def pathwayInfo(code):
# Function to get info about a pathway, from the code
# Intialize searcher
kSearcher = KEGG()
# Get result and parse it in a dictionnary
result = kSearcher.get(code)
# Add code at the begining of the list
dictResult = kSearcher.parse(result)
# Initialize an empty list
pathwayList = []
# If name exist as a key in dictionnary, else 'NA' insted
pathwayList.append(code)
if 'NAME' in dictResult.keys():
# If pathway name is a string comma separated, replace comma by semicolon
# Fix to avoid wrong column formating at the end of the script
nameStr = str(dictResult['NAME'][0].replace(',', ';'))
pathwayList.append(nameStr)
else:
pathwayList.append('NA')
# If class exist as a key in dictionnary, else 'NA' instead
if 'CLASS' in dictResult.keys():
# If pathway name is a string comma separated, replace comma by semicolon
# Fix to avoid wrong column formating at the end of the script
classStr = str(dictResult['CLASS']).replace(',', ';')
pathwayList.append(classStr)
else:
pathwayList.append('NA')
return pathwayList
def get_kegg_info(stId):
"""
Get kegg dict by pathway id.
"""
k = KEGG()
data = k.get(stId)
dict_data = k.parse(data)
return dict_data
def get_genes_from_kegg_pathway(pathway):
from bioservices.kegg import KEGG
k = KEGG()
k.organism = 'hsa'
pathway = k.get(pathway)
genes = k.parse(pathway)['GENE']
entrez, symbol = zip(*[i.split(' ') for i in genes])
return symbol
def get_single_compound_metadata_online(compound_id):
if compound_id.upper().startswith('C'):
s = KEGG()
res = s.get(compound_id)
return s.parse(res)
else:
ch = ChEBI()
res = ch.getCompleteEntity('CHEBI:'+compound_id)
return res
def extract_orthologs(filename):
'''
Create dictionnary with keggid as key and list of orthologs as value
arg: csv with keggids
return : dict with orthologs
'''
orthos_dict = {}
k = KEGG()
#get list of gammaproteobacteria from csv
df = pd.read_csv(filename, sep="\t", tupleize_cols=1)
df_gamma = pd.read_csv('gammaproteo.csv', sep="\t", tupleize_cols=1)
gamma_list = df_gamma['KEGG'].tolist()
#loop through keggid to get orthologs
for keggid in df['kegg_id']:
if keggid == "no":
continue
print(str(keggid))
ortho_list = []
#get orthologs on kegg
data = k.get(keggid)
dict_data = k.parse(data)
if isinstance(dict_data, int):
continue
#loop through kegg orthologs data and verify that organisms are gammaproteobacteria
for key, value in dict_data['GENES'].items():
if key.lower() in gamma_list:
# print(key.lower(), value.split('(')[0].split())
para_num = len(value.split('(')[0].split())
para_list = []
for i in range(0, para_num):
#print(value.split('(')[0].split()[i])
para_list.append(key.lower() + ":" +
value.split('(')[0].split()[i])
ortho_list.append(para_list)
orthos_dict[keggid] = ortho_list
return orthos_dict
def id2seq(self, hsa):
s = KEGG()
d = s.get(hsa)
dict_d = s.parse(d)
pattern = re.compile(r'\s+')
try:
seq = re.sub(pattern, '', dict_d['AASEQ'])
except:
seq = ''
#print('SEQ:', seq)
text_file = open("dummy.txt", "w")
text_file.write('>' + str(hsa) + '\n' + seq)
text_file.close()
return None
def get_metabs(KEGG, reac_id):
subs_list = []
prod_list = []
# Get reaction data from KEGG using a KEGG reaction ID
r_data = KEGG.get(reac_id)
# Parse the information retrieved
r_parsed = KEGG.parse(r_data)
# Split the equation into substrates and products
split_eq = re.split('<=>', r_parsed['EQUATION'])
# Remove the plus signs between the metabolites
subs_list = [s.strip() for s in split_eq[0].split('+')]
prod_list = [p.strip() for p in split_eq[1].split('+')]
return [subs_list, prod_list]
def get_compound_metadata_online(kegg_ids):
s = KEGG()
