def queryProsite(theSeqs):
print("Currently querying Prosite...")
prositeData = list()
for i in theSeqs:
handle = ScanProsite.scan(seq=i, skip="0")
result = ScanProsite.read(handle)
try:
handle = ExPASy.get_prosite_entry(result[0]["signature_ac"])
res = handle.read()
splitted = res.split("\n")
line = 0
for a in range(0, len(splitted)):
if splitted[a][0:2] == "DE":
line = a
prositeData.append(splitted[line][5:-1])
print(splitted[line][5:-1])
except IndexError:
prositeData.append(None)
print(None)
return prositeData
def Scan_Prosite(entry, sk="off"):
""" Scan_Prosite takes as arg :
entry = Uniprot ID ,PDB or SEQ
skip = (default ="off" ) if "on"
skip patterns and profiles with hight probabilty
returns :
df =(matchs and other features (start,end ,id,score...))
number_of_matchs
csv file corresponding to df
"""
handle = ScanProsite.scan(entry, skip=sk)
#By executing handle.read(), you can obtain the search results in raw XML format. Instead, let’s use
#Bio.ExPASy.ScanProsite.read to parse the raw XML into a Python object:
result = ScanProsite.read(handle)
data = {}
dict_list = [
'sequence_ac', 'start', 'stop', 'signature_ac', 'score', 'level'
]
data.fromkeys(dict_list)
data = {k: [] for k in dict_list}
df = pd.DataFrame(data)
for k in result:
df = df.append(k, ignore_index=True)
number_of_matchs = result.n_match
#df.to_csv("my_prosite_hits.csv")
return (df)
def prosite(self):
seq = input("select sequences id: ")
seq_select = seq.split(' ')
res = []
for iD in seq_select:
handle = ScanProsite.scan(seq=seq_select[iD]['seq'].seq)
res.append(ScanProsite.read(handle))
return res
def download_ProSite_motifs(query: str = 'P12004') -> defaultdict:
"""Performs ExPASy ProSite search for molecular signatures of the query protein. Accepts UniProt ID as a sole argument"""
in_handle = ScanProsite.scan(seq=query)
reader = ScanProsite.read(in_handle)
storage_dict = defaultdict(dict)
for motif in reader:
storage_dict[motif['signature_id']]['start'] = motif['start']
storage_dict[motif['signature_id']]['stop'] = motif['stop']
return storage_dict
def scan_motifs(self, ids_seqs):
'''
Funçao para pesquisa de motivos a partir da Prosite
:param ids_seqs: IDs das sequências do gestor
:return results: resultados do scan
'''
results = []
for id_seq in ids_seqs:
handle = ScanProsite.scan(seq=self.seqs[id_seq]['seq'].seq)
results.append(ScanProsite.read(handle))
return results
def SearchDomains(self,seqid):
if seqid in self.dseqs.keys():
fastaseq = ">SEQ1\n" + str(self.dseqs[seqid].seq)
result_handle = ScanProsite.scan(seq=fastaseq)
result = ScanProsite.read(result_handle)
if seqid in self.domains.keys():
self.domains[seqid].append(result)
else: self.domains[seqid] = [result]
return result
else:
print("Invalid ID")
return False
def Prosite_Domain(self):
from Bio import ExPASy
from Bio.ExPASy import Prosite, ScanProsite
try:
handle = ScanProsite.scan(seq=self.__seq_input)
result = ScanProsite.read(handle)
if len(result) != 0:
for res in range(len(result)):
prosite_acession = result[res]['signature_ac']
r = ExPASy.get_prosite_raw(prosite_acession)
html = Prosite.read(r)
r.close()
print('Foi encontrado um dominio %s.' % (html.name))
else:
print('Não foram encontradas correspondências.')
except:
print('A sequência fornecida não é uma sequência proteica.')
#deleting file
os.remove("prot.fasta")
print(missingProts)
np.save('Round2_protStructure.npy', seq_dict)
####4.Extract kinase domain and ATP binding pocket
kinaseDict = {}
ATPDict = {}
amissing, kmissing, count = 0, 0, 0
for key, sequence in seq_dict.items():
count = count + 1
print(key)
if count % 50 == 0:
time.sleep(60) # sleep 1 mn for very 50 query to avoid timeout
handle = ScanProsite.scan(seq=sequence)
result = ScanProsite.read(handle)
kinase, atp = 0, 0
for i in range(
len(result)
): #I am looping over all results but there should be only one that ha$
if result[i]['signature_ac'] == 'PS50011': # Protein kinase domain
kinaseDict[key] = sequence[result[i]['start']:result[i]['stop']]
kinase = 1
elif result[i]['signature_ac'] == 'PS00107': # ATP binding pocket
ATPDict[key] = sequence[result[i]['start']:result[i]['stop']]
atp = 1
if kinase == 0:
kmissing = kmissing + 1
print('kinase missing')
if atp == 0:
def scanProsite(dic, tag):
handle = ScanProsite.scan(seq=dic[tag].seq, lowscore=1)
result = ScanProsite.read(handle)
for i in range(len(result)):
print(result[i])
from pandas import DataFrame, read_csv
import pandas as pd
file = r'C:/Users/Kevin/Desktop/BIMM182_Project/Sequences.csv'
input = pd.read_csv(file)
motifset = []
for i in range(len(input)):
sequence = input.iloc[i, 2]
accessions = set()
motifs = ""
# Scan Prosite for matching motifs
handle = ScanProsite.scan(seq=sequence, skip="off")
result = ScanProsite.read(handle)
# Obtain all accession motifs
for hit in result:
acc = hit.get('signature_ac')
accessions.add(acc)
# Get descriptions from accession numbers
for accession in accessions:
prof = ExPASy.get_prosite_raw(accession)
text = prof.read()
text = text.splitlines()
desc = text[3]
desc = str.split(desc, 'DE ')
desc = desc[1]
from Bio import SeqIO
from Bio.ExPASy import ScanProsite
aligned_record = SeqIO.parse("/home/nadzhou/SEQs/spike_uniprot.fasta", "fasta")
start_end = []
for record in aligned_record:
prosite_handle = ScanProsite.scan(record.seq)
prosite_result = ScanProsite.read(prosite_handle)
for rec in prosite_result:
start_end.append((rec['start'], rec['stop']))
print(record.seq[rec['start']:rec['stop']])
print()
print(start_end)
def scanSequence(sequence):
handle = ScanProsite.scan(seq=sequence)
return ScanProsite.read(handle)
#!/usr/bin/env python
from __future__ import print_function
import os
from Bio.ExPASy import ScanProsite
if __name__ == "__main__":
with open(os.path.join('data', 'rosalind_prst.txt')) as dataset:
protein_string = dataset.readline().rstrip()
handle = ScanProsite.scan(protein_string)
result = ScanProsite.read(handle)
print(sorted(result, key=lambda x: x['start'])[-1]['signature_ac'])
