def composeTMP_nonTMP(ftmp, fnontmp, fpair, num):
'''
如果给了ftmp,fnontmp，就参考所给的列表组合
如果这个文件不存在，则从数据库中查询这两个列表
:param ftmp:
:param fnontmp:
:param fpair: 组合结果
:param num: 组合的总数
:return:
'''
if not (os.access(ftmp, os.F_OK) and os.access(fnontmp, os.F_OK)):
tmplist, nontmplist = generateCriterLists(ftmp, fnontmp)
else:
tmplist, nontmplist = readIDlist(ftmp), readIDlist(fnontmp)
sampL1 = min(len(tmplist), num)
sampL2 = min(len(nontmplist), num)
L1 = random.sample(range(0, len(tmplist)), sampL1)
L2 = random.sample(range(0, len(nontmplist)), sampL2)
with open(fpair, 'w') as fo:
for idx in range(num):
fo.write('%s\t%s\n' %
(tmplist[L1[(idx + random.randint(2, 9)) % sampL1]],
nontmplist[L2[(idx + random.randint(1, 5)) % sampL2]]))
fo.flush()
print(idx)
def getSingleInfo(fin, fout, fin_type='single', col=[0, 1]):
if fin_type == 'pair':
df = pd.read_table(fin, header=None)[col]
dat = df.to_numpy().reshape(1, -1)
proteins = set(dat[0])
elif fin_type == 'single':
proteins = readIDlist(fin)
else:
pass
do = DataOperation('uniprot', 'uniprot_sprot')
projection = {
'_id': True,
'sequence.@length': True,
'sequence.#text': True,
'keyword.@id': True,
'comment.subcellularLocation.location': True
}
prod = Protein()
with open(fout, 'w') as fo:
for AC in proteins:
pro = queryProtein(AC, do, projection=projection)
pro['accession'] = AC
if not prod.checkProtein(
pro['sequence']['#text'], 50, 2000, uncomm=True):
continue
proinfo = ensomblePortein(pro)
for v in proinfo.values():
fo.write(str(v))
fo.write('\t')
fo.write('\n')
fo.flush()
def queryPathway_Gene(foutPathway_Gene,
fpathway=None,
fpathwayInfo=None,
hsa='hsa'):
'''
:param foutPathway_Gene:
0 hsa00010 pathway id
1 3101 gene id
2 HK3 gene symbol
3 [K00844] KO id
4 [2.7.1.1] EC number
:param fpathway:
:param fpathwayInfo:
:param hsa:
:return:
'''
if os.access(fpathway, os.F_OK):
repair_pathways = readIDlist(fpathway)
else:
repair_pathways = queryAllPathway(fpathway=fpathway,
fpathwayInfo=fpathwayInfo,
hsa=hsa)
for idx, pathway in enumerate(repair_pathways):
geneID_geneName_KO_EC = []
print(idx, end='.')
if idx < 65: continue
for gene_id, gene_symbol, KO, EC in extractGeneFromPathway(pathway):
geneID_geneName_KO_EC.append(
(pathway, gene_id, gene_symbol, KO, EC))
saveList(geneID_geneName_KO_EC, foutPathway_Gene, file_mode='a')
def extractFasta(self,
fin_fasta,
fin_idlist,
fout_fasta,
in_multi=True,
out_multi=True):
oridict = self.getDict(fin_fasta, multi=in_multi)
desdict = {}
idlist = readIDlist(fin_idlist)
for id in idlist:
desdict[id] = oridict[id]
self.dict2fasta(desdict, fout_fasta, multi=out_multi)
def fullyComposeTMP_nonTMP(ftmp, fnontmp, fpair):
'''
如果给了ftmp,fnontmp，就参考所给的列表组合
如果这个文件不存在，则从数据库中查询这两个列表
:param ftmp:
:param fnontmp:
:param fpair: 组合结果
:param num: 组合的总数
:return:
'''
if not (os.access(ftmp, os.F_OK) and os.access(fnontmp, os.F_OK)):
print('Cretira not found, qury from Mongodb...')