metadata_map = {}
for i in range(len(kegg_ids)):
try:
if i % 10 == 0:
print("Retrieving %d/%d KEGG records" % (i, len(kegg_ids)))
kegg_id = kegg_ids[i]
res = s.get(kegg_id)
d = s.parse(res)
first_name = d['NAME'][0]
first_name = first_name.replace(';', '') # strip last ';' character
metadata_map[kegg_id] = {'display_name': first_name}
except TypeError:
print('kegg_id=%s parsed_data=%s' % (kegg_id, d))
return metadata_map
def get_seq(filename):
'''
Create dictionnary with species as keys and sequences as values for an alignment
arg: filename with gene name
return: organism dictionnary with sequences
'''
k = KEGG()
records = list(SeqIO.parse(os.path.join('alignments_nogaps/',filename), "fasta"))
idlist = []
orglist = []
seqlist = []
orgdict = {}
#go through sequences and search for organism name on kegg
for record in records:
idsplit = (record.id).split('_',1)
id = idsplit[0] + ':' + idsplit[1]
handle = k.get(id)
if isinstance( handle, int ):
print(id)
continue
org = k.parse(handle)['ORGANISM']
org = org.split()
org = org[1] +" "+ org[2]
seqlist.append(list(str(record.seq)))
orglist.append(org)
idlist.append(id)
duplist = set(orglist)
# create dict with organism as key and sequences for organism as values
for org in duplist:
indices = [i for i, x in enumerate(orglist) if x == org]
seqs = []
for e in indices:
seqs.append(seqlist[e])
orgdict[org] = seqs
#print(orgdict)
return orgdict
def queryKegg(theIDs):
print("Currently querying KEGG...")
k = KEGG()
keggData = list()
IDlist = list()
for id in theIDs:
ids = id[3:]
query = k.find("acb", ids)
query = query.split('\t')
finalQuery = query[0]
data = k.get(finalQuery)
dictData = k.parse(data)
keggData.append(dictData)
IDlist.append(ids)
return keggData, IDlist
def main():
# Start KEGG interface
k = KEGG()
# Create a dict to store final result
data = dict()
# Read in KEGG gene ID & gene symbol pairs
with open("hsa_gene_list.json", "r") as g:
gene_data = json.load(g)
for gene in gene_data.keys():
print gene
g_data = k.get(gene)
g_prsd = k.parse(g_data)
data[gene] = g_prsd
with open('ginfo.json', 'w') as fw:
json.dump(data, fw)
def get_reaction_ECs_from_kegg(self):
self.reaction_ECs = defaultdict(set)
kegg = KEGG()
for r in self.model.reactions:
ECs = []
try:
reacts = r.split(" ")
for i in reacts:
if i not in self.reaction_ECs:
print("KEGG reaction", i)
ECs += kegg.parse(kegg.get(i))['ENZYME']
for e in ECs:
self.reaction_ECs[i].add(e)
except Exception as inst:
print(inst)
#for e in ECs:
# self.reaction_ECs[r].add(e)
print("EC data loaded from KEGG")
def mapSpecies(mousepeptrackfilename):
#increase the field size of CSV
csv.field_size_limit(int(ctypes.c_ulong(-1).value // 2))
uniproturl = 'https://www.uniprot.org/uploadlists/'
RETRY_TIME = 20.0
mdf= pd.read_csv(mousepeptrackfilename, delimiter='\t')
mouseGenes=list(mdf['Gene'].unique())
mouseGenes=[g for g in mouseGenes if str(g) !='nan']
mousehumandic={}
for gx in range(0,len(mouseGenes),1000):
genecodes=' '.join(mouseGenes[gx:gx+1000])
geneuniprotparams = {
'from':'GENENAME',
'to':'ACC',
'format':'tab',
'query':genecodes,
'columns':'id,genes(PREFERRED),organism-id,reviewed'
}
while True:
try:
geneuniprotdata = urllib.urlencode(geneuniprotparams)
geneuniprotrequest = urllib2.Request(uniproturl, geneuniprotdata)
geneuniprotresponse = urllib2.urlopen(geneuniprotrequest)
for guniprotline in geneuniprotresponse:
gudata=guniprotline.strip()
if not gudata.startswith("Entry"):
guinfo=gudata.split("\t")
if '9606' == guinfo[2].lower() and 'reviewed' == guinfo[3].lower() and guinfo[-1].lower() ==guinfo[1].lower() and len(guinfo[0].strip())>1:
mousehumandic[guinfo[-1].strip()]=guinfo[0].strip()
break
except urllib2.HTTPError:
time.sleep(RETRY_TIME)
print ('Hey, I am trying again until succeeds to get data from uniprot data!',str(datetime.datetime.now()))
except httplib.BadStatusLine:
time.sleep(RETRY_TIME)
print ('Hey, I am trying again until succeeds to get data from uniprot data!',str(datetime.datetime.now()))
colname=['UniProtKB Accession','Protein','Gene','Organism','Peptide Sequence','Summary Concentration Range Data','All Concentration Range Data','All Concentration Range Data-Sample LLOQ Based','Peptide ID',\
'Special Residues','Molecular Weight','GRAVY Score','Transitions','Retention Time','Analytical inofrmation',\
'Gradients','AAA Concentration','CZE Purity','Panel','Knockout','LLOQ','ULOQ','Sample LLOQ','Protocol','Trypsin','QC. Conc. Data','Human UniProtKB Accession']
finalresult=[]
finalresult.append(colname)
humanUniprotID=[]
with open(mousepeptrackfilename) as csvfile:
reader = csv.DictReader(csvfile, delimiter='\t')
for row in reader:
templist=[]
for i in colname[:-1]:
tempdata=str(row[i]).strip()
templist.append(tempdata)
if len(str(templist[2]).strip())>0:
if templist[2] in mousehumandic:
huUniId=mousehumandic[templist[2]]
humanUniprotID.append(huUniId)
templist.append(huUniId)
else:
templist.append('NA')
finalresult.append(templist)
with open(mousepeptrackfilename,'wb') as pf:
pwriter =csv.writer(pf,delimiter='\t')
pwriter.writerows(finalresult)
unqhumanUniprotID=list(set(humanUniprotID))
humanUniprotfuncinfodic={}
countProt=0
for subcode in unqhumanUniprotID:
time.sleep(2)
drugbanklist=[]
PN='NA'
GN='NA'
OG='NA'
OGID='NA'
dislist=[]
GoIDList=[]
GoNamList=[]
GoTermList=[]
try:
countProt+=1
if countProt%1000 ==0:
print str(countProt), "th protein Protein Name, Gene, Organism Name,drug bank data,disease data job starts",str(datetime.datetime.now())
SGrequestURL="https://www.uniprot.org/uniprot/"+str(subcode)+".xml"
SGunifile=urllib.urlopen(SGrequestURL)
SGunidata= SGunifile.read()
SGunifile.close()
try:
SGunidata=minidom.parseString(SGunidata)
try:
drugdata=(SGunidata.getElementsByTagName('dbReference'))
for duItem in drugdata:
if (duItem.attributes['type'].value).upper() == 'DRUGBANK':
try:
drugname=(str(duItem.getElementsByTagName('property')[0].attributes['value'].value).strip())
drugid=str(duItem.attributes['id'].value).strip()
durl='<a target="_blank" href="https://www.drugbank.ca/drugs/'+drugid+'">'+drugname+'</a>'
drugbanklist.append(durl)
except:
pass
if (duItem.attributes['type'].value).strip() == 'NCBI Taxonomy':
try:
OGID=str(duItem.attributes['id'].value).strip()
except:
pass
except IndexError:
pass
try:
godata=(SGunidata.getElementsByTagName('dbReference'))
for gItem in godata:
if (gItem.attributes['type'].value).upper() == 'GO':
try:
gonamedetails=(str(gItem.getElementsByTagName('property')[0].attributes['value'].value).strip()).split(':')[1]
gotermdetails=(str(gItem.getElementsByTagName('property')[0].attributes['value'].value).strip()).split(':')[0]
GoNamList.append(gonamedetails)
goid=str(gItem.attributes['id'].value).strip()
GoIDList.append(goid)
if gotermdetails.lower()=='p':
GoTermList.append('Biological Process')
if gotermdetails.lower()=='f':
GoTermList.append('Molecular Function')
if gotermdetails.lower()=='c':
GoTermList.append('Cellular Component')
except:
pass
if (gItem.attributes['type'].value).strip() == 'NCBI Taxonomy':
try:
OGID=str(gItem.attributes['id'].value).strip()
except:
pass
except IndexError:
pass
try:
try:
PN=(((SGunidata.getElementsByTagName('protein')[0]).getElementsByTagName('recommendedName')[0]).getElementsByTagName('fullName')[0]).firstChild.nodeValue
