def fetchmol(qnmrfinal, qxrayfinal, qartfinal):
"""gets the molecular weight of each hit for each generated ID list and then generates a count
for the coming bar graph. mol count info stored in dicts. The info of what weight each ID is is not recorded,
rather, their existence is recorded if they are within a range of kDa"""
nmrmolrange = np.linspace(2.5, 37.5, 15, endpoint=True)
nmr = {i: 0 for i in nmrmolrange}
xrayartmolrange = np.linspace(20, 280, 14, endpoint=True)
xray = {i: 0 for i in xrayartmolrange}
xray[1000] = 0
art = {i: 0 for i in xrayartmolrange}
art[1000] = 0
for hit in qnmrfinal:
try:
test = pd.get_all_info(hit)
nmr = molcounternmr(test, nmr)
except:
continue
for hit in qxrayfinal:
try:
test = pd.get_all_info(hit)
xray = molcounterxray(test, xray)
except:
continue
for hit in qartfinal:
try:
test = pd.get_all_info(hit)
art = molcounterart(test, art)
except:
continue
print("NMR Mol Weight Tally:" + str(nmr))
print("Xray Mol Weight Tally:" + str(xray))
print("Artifact Mol Weight Tally:" + str(art))
return nmr, xray, art
def extractData(maxDepositionDate, foundStructures):
results = {}
moleculeName = []
entityTitle = ''
for entry in foundStructures:
entityInfo = describe_pdb(entry)
depositionDate = datetime.date(
parser.parse(entityInfo['deposition_date']))
if depositionDate >= maxDepositionDate:
entityStructureID = entityInfo['structureId']
entityTitle = entityInfo['title']
entityExtraInfo = get_all_info(entry)
if int(describe_pdb(entry)['nr_entities']) > 1:
for mol in entityExtraInfo['polymer']:
moleculeName.append(mol.get('macroMolecule').get('@name'))
elif int(describe_pdb(entry)['nr_entities']) == 1:
moleculeName = entityExtraInfo.get('polymer').get(
'macroMolecule').get('@name')
else:
moleculeName = 'No molecule name give'
results[entry] = (moleculeName, entityTitle, str(depositionDate))
return (results)
def fetch_info(self, pdb_ids): for pdb_id in tqdm(pdb_ids, disc="Checking PDB for new structures", unit="Structures "): try: info = get_all_info(pdb_id) enzyClass = info["polymer"]["enzClass"]["@ec"] length = int(info["polymer"]["@length"]) #ADD TO DATA FILE except: pass add_to_checked(pdb_id)
def get_organism(id):
nombres = list()
info = pypdb.get_all_info(id)
info_nombre = info.get('polymer')
#esto es porque a veces pueden tener mas de un organismo asociado
if (isinstance(info_nombre, list)):
for x in range(0, len(info_nombre)):
nombres.append(info.get('polymer')[x].get('Taxonomy').get('@name'))
nombres = list(dict.fromkeys(nombres))
return ' '.join(nombres)
return info_nombre.get('Taxonomy').get('@name')
def MakeTemplateList(fasta):
counter = 0
DataList = []
for file in os.listdir('.'):
if file.endswith('.table'):
source = file[-9:-6]
for line in open(file):
counter += 1
if counter == 2:
headerstring = line
headerstring = headerstring.replace(
'target name', 'target-name')
headerstring = headerstring.replace(
'query name', 'query-name')
headerstring = headerstring.replace(
'description of target', 'description-of-target')
headerstring = headerstring[1:].split()
headerstring.append('UniProtID')
headerstring.append('Source')
headerstring = list(MakeUniq(headerstring))
elif counter >= 4:
if line[0] == '#':
pass
else:
line2 = line.split()
description = ' '.join(line2[18:])
line3 = line2[:18]
line3.append(description)
if source == 'CSA':
line3[0] = line3[0][:6].lower().replace(':', '_')
PDBinfo = pypdb.get_all_info(line3[0][:4])['polymer']
if isinstance(PDBinfo, list):
UniProtID = PDBinfo[0]['macroMolecule'][
'accession']['@id']
else:
UniProtID = PDBinfo['macroMolecule']['accession'][
'@id']
line3.append(UniProtID)
line3.append(source)
DataList.append(line3)
df = pd.DataFrame(DataList, columns=headerstring)
df['E-value'] = df['E-value'].astype(float)
df.sort_values(by=['Source', 'E-value'], inplace=True)
df.to_csv('HmmerInfo/CSV/' + fasta + '_fullhmmer.csv')
df.drop_duplicates(subset='UniProtID', inplace=True)
df.to_csv('HmmerInfo/CSV/' + fasta + '_noduphmmer.csv')
targetlist = df['target-name'].tolist()
os.system('mv ' + fasta + '.MCSA.table HmmerInfo/MCSAhm')
os.system('mv ' + fasta + '.PDB.table HmmerInfo/PDBhm')