tmplist, nontmplist = generateCriterLists(ftmp, fnontmp)
else:
tmplist, nontmplist = readIDlist(ftmp), readIDlist(fnontmp)
with open(fpair, 'w') as fo:
for elem1 in tmplist:
for elem2 in nontmplist:
print(elem1, elem2)
fo.write('%s\t%s\n' % (elem1, elem2))
fo.flush()
def findKeyProtein(fin, fout, keyword):
# keyword = 'KW-0297'
tmplist = readIDlist(fin)
GPCRlist = []
do = DataOperation('uniprot', 'uniprot_sprot')
projection = {'_id': True}
for ac in tmplist:
dic = {'accession': ac, "keyword.@id": keyword}
result = do.QueryObj(dic, projection=projection)
for r in result:
GPCRlist.append(ac)
print(r)
saveList(GPCRlist, fout)
def getFasta(fin_all_protein, fout):
proteins = readIDlist(fin_all_protein)
do = DataOperation('uniprot', 'uniprot_sprot')
projection = {'sequence.#text': True}
with open(fout, 'w') as fo:
from Bio import SeqIO
from Bio.Seq import Seq
from Bio.SeqRecord import SeqRecord
records = []
for AC in proteins:
pro = queryProtein(AC, do, projection=projection)
if not pro: continue
record = SeqRecord(Seq(pro['sequence']['#text']),
id=AC,
description='')
records.append(record)
#fo.write('>%s\n%s\n'%(AC,pro['sequence']['#text']))
#fo.flush()
SeqIO.write(records, fout, 'fasta')
def findGProtein(fin, fout):
# {"protein.recommendedName.fullName":{$regex:/Guanine nucleotide-binding protein*/}}
# keyword = 'KW-0297'
tmplist = readIDlist(fin)
GPCRlist = []
do = DataOperation('uniprot', 'uniprot_sprot')
projection = {'_id': True}
count = 0
for ac in tmplist:
dic = {
'accession':
ac,
'$or': [{
"protein.recommendedName.fullName": {
'$regex': 'G protein '
}
}, {
"protein.recommendedName.fullName": {
'$regex': 'Guanine nucleotide-binding protein'
}
}, {
"protein.alternativeName.fullName": {
'$regex': 'G protein '
}
}, {
"protein.alternativeName.fullName": {
'$regex': 'Guanine nucleotide-binding protein'
}
}]
}
result = do.QueryObj(dic, projection=projection)
for r in result:
count = count + 1
GPCRlist.append(ac)
print(count, r)
saveList(GPCRlist, fout)
def getTmp_SpPair(tmpf,spf,finPair,foutPair,type1='TMP',type2='SP',crossover = False):
"""
cretira:/home/jjhnenu/data/PPI/release/criteria/ 20200701
tmpf='/home/jjhnenu/data/PPI/release/criteria/allcession_KW-0812_131609.list'
spf='/home/jjhnenu/data/PPI/release/criteria/allcession_soluble_614454.list'
:param tmpf:
:param spf:
:param finPair:
:param foutPair:
:return:
tmpf = r'E:\githubCode\BioComputeCodeStore\JiangJiuhong\data\PPI\tmp\\KW-0812.list'
spf = r'E:\githubCode\BioComputeCodeStore\JiangJiuhong\data\PPI\sp\\splist.list'
finPair = r'E:\githubCode\BioComputeCodeStore\JiangJiuhong\data\PPI\interactor_interaction\\noprepeat.txt'
foutPair = r'E:\githubCode\BioComputeCodeStore\JiangJiuhong\data\PPI\tmp_sp\\tmp_sp.txt'
getTmp_SpPair(tmpf, spf, finPair, foutPair)
"""
tmplist = readIDlist(tmpf)
splist = readIDlist(spf)
print(len(tmplist),len(splist))
# len(tmplist)<len(splist)
splitmark = '\\'
if '/' in foutPair:
splitmark = '/'
mypath = foutPair[:foutPair.rindex(splitmark)+1]
fSP_SP = mypath+'%s_%s.txt'%(type2,type2)
fTMP_TMP = mypath+'%s_%s.txt'%(type1,type1)
fnonTMP_SP = mypath + 'non%s_%s.txt' % (type1, type2)
fTmp_nonSP = mypath + '%s_non%s.txt' % (type1, type2)
fTMP_nonTmp = mypath+'%s_non%s.txt'%(type1,type1)