except:
PN=(((SGunidata.getElementsByTagName('protein')[0]).getElementsByTagName('submittedName')[0]).getElementsByTagName('fullName')[0]).firstChild.nodeValue
except IndexError:
pass
try:
try:
GN=((SGunidata.getElementsByTagName('gene')[0]).getElementsByTagName('name')[0]).firstChild.nodeValue
except:
GN='NA'
except IndexError:
pass
try:
try:
OG=((SGunidata.getElementsByTagName('organism')[0]).getElementsByTagName('name')[0]).firstChild.nodeValue
except:
OG='NA'
except IndexError:
pass
try:
disdata=SGunidata.getElementsByTagName('disease')
for dItem in disdata:
disname=''
disshort=''
try:
disname=(dItem.getElementsByTagName('name')[0]).firstChild.nodeValue
except:
pass
try:
disshort=(dItem.getElementsByTagName('acronym')[0]).firstChild.nodeValue
except:
pass
if len(disname.strip())>0:
dislist.append(str(disname.strip())+'('+str(disshort)+')')
except IndexError:
pass
except ExpatError:
pass
except IOError:
pass
drugbankdata='NA'
disdata='NA'
goiddata='NA'
gonamedata='NA'
gotermdata='NA'
if len(GoIDList)>0:
goiddata='|'.join(list(set(GoIDList)))
if len(GoNamList)>0:
gonamedata='|'.join(list(set(GoNamList)))
if len(GoTermList)>0:
gotermdata='|'.join(list(set(GoTermList)))
if len(drugbanklist)>0:
drugbankdata='|'.join(list(set(drugbanklist)))
if len(dislist)>0:
disdata='|'.join(list(set(dislist)))
humanUniprotfuncinfodic[subcode]=[PN,GN,OG,OGID,disdata,drugbankdata,goiddata,gonamedata,gotermdata]
hudicfile='humanUniprotfuncinfodic.obj'
hudicf = open(hudicfile, 'wb')
pickle.dump(humanUniprotfuncinfodic, hudicf , pickle.HIGHEST_PROTOCOL)
hudicf.close()
print ("Extracting KEGG pathway name, job starts",str(datetime.datetime.now()))
hkeggdictfile={}
hk = KEGG()
for hkx in range(0,len(unqhumanUniprotID),2000):
countProt+=hkx+2000
if countProt%2000 ==0:
print (str(countProt), "th protein kegg job starts",str(datetime.datetime.now()))
huniprotcodes=' '.join(unqhumanUniprotID[hkx:hkx+2000])
huniprotparams = {
'from':'ACC',
'to':'KEGG_ID',
'format':'tab',
'query':huniprotcodes
}
while True:
try:
hkuniprotdata = urllib.urlencode(huniprotparams)
hkuniprotrequest = urllib2.Request(uniproturl, hkuniprotdata)
hkuniprotresponse = urllib2.urlopen(hkuniprotrequest)
for hkuniprotline in hkuniprotresponse:
hkudata=hkuniprotline.strip()
if not hkudata.startswith("From"):
hkuinfo=hkudata.split("\t")
if len(hkuinfo[1].strip()):
hkegg=hk.get(hkuinfo[1].strip())
hkudict_data = hk.parse(hkegg)
try:
try:
if len(str(hkuinfo[0]).strip()) >5:
hkeggdictfile[hkuinfo[0].strip()]=hkudict_data['PATHWAY'].values()
except TypeError:
pass
except KeyError:
pass
break
except urllib2.HTTPError:
time.sleep(RETRY_TIME)
print ('Hey, I am trying again until succeeds to get data from KEGG!',str(datetime.datetime.now()))
pass
hkdicfile='humankeggdic.obj'
hkdicf = open(hkdicfile, 'wb')
pickle.dump(hkeggdictfile, hkdicf , pickle.HIGHEST_PROTOCOL)
hkdicf.close()
from bioservices.kegg import KEGG
k = KEGG()
path = k.get("K00855")
kdict = k.parse(path)
print(kdict)
help(kdict)
with open("play.out", "wt") as result:
result.write("\n".join(kdict.keys()))
ko
) #here append may be used, bc only one item is added at a time
else:
EC2KO_dic[item] = [ko]
else:
KO_dic[ko] = [definition]
#here a list of KOs is made to be searched for - if pathway is defined
kostoget = {}
if args.pathway != 'none':
output = open('eclist.txt', 'w')
kegg = KEGG()
pathway = kegg.get(args.pathway)
dict_data = kegg.parse(pathway)
if dict_data == 404:
print("WARNING: BAD PATHWAY SUBMITTED TO KEGG!")
elif dict_data == None:
print("WARNING: ERROR CONNECTING TO KEGG SERVER!")
