_out['decoy']]
df_out['FC_pass'] = (df_out['FC'].abs() >= fold_change) & (df_out['v_arr']
> 0)
df_out_BH_multiplier = df_out['FC_pass'].sum()
qval_threshold = args['qval']
df_out['p-value'] = 1.0
df_out['BH_pass'] = False
df_out = df_out.sort_values(by='score', ascending=False)
df_out.loc[df_out['FC_pass'], 'BH_threshold'] = -np.log10(
df_out.loc[df_out['FC_pass'], 'score'].rank(ascending=False,
method='max') *
qval_threshold / df_out_BH_multiplier)
df_out.loc[df_out['FC_pass'],
'BH_pass'] = df_out.loc[df_out['FC_pass'],
'score'] > df_out.loc[df_out['FC_pass'],
'BH_threshold']
df_out.loc[df_out['FC_pass'],
'p-value'] = 10**(-df_out.loc[df_out['FC_pass'], 'score'])
score_threshold = df_out[df_out['BH_pass']]['score'].min()
df_out.loc[df_out['FC_pass'],
'BH_pass'] = df_out.loc[df_out['FC_pass'],
'score'] >= score_threshold
df_out.to_csv(path_or_buf=args['out'] + '_quant_full.tsv',
sep='\t',
index=False)
df_out_f = df_out[(df_out['BH_pass']) & (df_out['FC_pass'])]
df_out_f.to_csv(path_or_buf=args['out'] + '.tsv', sep='\t', index=False)
genes_map = {}
if args['d']:
for prot, protseq in fasta.read(args['d']):
try:
prot_name = prot.split('|')[1]
except:
prot_name = prot
try:
gene_name = prot.split('GN=')[1].split(' ')[0]
except:
gene_name = prot
genes_map[prot_name] = gene_name
for z in set(df_out_f['dbname']):
try:
prot_name = z.split('|')[1]
except:
prot_name = z
gene_name = genes_map.get(prot_name, prot_name)
total_set.add(prot_name)
total_set_genes.add(gene_name)
logger.info('Total number of significantly changed proteins: %d',
len(total_set))
logger.info('Total number of significantly changed genes: %d',
len(total_set_genes))
f1 = open(args['out'] + '_proteins_for_stringdb.txt', 'w')
for z in total_set:
f1.write(z + '\n')
f1.close()
f1 = open(args['out'] + '_genes_for_stringdb.txt', 'w')
for z in total_set_genes:
f1.write(z + '\n')
f1.close()
plt.rcParams["font.family"] = "Times New Roman"
parser = argparse.ArgumentParser(description='Use to find kmers')
parser.add_argument('-modifa', type=str, help='modifying_fasta', required=True)
parser.add_argument('-reffa', type=str, help='reference_fasta', required=True)
parser.add_argument('-save_way', type=str, help='path_of_save', required=True)
parser.add_argument('-image_save',
type=str,
help='image_save_path',
required=True)
args = parser.parse_args()
ref_path = args.reffa
mpl.rcParams['figure.figsize'] = [12, 8]
print('Analizing of motive abundance of E.coli ...\n')
genome = read(args.reffa)
syntheny_bloks = read(args.modifa)
alphablet = ['A', 'T', 'G', 'C']
new_genome = ''
for line in genome:
new_genome += line.sequence
def create_lib(r):
return set([''.join(i) for i in permutations(alphablet * r, r=r)])
all_ends = ''
for line in syntheny_bloks:
all_ends += line.sequence
check_list = {}
checker = []
for i in range(len(p)):
if str(p['describtion'][i]) != 'nan':
if str(p['describtion'][i]) != 'Description':
check_list[str(p['describtion'][i])] = {}
checker.append(str(p['describtion'][i]))
else:
break
strains = args.refpa
for sta in strains:
if '.DS_Store' in sta:
continue
op_pa = read(args.refpa + '/All/{}'.format(sta))
for line in op_pa:
for num in check_list:
if num in line.description:
check_list[num] = [sta[:-12]]
last = []
for ind in check_list:
last.append(check_list[ind])
print('Write graph ...')