fnonTmp_nonTmp = mypath+'non%s_non%s.txt'%(type1,type1)
fnonsp_sp = mypath+'non%s_%s.txt'%(type2,type2)
fnonsp_nonsp = mypath+'non%s_non%s.txt'%(type2,type2)
fdrop = mypath+'drop.txt'
with open(finPair, 'r') as fin, \
open(foutPair, 'w') as fout,\
open(fSP_SP,'w') as foutsp,\
open(fTMP_TMP,'w') as fouttmp,\
open(fnonTMP_SP,'w') as foutnonTMP_SP,\
open(fTmp_nonSP,'w') as foutTmp_nonSP,\
open(fTMP_nonTmp,'w') as foutTMP_nonTmp,\
open(fnonTmp_nonTmp,'w') as foutnonTmp_nonTmp,\
open(fnonsp_sp, 'w') as fout_nonsp_sp,\
open(fnonsp_nonsp, 'w') as fout_nonsp_nonsp,\
open(fdrop, 'w') as fout_drop\
:
line = fin.readline()
while (line):
pair = line.split('\t')
try:
a = pair[0]
b = pair[1][:-1]
except:
print(pair)
# Atmp = a in tmplist
# Asp = False if Atmp and not crossover else a in splist
# Btmp = b in tmplist
# Bsp = False if Btmp and not crossover else b in splist
Atmp = a in tmplist
Asp = a in splist
Btmp = b in tmplist
Bsp = b in splist
if Atmp and Bsp: # TMP_SP
fout.write(a + '\t' + b + '\n')
fout.flush()
print('%s %s save this pair' % (a, b))
elif Asp and Btmp: # TMP_SP
fout.write(b + '\t' + a + '\n')
fout.flush()
print('%s %s save this pair' % (a, b))
elif Asp and Bsp: # SP_SP
foutsp.write(a + '\t' + b + '\n')
foutsp.flush()
print('%s %s save this pair %s_%s' % (a, b,type2,type2))
elif Atmp and Btmp: # TMP_TMP
fouttmp.write(a + '\t' + b + '\n')
fouttmp.flush()
print('%s\t%s save this pair %s_%s' % (a, b,type1,type1))
else:
pass
if not Atmp and Bsp:
foutnonTMP_SP.write(a + '\t' + b + '\n')
foutnonTMP_SP.flush()
print('%s %s save this pair %s_non%s' % (a, b, type1, type2))
elif Atmp and not Bsp:
foutTmp_nonSP.write(a + '\t' + b + '\n')
foutTmp_nonSP.flush()
print('%s %s save this pair %s_non%s' % (a, b, type1, type2))
else:pass
# if (a in tmplist and b not in tmplist) or (a not in tmplist and b in tmplist):
# foutTMP_nonTmp.write(a + '\t' + b + '\n')
# foutTMP_nonTmp.flush()
# print('%s %s save this pair TMP_nonTmp' % (a, b))
# if a not in tmplist and b not in tmplist:
# foutnonTmp_nonTmp.write(a + '\t' + b + '\n')
# foutnonTmp_nonTmp.flush()
# print('%s %s save this pair nonTMP_nonTmp' % (a, b))
if Atmp and not Btmp: # TMP_nonTMP
foutTMP_nonTmp.write(a + '\t' + b + '\n')
foutTMP_nonTmp.flush()
print('%s %s save this pair %s_non%s' % (a, b,type1,type1))
elif not Atmp and Btmp: # TMP_nonTMP
foutTMP_nonTmp.write(b + '\t' + a + '\n')
foutTMP_nonTmp.flush()
print('%s %s save this pair %s_non%s' % (a, b,type1,type1))
elif not Atmp and not Btmp:
foutnonTmp_nonTmp.write(a + '\t' + b + '\n')
foutnonTmp_nonTmp.flush()
print('%s %s save this pair non%s_non%s' % (a, b,type1,type1))
else:
pass
if not Asp and Bsp:
fout_nonsp_sp.write(a + '\t' + b + '\n')
fout_nonsp_sp.flush()
print('%s %s save this pair non%s_%s' % (a, b,type2,type2))
elif Asp and not Bsp:
fout_nonsp_sp.write(b + '\t' + a + '\n')
fout_nonsp_sp.flush()
print('%s %s save this pair non%s_%s' % (a, b,type2,type2))
elif not Asp and not Bsp:
fout_nonsp_nonsp.write(a + '\t' + b + '\n')
fout_nonsp_nonsp.flush()
print('%s %s save this pair non%s_non%s' % (a, b,type2,type2))
else:
# fout_drop.write(a + '\t' + b + '\n')
# print('%s %s drop this pair' % (a, b))
pass
line = fin.readline()
func = countpair
handledir(mypath, func)