#print(dict_data)
print("ECQUERY: PROCESSING PATHWAY/MODULE")
try:
for key in dict_data['ORTHOLOGY'].keys():
print("adding ortholog {} to eclist".format(key))
value = dict_data['ORTHOLOGY'][key]
print(value)
if "," in key:
for item in key.split(","):
kostoget[item] = value
output.write(item + '\n')
from bioservices.kegg import KEGG
kegg = KEGG()
pathway = kegg.get("ko01230")
dict_data = kegg.parse(pathway)
print(dict_data)
output = open("modules.txt", "w")
modules_dict = {}
for key in dict_data['MODULE'].keys():
pathway = kegg.get(key)
module_data = kegg.parse(pathway)
#print(module_data)
orthologs = []
for ortholog in module_data['ORTHOLOGY'].keys():
data = [ortholog, module_data['ORTHOLOGY'][ortholog]]
orthologs.append("_".join(data))
modules_dict[ortholog] = key
output.write('{}\t{}\t{}\t{}\n'.format(key, module_data['NAME'],
module_data['DEFINITION'],
"//".join(orthologs)))
def enzymeInfo(code, ignored, stats, verbosity):
# Function to get info about an enzyme, from the code
# This function return a double list
# Intialize KEGG searcher
kSearch = KEGG(verbose=verbosity)
# Get result and parse it in a dictionnary
print(f"[+] Get info about enzyme {code}")
result = kSearch.get(code)
# If KEGG return an int, the enzyme code doesn't match in databases
if type(result) is int:
return False
else:
dictResult = kSearch.parse(result)
# Create prefix list, info about enzyme herself
prefixList = []
# Add code at the begining of the list
prefixList.append(code)
# If name is present as key, else 'NA' insted
if 'NAME' in dictResult.keys():
#prefixList.append(dictResult['NAME'])
# Convert names from list into a string
# with strop '[]' part, and replace initial separator , by ;
namesStr = str(dictResult['NAME']).strip("'[]'").replace(',', ';')
prefixList.append(namesStr)
else:
prefixList.append('NA')
# If definition is present as key, else 'NA' insted
if 'DEFINITION' in dictResult.keys():
# If definition is a string comma separated, replace comma by semicolon
# Fix to avoid wrong column formating at the end of the script
definitionStr = str(dictResult['DEFINITION']).replace(',', ';')
prefixList.append(definitionStr)
else:
prefixList.append('NA')
# If pathway exist as a key in result
if 'PATHWAY' in dictResult.keys():
# Get all pathways as keys in dictionnary
pathwayList = list(dictResult['PATHWAY'].keys())
# Create final list, which contain :
# - prefix (info about enzyme)
# - suffix list (info about each enzyme's pathways)
finalList = []
# Add suffix in final list
finalList.append(prefixList)
for pathway in pathwayList:
# If pathway not in ignored list
if pathway not in ignored:
print(f" [-] Get info about {pathway} pathway")
suffixList = pathwayInfo(pathway)
# Add number of pathway for stats
stats['NB_PATHWAY'] = stats['NB_PATHWAY'] + 1
# Add suffix of pathway in final list
finalList.append(suffixList)
