def var_parser(outputfile, emu_out):
'''output: pmid, string, mutation_entry'''
'''mutation integration'''
with open(outputfile, 'r') as f:
mutation_results = [i.split('\n') for i in f.read().split('\n\n') if i]
emu = defaultdict(lambda: defaultdict(str))
for pmid, string, norm in emu_to_tmvar(emu_out):
emu[int(pmid)][string] = norm
for article_result in mutation_results:
text = ''
pmid = False
mutation_entry_ls = []
for record in article_result:
is_text = re.match(r'(\d+)\|a\|(.+)$', record)
if is_text:
# id, string
text = is_text.group(2)
pmid = int(is_text.group(1))
else:
rpmid, start, end, string, Mtype, norm = record.split('\t')
start = int(start)
end = int(end)
if re.search('[A-T]\d+[A-T]', string.upper()):
if re.match('c', norm):
norm = norm.replace('c', 'p', 1)
try:
my_nm = normalization.mutation_string_normalizition(
string, norm)
except ValueError as e:
print('{0}\t{1:10}'.format(e, 'skip'))
continue
mutation_entry_ls.append(
Base.BioEntry(start, end, 'mutation', string, norm, my_nm))
for string in emu[pmid].keys():
emu_nm = emu[pmid][string]
is_exist = 0
for tmvar_mut in mutation_entry_ls:
tm_nm = tmvar_mut.id
if emu_nm in tm_nm or tm_nm in emu_nm: # 重复
is_exist = 1
break
if not is_exist:
'''add mutation entry'''
for start, end in __emu_get_pos(string, text):
my_nm = normalization.mutation_string_normalizition(
string, emu_nm)
mutation_entry_ls.append(
Base.BioEntry(start, end, 'mutation', string, emu_nm,
my_nm))
yield (pmid, text, mutation_entry_ls)
def __GNormPlus_parser_old(outputdir):
'''output: pmid, gene_entry, article_part'''
gnor = normalization.gene_normor('../data/hgnc_complete_set.txt')
for root, dirs, files in os.walk(outputdir):
for fn in files:
with open(os.path.join(root, fn), 'r') as f:
records = [i for i in f.read().split('\n') if i]
for r in records:
is_text = re.match(r'\d+\|a\|(.+)$', r)
if is_text:
continue
else:
pmid, start, end, string, type, ids = r.split('\t')
if not re.match('gene|protein', type, re.IGNORECASE):
continue
pmid = int(pmid)
start = int(start)
end = int(end)
for did in ids.split(';'):
'''one tring map to mutil-gene'''
try:
norm = gnor.norm(did)
except KeyError as e:
print('gene {0} skip: no gene in hgnc'.format(
string))
continue
yield (pmid,
Base.BioEntry(start, end, type, string, id,
norm), fn)
def GNormPlus_parser(outputfile):
gnor = normalization.gene_normor('../data/hgnc_complete_set.txt')
with open(outputfile, 'r') as f:
gene_results = [i.split('\n') for i in f.read().split('\n\n') if i]
for article_result in gene_results:
text = ''
pmid = False
mutation_entry_ls = []
for record in article_result:
is_text = re.match(r'(\d+)\|a\|(.+)$', record)
if is_text:
# id, string
text = is_text.group(2)
pmid = int(is_text.group(1))
else:
pmid, start, end, string, _type, ids = record.split('\t')
if not re.match('gene|protein', _type, re.IGNORECASE):
continue
pmid = int(pmid)
start = int(start)
end = int(end)
for did in ids.split(';'):
'''one tring map to mutil-gene'''
try:
norm = gnor.norm(did)
except KeyError as e:
print('gene {0} skip: no gene in hgnc'.format(string))
continue
yield (pmid,
Base.BioEntry(start, end, _type, string, id, norm))
def DNorm_parser(outputfile):
'''output: pmid, disease_entry'''
with open(outputfile, 'r') as f:
dnorm_results = [i.split('\t') for i in f.read().split('\n') if i]
for i in dnorm_results:
if len(i) == 5:
pmid, start, end, string, id = i
else:
continue
# pmid, start, end, string = i
# id = 'null'
pmid = int(pmid)
start = int(start)
end = int(end)
yield (pmid, Base.BioEntry(start, end, 'disease', string, id))
def GNormPlus_parser(outputdir):
'''output: pmid, gene_entry, article_part'''
for root, dirs, files in os.walk(outputdir):
for fn in files:
with open(os.path.join(root, fn), 'r') as f:
records = [i for i in f.read().split('\n') if i]
for r in records:
is_text = re.match(r'\d+\|a\|(.+)$', r)
if is_text:
continue
else:
pmid, start, end, string, type, id = r.split('\t')
if not re.match('gene|protein', type, re.IGNORECASE):
continue
pmid = int(pmid)
start = int(start)
end = int(end)
yield (pmid, Base.BioEntry(start, end, type, string,
id), fn)
def tmvar_parser(outputfile):
'''output: pmid, string, mutation_entry'''
with open(outputfile, 'r') as f:
mutation_results = [i.split('\n') for i in f.read().split('\n\n') if i]
for article_result in mutation_results:
text = ''
pmid = False
mutation_entry_ls = []
for record in article_result:
is_text = re.match(r'(\d+)\|a\|(.+)$', record)
if is_text:
# id, string
text = is_text.group(2)
pmid = int(is_text.group(1))
else:
rpmid, start, end, string, Mtype, norm = record.split('\t')
start = int(start)
end = int(end)
mutation_entry_ls.append(
Base.BioEntry(start, end, 'mutation', string, norm))
yield (pmid, text, mutation_entry_ls)