for seq_id in graph:
graph[seq_id]['blocks'].sort()
blocks = graph[seq_id]['blocks']
for i in range(len(blocks) - 1):
if args.refpa != None:
out.write('sb{}\tsb{}\t{}\t{}\t{}\t{}\n'.format(
ratio_all_kmers = []
ratio_all_GATC = []
all_p_values_kmers = []
all_p_values_GATC = []
for kmer in create_lib(4):
if kmer != 'GATC':
for org in organisms:
cake = ''
end_genome = ''
mid_genome = ''
if '.DS_Store' in org:
continue
print('Analyzin of {} ...'.format(org))
mid_genome = read('{}{}/{}_middle_SB.fasta'.format(
args.orgs, org, org))
for line in mid_genome:
mid_genome = line.sequence
end_genome = read('{}{}/{}_ENDS_SB.fasta'.format(
args.orgs, org, org))
for line in end_genome:
end_genome = line.sequence
print('Counting {} ...'.format(kmer))
lable_k.append(kmer)
mid = mid_genome.count(kmer) * 100 / (len(mid_genome) - 5)
end = end_genome.count(kmer) * 100 / (len(end_genome) - 5)
ratio = end / mid
ratio_all_kmers.append(ratio)
cake += 'Frequency in the ends for every 4-mers\t{}\n'.format(end)
def test_read_and_write_long(self):
with tempfile.TemporaryFile(mode='r+') as new_fasta_file:
fasta.write(fasta.read(self.fasta_file), new_fasta_file)
new_fasta_file.seek(0)
new_entries = list(fasta.read(new_fasta_file))
self.assertEqual(new_entries, self.fasta_entries_long)
def run():
parser = argparse.ArgumentParser(
description='calculate NSAF for scavager results',
epilog='''
Example usage
-------------
$ scav2nsaf -S1 sample1_1_proteins.tsv sample1_n_proteins.tsv -S2 sample2_1_proteins.tsv sample2_n_proteins.tsv
-------------
''',
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
parser.add_argument('-S1',
nargs='+',
metavar='FILE',
help='input files for S1 sample',
required=True)
parser.add_argument('-S2',
nargs='+',
metavar='FILE',
help='input files for S2 sample')
parser.add_argument('-S3',
nargs='+',
metavar='FILE',
help='input files for S3 sample')
parser.add_argument('-S4',
nargs='+',
metavar='FILE',
help='input files for S4 sample')
parser.add_argument('-u',
'--union',
help='pool the files together for the samples',
action='store_true')
group = parser.add_mutually_exclusive_group()
group.add_argument(
'-a',
'--autolabel',
help='in union mode, derive sample labels from common name prefixes',
action='store_true')
group.add_argument(
'--labels',
nargs='+',
metavar='LABEL',
help='labels for samples in union mode (same number as samples)')
parser.add_argument('-db',
metavar='FILE',
help='path to fasta file',
required=True)
parser.add_argument('-out',
metavar='FILE',
help='name of nsaf output file',
default='nsaf_out.txt')
parser.add_argument('-version',
action='version',
version='%s' %
(pkg_resources.require("scavager")[0], ))
args = vars(parser.parse_args())
samples = ['S1', 'S2', 'S3', 'S4']
labels = args['labels'] if args['labels'] else samples
df_final = False
allowed_prots = set()
allowed_peptides = set()
for sample_num in samples:
if args[sample_num]:
for z in args[sample_num]:
df0 = pd.read_csv(z, sep='\t')
allowed_prots.update(df0['dbname'])
for sample_num in samples:
if args[sample_num]:
for z in args[sample_num]:
df0 = pd.read_csv(z.replace('_proteins.tsv', '_peptides.tsv'),
sep='\t')
allowed_peptides.update(df0['peptide'])
for sample_num, label in zip(samples, labels):
if args[sample_num]:
if not args['union']:
for z in args[sample_num]:
df1 = read_table(z, allowed_peptides, allowed_prots)
if df_final is False:
df_final = df1
else:
df_final = df_final.reset_index().merge(
df1.reset_index(), on='peptide',
how='outer') #.set_index('peptide')
df_final.protein_x.fillna(value=df_final.protein_y,
inplace=True)
df_final['protein'] = df_final['protein_x']
df_final = df_final.drop(columns=[
'protein_x', 'protein_y', 'index_x', 'index_y'
])
else:
if args['autolabel']:
label = os.path.commonprefix(args[sample_num]).rstrip('_')
df1 = read_table(args[sample_num],
allowed_peptides,
allowed_prots,
label=label)
if df_final is False:
df_final = df1
