def __get_dna(self, ice_id):
'''Gets the sequence ICE entry.'''
url = self.__url + '/rest/file/' + self.__get_ice_number(ice_id) + \
'/sequence/sbol1?sid=' + self.__sid
temp_file = tempfile.NamedTemporaryFile(delete=False)
with codecs.open(temp_file.name, 'w', 'utf-8') as text_file:
text_file.write(net_utils.get(url))
return sbol_utils.read(temp_file.name)
def get_genbank(self, ice_id, out=None):
'''Get Genbank file.'''
url = self.__url + '/rest/file/' + self.__get_ice_number(ice_id) + \
'/sequence/genbank'
genbank = net_utils.get(url, self.__headers)
if out:
with open(out, 'w') as out_file:
out_file.write(genbank)
return genbank
def __get_sbol_doc(self, ice_id):
'''Gets the sequence ICE entry.'''
url = self.__url + '/rest/file/' + self.__get_ice_number(ice_id) + \
'/sequence/sbol?sid=' + self.__sid
temp_file = tempfile.NamedTemporaryFile(delete=False)
with open(temp_file.name, 'w') as text_file:
text_file.write(net_utils.get(url))
document = sbol.Document()
document.read(temp_file.name)
return document
def get_fasta(prefix,
suffix,
entries,
out_file='out.fasta',
organism='4932',
seqs_per_class=500):
'''Get fasta file of protein variants.'''
base_url = 'http://codon.synbiochem.co.uk/codons?codon=%s&organism=%s'
class_seqs = defaultdict(list)
for entry in entries:
seq_class, nucl_seq = entry.split('_')
whole_seq = prefix + nucl_seq + suffix
num_of_codons = len(whole_seq) // 3
codon_indexes = range(0, num_of_codons * 3, 3)
aa_seqs = []
for codon_index in codon_indexes:
codon = whole_seq[codon_index:codon_index + 3]
url = base_url % (codon, organism)
resp = json.loads(net_utils.get(url, headers=None, verify=True))
aas = [
aa['amino_acid'] for aa in resp[0]['amino_acids']
if aa['amino_acid'] != 'Stop'
]
if aa_seqs:
new_aa_seq = []
for aa in aas:
for aa_seq in aa_seqs:
new_aa_seq.append(aa_seq + aa)
aa_seqs = new_aa_seq
else:
aa_seqs = aas
class_seqs[seq_class].extend(aa_seqs)
with open(out_file, 'w') as out:
for seq_class, seqs in class_seqs.items():
seqs = [
choice(seqs) for _ in range(min(len(seqs), seqs_per_class))
]
SeqIO.write([
SeqRecord(Seq(aa_seq, generic_protein), '%s_%i' %
(seq_class, aa_idx + 1), '', '')
for aa_idx, aa_seq in enumerate(seqs)
], out, 'fasta')
def __get_meta_data(self, ice_id):
'''Returns an ICE entry metadata.'''
return _read_resp(net_utils.get(
self.__url + '/rest/parts/' + self.__get_ice_number(ice_id),
self.__headers))
def search(self, term, limit=5):
'''Searches ICE.'''
url = self.__url + '/rest/search?offset=0&limit=' + str(limit) + \
'&sort=relevance&q="' + term + '"'
return _read_resp(net_utils.get(url, self.__headers))
def search_groups(self, term):
'''Gets groups from search terms.'''
url = self.__url + '/rest/groups/autocomplete?token=' + term
return _read_resp(net_utils.get(url, headers=self.__headers))
def get_groups(self):
'''Gets a group name to id map.'''
url = self.__url + '/rest/groups?offset=0&limit=4096'
groups = _read_resp(net_utils.get(url, headers=self.__headers))
return {group['label']: group['id'] for group in groups['data']}
def __get_meta_data(self, ice_id):
'''Returns an ICE entry metadata.'''
return _read_resp(net_utils.get(
self.__url + '/rest/parts/' + self.__get_ice_number(ice_id),
self.__headers))
df = pd.DataFrame(resp['data'])
return df, resp['resultCount']
if __name__ == '__main__':
parser = arguments()
arg = parser.parse_args()
iceurl = 'https://ice.synbiochem.co.uk'
client = ice_utils.ICEClient(iceurl, os.environ['ICE_USERNAME'],
os.environ['ICE_PASSWORD'])
dfs = []
for cole in arg.collection.split(','):
print(cole)
url = iceurl + '/rest/collections/' + cole + '/entries?limit=' + str(
arg.chunk)
response = net_utils.get(url, headers=client._ICEClient__headers)
df, total = response2df(response)
while df.shape[0] < total and df.shape[0] < arg.limit:
print(df.shape[0])
url = iceurl + '/rest/collections/' + cole + '/entries?limit=' + str(
arg.chunk) + '&offset=' + str(df.shape[0])
response = net_utils.get(url, headers=client._ICEClient__headers)
df1, total = response2df(response)
df = pd.concat([df, df1], ignore_index=True, sort=True)
if 'Name' not in df.columns:
# Use both namings to avoid issues
try:
df['Name'] = df['name']
df['Part ID'] = df['partId']
except:
pass