print('get%s_%sPair end'%(type1,type2))
# Title : tesKEGGAPI.py # Created by: [email protected] # Created on: 2021/2/4 15:23 # des : TODO import os from common import readIDlist if __name__ == '__main__': from Bio.KEGG import REST dirout = 'file/6bioAnalysis/keggDB/pathwayInfo' human_pathways = REST.kegg_list("pathway", "hsa").read() repair_pathways = readIDlist('file/6bioAnalysis/keggDB/1pathway_human.tsv') # Get the genes for pathways and add them to a list repair_genes = [] for idx,pathway in enumerate(repair_pathways): print(idx,pathway) pathway_file = REST.kegg_get(pathway).read() # query and read each pathway with open(os.path.join(dirout,'%s.txt'%pathway),'w') as fo: fo.write(pathway_file) fo.flush()
def handlePair(foutdir,
sep=',',
dbname=None,
checkTMP=True,
jumpStep=None,
fin=None,
f2tmp_nonTtmp_info_qualified=None,
keepOne=False):
'''
数据量较少，直接逐行查询，很多蛋白被查询了多次
:param foutdir:
:param sep: sep of fin file
:parameter dbname: name of mongodb
:parameter jumpStep: skip some step in this method [1,2,3,4]
:parameter fin:ignore this parameter when 1 in jumpStep
:parameter f2tmp_nonTtmp_info_qualified: sign this path in the dir
:return:
fin = 'file/1intAct_pair_norepeat.txt'
foutdir = 'file/1positive'
handlePair(fin,foutdir)
'''
'''
config path
'''
f1tmp_nontmp_info = os.path.join(foutdir, '1tmp_nontmp_info.tsv')
f1TMP_nonTMP = os.path.join(foutdir, '1tmp_nontmp.tsv')
if not f2tmp_nonTtmp_info_qualified:
f2tmp_nonTtmp_info_qualified = os.path.join(
foutdir, '2tmp_nontmp_info_qualified.tsv')
fout_fasta = os.path.join(foutdir, '2pair.fasta')
fout_tmp_fasta = os.path.join(foutdir, '2tmp.fasta')
fout_nontmp_fasta = os.path.join(foutdir, '2nontmp.fasta')
f2positive = os.path.join(foutdir, '2pair.tsv')
f2tmp = os.path.join(foutdir, '2tmp.list')
f2nontmp = os.path.join(foutdir, '2nontmp.list')
f2all = os.path.join(foutdir, '2all.list')
f2tmp_info = os.path.join(foutdir, '2tmp_info.tsv')
f2nontmp_info = os.path.join(foutdir, '2nontmp_info.tsv')
f2all_info = os.path.join(foutdir, '2all_info.tsv')
f3subcell = os.path.join(foutdir, '3subcellular.tsv')
'''
1. get tmp nontmp pair
time 1766.4131457805634
'''
if jumpStep == None or 1 not in jumpStep:
getPairInfo_TMP_nonTMP(fin,
f1tmp_nontmp_info,
sep=sep,
checkTMP=checkTMP,
keepOne=keepOne)
simplifyTable(f1tmp_nontmp_info, f1TMP_nonTMP)
'''
2. get qualified tmp nontmp pair
'''
if jumpStep == None or 2 not in jumpStep:
saveQualified(f1tmp_nontmp_info, f2tmp_nonTtmp_info_qualified)
'''
3. get related list,fasta,pair
'''
if jumpStep == None or 3 not in jumpStep:
simplifyTable(f2tmp_nonTtmp_info_qualified, f2positive)
extractPairAndFasta(f2tmp_nonTtmp_info_qualified,
fout_fasta,
fout_tmp_fasta=fout_tmp_fasta,
fout_nontmp_fasta=fout_nontmp_fasta)
getproteinlist(f2tmp_nonTtmp_info_qualified,
ftmp=f2tmp,
fnontmp=f2nontmp,
fall=f2all,
ftmp_info=f2tmp_info,
ftmp_nontmp_info=f2nontmp_info,
fall_info=f2all_info)
'''
4. save to mongodb
'''
if jumpStep == None or (4 not in jumpStep and dbname):
notsvaelist = save(readIDlist(f2all), dbname)
print('those protein not save in the mongodb', notsvaelist)
'''
5. calcu subcellular
'''
if jumpStep == None or 5 not in jumpStep:
handleRow(f2tmp_nonTtmp_info_qualified, f3subcell, calcuSubcell)