# If enzyme have only 1 pathway and this pathway is in ignored list
# Bad luck !
elif len(pathwayList) == 1 and pathway in ignored:
print(
f" [!] Enzyme {code} have only 1 pathway : {pathway}")
print(f" [!] and this pathway is ignored")
# Add entries for stats
stats['ENZYME_ONLY_IGNORED_PATHWAY'] = stats[
'ENZYME_ONLY_IGNORED_PATHWAY'] + 1
stats['LIST_ENZYME_ONLY_IGNORED_PATHWAY'].append(code)
# Artificially create pathway entry, but empty
suffixList = ['NA']
# Add suffix of pathway in final list
finalList.append(suffixList)
else:
print(f" [!] Ignored pathway : {pathway}")
# Else, if pathway doesn't exist as a key in result
elif 'PATHWAY' not in dictResult.keys():
# Initialize an empty list
finalList = []
# Add suffix in final list
finalList.append(prefixList)
# Display a alert message
print(f"[!] No pathway detected for enzyme {code}\n")
# Increment number of failed pathway in stats and add enzyme in list
stats['MISSING_PATHWAY_IN_KEGG'] = stats[
'MISSING_PATHWAY_IN_KEGG'] + 1
stats['LIST_MISSING_PATHWAY_IN_KEGG'].append(code)
# Artificially create pathway entry, but empty
suffixList = ['NA']
# Add suffix of pathway in final list
finalList.append(suffixList)
return finalList
def mapSpecies(mousepeptrackfilename):
RETRY_TIME = 20.0
mouseTohumanfilepath = os.path.join(os.getcwd(), 'MouseToHuman.tsv')
print("Extracting Mouse to Human Map data, job starts",
str(datetime.datetime.now()))
#increase the field size of CSV
csv.field_size_limit(int(ctypes.c_ulong(-1).value // 2))
try:
urllib.urlretrieve(
'http://www.informatics.jax.org/downloads/reports/HOM_MouseHumanSequence.rpt',
mouseTohumanfilepath)
urllib.urlcleanup()
except:
print("Can't able to download MouseToHuman.tsv file!!")
colnameMousHu = [
'HomoloGene ID', 'Common Organism Name', 'NCBI Taxon ID', 'Symbol',
'EntrezGene ID', 'Mouse MGI ID', 'HGNC ID', 'OMIM Gene ID',
'Genetic Location', 'Genomic Coordinates (mouse: , human: )',
'Nucleotide RefSeq IDs', 'Protein RefSeq IDs', 'SWISS_PROT IDs'
]
mouseHumandata = []
homologID = []
with open(mouseTohumanfilepath) as mhtsvfile:
mhreader = csv.DictReader(mhtsvfile, delimiter='\t')
for mhrow in mhreader:
mhtemplist = []
for i in colnameMousHu:
mhtempdata = str(mhrow[i]).strip()
mhtemplist.append(mhtempdata)
if len(mhtemplist[-1].strip()) > 0:
homologID.append(mhtemplist[0])
mouseHumandata.append(mhtemplist)
homologID = list(set(homologID))
homologID.sort()
mousehumandic = {}
for homologidItem in homologID:
tempHumanHomoUniID = ''
tempMouseHomoUniID = ''
for item in mouseHumandata:
if homologidItem == item[0]:
if 'mouse' in item[1].strip().lower():
tempMouseHomoUniID = item[-1].strip()
else:
tempHumanHomoUniID = item[-1].strip()
if len(tempMouseHomoUniID.strip()) > 0 and len(
tempHumanHomoUniID.strip()) > 0 and tempHumanHomoUniID.strip(
).upper() != 'NA':
mousehumandic[tempMouseHomoUniID] = tempHumanHomoUniID
colname=['UniProtKB Accession','Protein','Gene','Organism','Peptide Sequence','Summary Concentration Range Data','All Concentration Range Data','All Concentration Range Data-Sample LLOQ Based','Peptide ID',\
'Special Residues','Molecular Weight','GRAVY Score','Transitions','Retention Time','Analytical inofrmation',\
'Gradients','AAA Concentration','CZE Purity','Panel','Knockout','LLOQ','ULOQ','Sample LLOQ','Protocol','Trypsin','QC. Conc. Data','Human UniProtKB Accession']
finalresult = []
finalresult.append(colname)
humanUniprotID = []
with open(mousepeptrackfilename) as csvfile:
reader = csv.DictReader(csvfile, delimiter='\t')
for row in reader:
templist = []
for i in colname[:-1]:
tempdata = str(row[i]).strip()
templist.append(tempdata)
if len(str(templist[0]).strip()) > 0:
if templist[0].split('-')[0] in mousehumandic:
humanUniprotID.append(
mousehumandic[templist[0].split('-')[0]])
templist.append(mousehumandic[templist[0].split('-')[0]])
else:
templist.append('NA')
finalresult.append(templist)
with open(mousepeptrackfilename, 'wb') as pf:
pwriter = csv.writer(pf, delimiter='\t')
pwriter.writerows(finalresult)
disGenDataDicName = disGenData()
#disGenDataDicName='disGen.obj'
disGenDataDic = cPickle.load(open(disGenDataDicName, 'rb'))
unqhumanUniprotID = list(set(humanUniprotID))
humanUniprotfuncinfodic = {}
countProt = 0
for subcode in unqhumanUniprotID:
time.sleep(2)
drugbanklist = []
PN = 'NA'
GN = 'NA'
OG = 'NA'
OGID = 'NA'
dislist = []
unidislist = []
unidisURLlist = []
disgendislist = []
disgendisURLlist = []
GoIDList = []
GoNamList = []
GoTermList = []
GOinfo = []
try:
countProt += 1
if countProt % 1000 == 0:
print str(
countProt
), "th protein Protein Name, Gene, Organism Name,drug bank data,disease data job starts", str(
datetime.datetime.now())
SGrequestURL = "https://www.uniprot.org/uniprot/" + str(
subcode) + ".xml"
SGunifile = urllib.urlopen(SGrequestURL)
SGunidata = SGunifile.read()
SGunifile.close()
try:
SGunidata = minidom.parseString(SGunidata)
try:
drugdata = (SGunidata.getElementsByTagName('dbReference'))
for duItem in drugdata:
if (duItem.attributes['type'].value
).upper() == 'DRUGBANK':
try:
drugname = (str(
duItem.getElementsByTagName('property')
[0].attributes['value'].value).strip())
drugid = str(
duItem.attributes['id'].value).strip()
durl = '<a target="_blank" href="https://www.drugbank.ca/drugs/' + drugid + '">' + drugname + '</a>'
drugbanklist.append(durl)
except:
pass
if (duItem.attributes['type'].value
).strip() == 'NCBI Taxonomy':
try:
OGID = str(
duItem.attributes['id'].value).strip()
except:
pass
except IndexError:
pass
try:
godata = (SGunidata.getElementsByTagName('dbReference'))
for gItem in godata:
if (gItem.attributes['type'].value).upper() == 'GO':
try:
gonamedetails = (str(
gItem.getElementsByTagName('property')
[0].attributes['value'].value).strip()
).split(':')[1]
gotermdetails = (str(
gItem.getElementsByTagName('property')
[0].attributes['value'].value).strip()
).split(':')[0]
GoNamList.append(gonamedetails)
goid = str(
gItem.attributes['id'].value).strip()
GoIDList.append(goid)
tempGoTerm = None
if gotermdetails.lower() == 'p':
tempGoTerm = 'Biological Process'
if gotermdetails.lower() == 'f':
tempGoTerm = 'Molecular Function'
if gotermdetails.lower() == 'c':
tempGoTerm = 'Cellular Component'
GoTermList.append(tempGoTerm)
tempGOData = gonamedetails + ';' + goid + ';' + tempGoTerm
GOinfo.append(tempGOData)
except:
pass
if (gItem.attributes['type'].value
).strip() == 'NCBI Taxonomy':
try:
OGID = str(
gItem.attributes['id'].value).strip()
except:
pass
except IndexError:
pass
try:
try:
PN = (((SGunidata.getElementsByTagName('protein')[0]
).getElementsByTagName('recommendedName')[0]
).getElementsByTagName('fullName')[0]
).firstChild.nodeValue
except:
PN = (((SGunidata.getElementsByTagName('protein')[0]
).getElementsByTagName('submittedName')[0]
).getElementsByTagName('fullName')[0]
).firstChild.nodeValue
except IndexError:
pass
try:
try:
GN = ((
SGunidata.getElementsByTagName('gene')[0]
).getElementsByTagName('name')[0]).firstChild.nodeValue
except:
GN = 'NA'
except IndexError:
pass
try:
try:
OG = ((
SGunidata.getElementsByTagName('organism')[0]
).getElementsByTagName('name')[0]).firstChild.nodeValue
except:
OG = 'NA'
except IndexError:
pass
try:
disdata = SGunidata.getElementsByTagName('disease')
for dItem in disdata:
disname = ''
disshort = ''
disURL = ''
disID = ''
try:
disname = (dItem.getElementsByTagName('name')[0]