else:
df_final = df_final.reset_index().merge(
df1.reset_index(), on='peptide',
how='outer') #.set_index('peptide')
df_final.protein_x.fillna(value=df_final.protein_y,
inplace=True)
df_final['protein'] = df_final['protein_x']
df_final = df_final.drop(columns=[
'protein_x', 'protein_y', 'index_x', 'index_y'
])
df_final = df_final.set_index('peptide')
df_final['proteins'] = df_final['protein']
df_final = df_final.drop(columns=['protein'])
cols = df_final.columns.tolist()
cols.remove('proteins')
cols.insert(0, 'proteins')
df_final = df_final[cols]
df_final.fillna(value='')
cols = df_final.columns.difference(['proteins'])
genres = df_final['proteins'] #.str.split(';')
df_final = (
df_final.loc[df_final.index.repeat(genres.str.len()), cols].assign(
dbname=list(chain.from_iterable(genres.tolist()))))
df_final = df_final.groupby('dbname').sum()
df_final.reset_index(level=0, inplace=True)
protsL = {}
for p in fasta.read(args['db']):
dbn = p[0].split()[0]
protsL[dbn] = len(p[1])
df_final['Length'] = df_final['dbname'].apply(lambda z: protsL[z])
for cc in df_final.columns:
if cc not in ['dbname', 'Length']:
df_final[cc] = df_final[cc] / df_final['Length']
for cc in df_final.columns:
if cc not in ['dbname', 'Length']:
df_final[cc] = df_final[cc] / df_final[cc].sum()
df_final[cc] = df_final[cc].replace(0, np.nan)
min_val = np.nanmin(df_final[cc].values)
df_final[cc] = df_final[cc].replace(np.nan, min_val)
df_final.drop(columns=[
'Length',
], inplace=True)
df_final.to_csv(args['out'], sep='\t', index=False)
#file for getting peptides from other enzymes
##import libraries
import matplotlib.pyplot as plt
import numpy as np
from pyteomics import fasta, parser
print('hello world')
databasePath = 'C:/Users/chris/OneDrive/Documents/bccrc/projectsRepository/sorensenLab/relatedToYbx1/design20200718_multipleProteaseSp3/uniprotHumanJul2020.fasta'
print('Cleaving the proteins with trypsin...')
uniquePeptides = set()
for description, sequence in fasta.read(databasePath):
newPeptides = parser.cleave(sequence, 'K', min_length=6)
uniquePeptides.update(newPeptides)
print('Done, {0} sequences obtained!'.format(len(uniquePeptides)))
import csv
from pyteomics.fasta import read
import matplotlib.pyplot as plt
from scipy.stats import mannwhitneyu
import sys
plt.style.use('seaborn')
parser = argparse.ArgumentParser('Mann-whitney test')
parser.add_argument('-stain', help='Path to organisms', required=True)
parser.add_argument('-coord', help='Path to coordinate', required=True)
parser.add_argument('-save_way', help='Path to save fold', required=True)
args = parser.parse_args()
print('Starting of analyze ...')
fasta = read(args.stain)
coord = open(args.coord)
read_tsv = pd.DataFrame(csv.reader(coord, delimiter="\t"))
read_tsv = read_tsv.T.drop(0)
Methyla_coordinates = []
for ind in range(len(read_tsv)):
if ind == 0:
continue
Methyla_coordinates.append([int(read_tsv[0][ind]), int(read_tsv[1][ind])])
SB_coordinates = []
SBs = read(args.stain)
SB_dict = {}
for line in SBs:
if 'GCF_000005845.2_ASM584v2' in line.description:
# In python 3, use 'from urllib.request import urlretrieve' instead
import gzip
import matplotlib.pyplot as plt
import numpy as np
from pyteomics import fasta, parser, mass, achrom, electrochem, auxiliary
if not os.path.isfile('yeast.fasta.gz'):
print 'Downloading the FASTA file for Saccharomyces cerevisiae...'
urlretrieve(
'ftp://ftp.uniprot.org/pub/databases/uniprot/current_release/'
'knowledgebase/proteomes/YEAST.fasta.gz', 'yeast.fasta.gz')
print 'Done!'
print 'Cleaving the proteins with trypsin...'
unique_peptides = set()
for description, sequence in fasta.read(gzip.open('yeast.fasta.gz')):
new_peptides = parser.cleave(sequence, parser.expasy_rules['trypsin'])
unique_peptides.update(new_peptides)
print 'Done, {0} sequences obtained!'.format(len(unique_peptides))
peptides = [{'sequence': i} for i in unique_peptides]
print 'Parsing peptide sequences...'