return targetlist
def fetch_and_name_chains(cif_id):
'''
DESCRIPTION
1. Fecth a cif structure from the pdb
2. Rename the cif file accoring to its 4-letter PDB-code followed by its Genus species name
3. Create a PyMOL object for each of its chains and name them according to whats on the PDB
'''
polymers = pypdb.get_all_info(cif_id)['polymer']
taxonomy = set()
chain_to_name = {}
for poly in range(len(polymers)):
taxonomy.add(polymers[poly]['Taxonomy']['@name'])
polymer_description = polymers[poly]['polymerDescription'][
'@description'].split(',')[0]
entry_name = re.sub('\(|\)', '', polymer_description)
chain = polymers[poly]['chain']['@id']
name = re.sub('ribosomal\s|protein\s|60S\s|40S\s|subunit\s|\sprotein',
'',
entry_name,
flags=re.IGNORECASE)
chain_to_name[name] = chain
if len(taxonomy) > 1:
print('multiple species in structure')
exit
for organism in taxonomy:
Genus_species = organism.replace(' ', '_')
ge_sp = organism.split()
GeSp = ge_sp[0][:2] + ge_sp[1][0].capitalize() + ge_sp[1][1]
cmd.fetch(cif_id, cif_id + '_' + GeSp)
os.rename(cif_id + '.cif', cif_id + '_' + Genus_species + '.cif')
# os.remove(cif_id+'.cif') # Not sure why this isn't working
for obj_name in chain_to_name:
cmd.create(
GeSp + '_' + cif_id + '_' + obj_name,
cif_id + '_' + GeSp + ' and chain ' + chain_to_name[obj_name])
cmd.show_as('cartoon')
def _get_edge_cluster(pdb,pdb_names,percent):
edges = []
# get info and chain labes
polymer = pypdb.get_all_info(pdb)['polymer']
chain_labels = []
for p in polymer:
chain = p['chain']
if not isinstance(chain,list):
chain = [chain]
for c in chain:
chain_labels.append(c['@id'])
# get all the neighbors for the all the chains
for chain in chain_labels:
check, niter = False, 0
# get the cluster
while not check and niter < 10:
try:
cluster,check = pypdb.get_seq_cluster_percent(pdb+'.'+chain, percent=percent)
cluster = cluster['pdbChain']
except Exception as e:
print(str(e))
print('Request failed for %s.%s -> Trying again' %(pdb,chain))
niter += 1
# add the (pdb,pdbneighbor) to the edge list
if len(cluster)>0:
if not isinstance(cluster,list):
cluster = [cluster]
for n in cluster:
pdbid = n['@name'].split('.')[0]
# make sure the neighbor is in the pdb_names
if pdbid in pdb_names:
edges.append((pdb,pdbid))
return edges
def screen_pdb(pdb, dict_cond=None):
if dict_cond is None:
dict_cond = {
'method': 'xray',
'resolution': np.float('Inf'),
'number_of_entity': 2,
'type': ['protein'],
'len_min': 50,
'len_max': 5000
}
check = True
# info = pypdb.get_entity_info(pdb)
info = pypdb.get_all_info(pdb)['rcsb_entry_info']
# method
check *= info['experimental_method'] == dict_cond['method']
if not check:
reason = 'Incorrect Method : %s' % info['experimental_method']
return check, reason
if 'diffrn_resolution_high' not in info:
info['diffrn_resolution_high'] = {
'provenance_source': None,
'value': np.Inf
}
check *= float(
info['diffrn_resolution_high']['value']) <= dict_cond['resolution']
if not check:
reason = 'Low Resolution (%1.2f)' % float(info['resolution'])
return check, reason
# number of entity
# entity = _make_list(info['Entity'])
# check *= len(entity) == dict_cond['number_of_entity']
# if not check:
# reason = 'Wrong number of entitites %d' % len(entity)
# return check, reason
# number/type of chain
types = info['selected_polymer_entity_types']
check *= types in dict_cond['type']
if not check:
reason = 'Incorrect chain Type %s' % types
return check, reason
# chain = _make_list(e['Chain'])
# check *= len(chain) == bioAss
# if not check:
# reason = 'Incorrect Number of Chain %d/%d' % (
# len(chain), bioAss)
# return check, reason
# lentgth
l = info['deposited_polymer_monomer_count']
check *= (l >= dict_cond['len_min'] and l <= dict_cond['len_max'])
if not check:
reason = 'Incorrect chain length %d' % l
return check, reason
return check, 'Entries accepted'
"""this code is showing two methods to establish the relations between uniprotID and pdb chainID"""
import os ##for directory
import pypdb
import pprint
#First method
os.chdir('/Users/luho/PycharmProjects/pdb/code')