).firstChild.nodeValue
disID = (dItem.attributes['id'].value).upper()
except:
pass
try:
disshort = (dItem.getElementsByTagName('acronym')
[0]).firstChild.nodeValue
except:
pass
if len(disname.strip()) > 0:
disURL = '<a target="_blank" href="https://www.uniprot.org/diseases/' + disID + '">' + str(
disname.strip()) + '(' + str(
disshort) + ')' + '</a>'
dislist.append(
str(disname.strip()) + '(' + str(disshort) +
')')
unidislist.append(
str(disname.strip()) + '(' + str(disshort) +
')')
unidisURLlist.append(disURL)
except IndexError:
pass
except ExpatError:
pass
except IOError:
pass
drugbankdata = 'NA'
disdata = 'NA'
uniDisData = 'NA'
uniDisURLData = 'NA'
disgenDisData = 'NA'
disgenDisURLData = 'NA'
goiddata = 'NA'
gonamedata = 'NA'
gotermdata = 'NA'
goData = 'NA'
if GN != 'NA' and GN in disGenDataDic:
disgendislist = disGenDataDic[GN][0]
disgendisURLlist = disGenDataDic[GN][1]
if len(dislist) > 0:
dislist = dislist + disGenDataDic[GN][0]
else:
dislist = disGenDataDic[GN][0]
if len(GoIDList) > 0:
goiddata = '|'.join(list(set(GoIDList)))
if len(GoNamList) > 0:
gonamedata = '|'.join(list(set(GoNamList)))
if len(GoTermList) > 0:
gotermdata = '|'.join(list(set(GoTermList)))
if len(GOinfo) > 0:
goData = '|'.join(list(set(GOinfo)))
if len(drugbanklist) > 0:
drugbankdata = '|'.join(list(set(drugbanklist)))
if len(dislist) > 0:
disdata = '|'.join(list(set(dislist)))
if len(unidislist) > 0:
uniDisData = '|'.join(list(set(unidislist)))
if len(unidisURLlist) > 0:
uniDisURLData = '|'.join(list(set(unidisURLlist)))
if len(disgendislist) > 0:
disgenDisData = '|'.join(list(set(disgendislist)))
if len(disgendisURLlist) > 0:
disgenDisURLData = '|'.join(list(set(disgendisURLlist)))
humanUniprotfuncinfodic[subcode] = [
PN, GN, OG, OGID, disdata, uniDisData, uniDisURLData,
disgenDisData, disgenDisURLData, drugbankdata, goiddata,
gonamedata, gotermdata, goData
]
hudicfile = 'humanUniprotfuncinfodic.obj'
hudicf = open(hudicfile, 'wb')
pickle.dump(humanUniprotfuncinfodic, hudicf, pickle.HIGHEST_PROTOCOL)
hudicf.close()
print("Extracting KEGG pathway name, job starts",
str(datetime.datetime.now()))
hkeggdictfile = {}
huniproturl = 'https://www.uniprot.org/uploadlists/'
hk = KEGG()
for hkx in range(0, len(unqhumanUniprotID), 2000):
countProt += hkx + 2000
if countProt % 2000 == 0:
print(str(countProt), "th protein kegg job starts",
str(datetime.datetime.now()))
huniprotcodes = ' '.join(unqhumanUniprotID[hkx:hkx + 2000])
huniprotparams = {
'from': 'ACC',
'to': 'KEGG_ID',
'format': 'tab',
'query': huniprotcodes
}
while True:
try:
hkuniprotdata = urllib.urlencode(huniprotparams)
hkuniprotrequest = urllib2.Request(huniproturl, hkuniprotdata)
hkuniprotresponse = urllib2.urlopen(hkuniprotrequest)
for hkuniprotline in hkuniprotresponse:
hkudata = hkuniprotline.strip()
if not hkudata.startswith("From"):
hkuinfo = hkudata.split("\t")
if len(hkuinfo[1].strip()):
hkegg = hk.get(hkuinfo[1].strip())
hkudict_data = hk.parse(hkegg)
try:
try:
if len(str(hkuinfo[0]).strip()) > 5:
tempkeggData = '|'.join(
'{};{}'.format(key, value)
for key, value in
hkudict_data['PATHWAY'].items())
hkeggdictfile[hkuinfo[0].strip()] = [
hkudict_data['PATHWAY'].values(),
tempkeggData
]
except TypeError:
pass
except KeyError:
pass
break
except urllib2.HTTPError:
time.sleep(RETRY_TIME)
print(
'Hey, I am trying again until succeeds to get data from KEGG!',
str(datetime.datetime.now()))
pass
hkdicfile = 'humankeggdic.obj'
hkdicf = open(hkdicfile, 'wb')
pickle.dump(hkeggdictfile, hkdicf, pickle.HIGHEST_PROTOCOL)
hkdicf.close()
def get_kegg_info(stId):
k = KEGG()
data = k.get(stId)
dict_data = k.parse(data)
return dict_data