for peptide in peptides:
peptide['parsed_sequence'] = parser.parse(peptide['sequence'],
show_unmodified_termini=True)
peptide['length'] = parser.length(peptide['parsed_sequence'])
print 'Done!'
peptides = [peptide for peptide in peptides if peptide['length'] <= 100]
res_df.columns = [
"ID1", "ID2", "mass", "mz(z1)", "mz(z2)", "mz(z3)", "mz(z4)"
]
res_df['ID1'] = res_df['ID1'].astype(int)
res_df['ID2'] = res_df['ID2'].astype(int)
b = time.time()
print(np.round((b - a) / 60., 2))
return (res_df)
#set input params
idname = "test"
fasta_file = "test.fasta"
min_length = 6
protease = "trypsin"
crosslinker_mass = 138.06807961
#read database
target = fasta.read(fasta_file)
decoys = fasta.read(fasta_file)
#generate peptides
target_df = digest(target, "Target", min_length=min_length, protease=protease)
decoy_df = digest(decoys, "Decoy", min_length=min_length, protease=protease)
df = pd.concat([target_df, decoy_df])
df = df.reset_index()
df.to_csv("{}_PeptidesDigest.csv".format(idname))
pairs = generate_peptidepairs(df, crosslinker_mass, idname)
#df_masses = pd.read_csv("test_masses.csv")
arg_parser.add_argument('-m', '--min', type=int, default=6,
help='minimal length of the peptide to report')
arg_parser.add_argument('-M', '--missed', type=int, default=0,
help='number of missed cleavages')
arg_parser.add_argument('-e', '--enz', type=str, default='trypsin',
help='protease to use for digestion')
arg_parser.add_argument('-v', '--verbosity', action='count',
help='increase output verbosity')
args = arg_parser.parse_args()
# TODO: Do it proper way - using os.path
out_file = args.fasta_file + '.peptides'
peptides = []
with fasta.read(args.fasta_file) as reader, open(out_file,'w') as writer:
# Build a set of peptides for each fasta sequence
if args.verbosity >= 1:
print 'Building digests...'
for description, sequence in reader:
peps = parser.cleave(sequence, parser.expasy_rules[args.enz], args.missed)
peps = [x for x in peps if len(x) > args.min]
writer.write('Peptides for {seq} ({enz} cleavage)\n'.format(
seq=description, enz=args.enz))
writer.write('...\t{n} missed cleavages\n'.format(n=args.missed))
writer.write('\n'.join(peps)+'\n')
peptides.append(set(peps))
if args.verbosity >= 2:
print '...\t{n} peptides for {prot}'.format(n=len(peps),prot=description)
orgsnisms = os.listdir(args.orgs)
part = ''
condition = ''
coordinate_of_split = ''
check_list = {}
#Maiking spike
for org in orgsnisms:
if '.DS_Store' in org:
continue
strains = os.listdir('{}{}/All/'.format(args.orgs, org))
for strain in strains:
if '.DS_Store' in strain:
continue
check_list[strain[:-12]] = {}
fna = read('{}{}/All/{}'.format(args.orgs, org, strain))
counter = 0
index = 0
for line in fna:
counter += len(line.sequence)
if index != 0:
check_list[strain[:-12]][line.description.split()[0]] = counter
else:
check_list[strain[:-12]][line.description.split()[0]] = 0
index += 1
for organism in SBs:
if '.DS_store' in organism:
continue
for an_org in coord_prot:
if '.DS_Store' in an_org:
continue
if 'LCB' in miorg:
print('Analyzing {} ...'.format(miorg[4:-6]))
save_file = open(
args.save_way +
'{}_{}&{}.txt'.format(miorg[4:-6], k, k[::-1]), 'w')
elif 'SB' in miorg:
print('Analyzing {} ...'.format(miorg[3:-6]))
save_file = open(
args.save_way +
'{}_{}&{}.txt'.format(miorg[3:-6], k, k[::-1]), 'w')
else:
print('Analyzing ...')
save_file = open(args.save_way, 'w')
fasta = read(args.genome + miorg)
for SB in fasta:
length = len(SB.sequence)
if length > 1000:
rich = (
(SB.sequence.count(k[::-1]) + SB.sequence.count(k)) *
100) / (length - (len(k) - 1))
save_file.write(
split(split(SB.description)[0])[1] + '\t' +
str(len(SB.sequence)) + '\t' + str(rich) + '\n')
save_file.close()
print('Text file was saved!')