all_info = pypdb.get_all_info('6fai')
pprint.pprint(all_info)
#second method
#download all the Structure-chainID-UniprotID mapping from http://www.ebi.ac.uk/pdbe/docs/sifts/
for i in tqdm(checked_list):
if i in pdb_ids:
pdb_ids.remove(i)
for pdb_id in tqdm(pdb_ids):
pdb_info = describe_pdb(pdb_id)
AAs = int(pdb_info["nr_residues"])
if AAs <= 400:
if AAs > 200:
info = get_all_info(pdb_id)
try:
enzyClass = info["polymer"]["enzClass"]["@ec"]
ec.write("{}, ".format(pdb_id))
except:
pass
else:
pass
else:
pass
checked.write("{}, ".format(pdb_id))
def read_list(path):
"Reads list stored in txt file"
list_enzymes = open(path,'r')
def get_protein_cluster_graph(cluster, percent):
edges, nodes, dict_chains = {}, {}, {}
Edge = namedtuple('Edge', ['weight', 'txt'])
Node = namedtuple('Node', ['number', 'txt'])
pdbid = None
for pdb in tqdm(cluster):
print_id(pdb, pdb, pdbid)
# get the polymer infos
check, niter = True, 0
while check and niter < 10:
try:
polymer = pypdb.get_all_info(pdb)['polymer']
check = False
except:
print('PDBUnique -> Issue getting info for :', pdb)
print('PDBUnique -> Trying again in 5 sec')
time.sleep(5)
niter += 1
if check:
print('PDBUnique -> Entry %s ignored' % pdb)
continue
# get the chain labels
chain_labels, chain_entity = [], []
for ip, p in enumerate(polymer):
chain = p['chain']
# only conserve the first chain
if isinstance(chain, list):
chain = chain[0]
chain_labels.append(chain['@id'])
chain_entity.append(ip)
# init the names
names = [None] * len(chain_labels)
print_id(chain, pdb, pdbid)
nup = 0
# enumerate chans
for ic, (chain, ip) in enumerate(zip(chain_labels, chain_entity)):
# pdb.chain ID
id_chain = pdb + '.' + chain
# add the pdb.chain ID to the general dict
# {pdb.chain: prot name}
if id_chain not in dict_chains:
# use the macromolecule or polymer description name
for name_option, tag in zip(
['polymerDescription', 'macroMolecule'],
['@description', '@name']):
if name_option in polymer[ip]:
if isinstance(polymer[ip][name_option], list):
names[ic] = polymer[ip][name_option][0][tag]
else:
names[ic] = polymer[ip][name_option][tag]
break
if names[ic] == 'Uncharacterized Protein':
names[ic] = 'UP_%03d' % nup
nup += 1
# add the pdb.chain to the dict
dict_chains[id_chain] = names[ic]
# get the seq similarity of the chain
check, niter = True, 0
while check and niter < 10:
try:
cluster, _ = pypdb.get_seq_cluster_percent(
id_chain, percent=percent)
check = False
except:
print('PDBUnique -> Issue getting cluster for :',
id_chain)
print('PDBUnique -> Trying again in 5 sec')
time.sleep(5)
niter += 1
if check:
print('PDBUnique -> Entry %s ignored' % id_chain)
cluster = []
else:
cluster = cluster['pdbChain']
print_id(cluster, pdb, pdbid)
# add all the chains with similar seq
# to the dict_chain {pdb.chain: prot_name}
if len(cluster) > 0:
if not isinstance(cluster, list):
cluster = [cluster]
for n in cluster:
dict_chains[n['@name']] = names[ic]
# reuse the previously defined entry
else:
names[ic] = dict_chains[id_chain]
# add the node to the dict of Node namedtuples
key = names[ic]
print_id(key, pdb, pdbid)
if key not in nodes:
nodes[key] = Node(number=1, txt=id_chain)
else:
nodes[key] = nodes[key]._replace(number=nodes[key].number +
1)
if nodes[key].number < 35:
nodes[key] = nodes[key]._replace(txt=nodes[key].txt +
'<br>' + id_chain)
elif nodes[key] == 35:
nodes[key] = nodes[key]._replace(txt=nodes[key].txt +
'<br>' + '...')
# add the edge to the dict of Edge namedtuples
names.sort()
key = tuple(names)
print_id(key, pdb, pdbid)
if key not in edges:
edges[key] = Edge(weight=1, txt=pdb)
else:
edges[key] = edges[key]._replace(weight=edges[key].weight + 1)
edges[key] = edges[key]._replace(txt=edges[key].txt + '<br>' +
pdb)
# Create the graph
g = nx.Graph()
for node_key, node_val in nodes.items():
g.add_node(node_key, number=node_val.number, txt=node_val.txt)
for edge_key, edge_val in edges.items():
g.add_edge(edge_key[0],
edge_key[1],
weight=edge_val.weight,
txt=edge_val.txt)
return g
def _get_pdb_length(pdb):
polymer = pypdb.get_all_info(pdb)['polymer']
l = []
for p in polymer:
l.append(float(p['@length']))
return l