else:
for k in mixer(kmer):
if k in controler:
def prot_gen(args):
db = args['d']
with fasta.read(db) as f:
for p in f:
yield p
ref_path = os.listdir(args.reffa)
k = args.k
if args.data_type == 'fold' and k <= 3:
while not k > 3:
if k > 2:
mpl.rcParams['font.size'] = 5
if k > 1:
mpl.rcParams['figure.figsize'] = [25, 8]
else:
mpl.rcParams['figure.figsize'] = [15, 8]
for path in ref_path:
if '.DS_Store' in path:
continue
print('Analizing of {}\n'.format(path))
genome = read(args.reffa + '{}/{}_middle_SB.fasta'.format(path, path))
syntheny_bloks = read(args.modifa + '{}/{}_ENDS_SB.fasta'.format(path, path))
alphablet = ['A', 'T', 'G' ,'C']
mid_genome = ''
for line in genome:
mid_genome += line.sequence
def create_lib(r):
return set([''.join(i) for i in permutations(alphablet * r, r=r)])
all_ends = ''
for line in syntheny_bloks:
all_ends += line.sequence
print('Waiting...\n')
def get_cond_peps(file_in, fasta_file):
cols = ['Sequence', 'Proteins', 'Experiment', 'Intensity (normalized)']
raw, headers, inds = openfile(file_in, cols)
unique_prots = []
unique_conds = []
unique_bio_reps = ['']
count = 0
prot_seqs = []
no_fasta = []
for a in raw:
prots = a[inds[1]].split(';')
count += 1
for b in prots:
prot = b.split('CON__')[-1]
prot_seq = ''
if prot not in unique_prots and [prot] not in no_fasta:
fasta_test = False
# print prot
for description, sequence in fasta.read(fasta_file):
if '|' + prot + '|' in description:
fasta_test = True
print 'Fasta match found:', prot, count, '/', len(raw)
prot_seq = sequence
if fasta_test == True:
unique_prots.append(prot)
prot_seqs.append([prot, prot_seq])
else:
print 'No FASTA match:', prot
no_fasta.append([prot])
if a[inds[2]] not in unique_conds:
unique_conds.append(a[inds[2]])
#if a[inds[4]] not in unique_bio_reps:
# unique_bio_reps.append(a[inds[4]])
headers_out = [
'Protein', 'Experiment', 'Passing Peptides', 'FASTA seq',
'Intensities (normalized)'
]
out = []
for prot in unique_prots:
for cond in unique_conds:
for bio_rep in unique_bio_reps:
pep_list = []
intens_list = []
for a in raw:
if prot in a[inds[1]]:
if cond == a[inds[2]]:
if 1 == 1:
pep_list.append(a[inds[0]])
intens_list.append(a[inds[3]])
pep_str = ''
for i in pep_list:
pep_str += i + ';'
pep_str = pep_str[:-1]
intens_str = ''
for i in intens_list:
intens_str += i + ';'
intens_str = intens_str[:-1]
for b in prot_seqs:
if b[0] == prot:
prot_seq = b[1]
entry = [prot, cond, pep_str, prot_seq, intens_str]
out.append(entry)
print entry
file_out = file_in.split('.txt')[0] + '_passpeps.txt'
savefile(file_out, out, headers_out)
description was replaced by the gene name.
'''
descr, seq = e
m = GN_PATTERN.search(descr)
new_descr = m.group('gene') or descr
return Protein(new_descr, seq)
parser = argparse.ArgumentParser()
parser.add_argument('files', type=str, nargs='+',
help='.fasta files containing plink output')
parser.add_argument('-o', '--output', type=str, default=None,
help='file to write output to.')
parser.add_argument('--gname', action='store_true',
help='only leave gene name (GN) in the output fasta header')
parser.add_argument('-v', '--verbosity', action='count',
help='increase output verbosity')
args = parser.parse_args()
unique_entries = set()
for filename in args.files:
if args.verbosity > 0:
print 'Processing {0}...'.format(filename)
for entry in fasta.read(open(filename)):
if args.gname:
entry = get_gene_name(entry)
unique_entries.update({entry,})
if args.verbosity > 1:
print '\t... found {0} unique entries so far'.format(len(unique_entries))
fasta.write(unique_entries, output=args.output)
