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 loaddata(ftmp_fa, fnontmp_fa):
'''
:param ftmp_fa: tmp fasta file
:param fnontmp_fa: nontmp fasta file
:return: idpair, feature
# ftmp_fa = 'file/8humanPredict/3tmp.fasta'
# fnontmp_fa = 'file/8humanPredict/3nontmp.fasta'
'''
fd = FastaDealer()
bf = BaseFeature()
p = Protein()
# flag = False
# flagA = False
# flagB = False
for pa in SeqIO.parse(ftmp_fa, 'fasta'):
# if pa.id == 'Q8TCW7':flagA = True # 程序中断之后，需要用这个代码控制起始预测位置
# if not flagA:continue
if not p.checkProtein(pa.seq, 50, 2000, uncomm=True): continue
a = fd.phsi_blos(pa.seq)
for pb in SeqIO.parse(fnontmp_fa, 'fasta'):
# if pb.id == 'Q15942':flagB = True
# if not flagB:continue
if not p.checkProtein(pb.seq, 50, 2000, uncomm=True): continue
b = fd.phsi_blos(pb.seq)
c = bf.padding_PSSM(a, b, vstack=True, shape=(2000, 25))
yield '%s-%s' % (pa.id, pb.id), c
def getNpy(self, fin_fasta, out_dir, multi=True, checkprotein=True):
check_path(out_dir)
p = Protein()
for ID, seq in self.getYield(fin_fasta, multi=multi):
if not checkprotein or p.checkProtein(seq, 50, 2000, uncomm=True):
filename = os.path.join(out_dir, "%s.npy" % ID)
result = self.seq2num(seq)
if len(result) != 0:
np.save(filename, result)
def ensomblePortein(pro):
ens_pro = {}
prd = Protein()
ens_pro['accession'] = pro['accession']
ens_pro['name'] = pro['_id']
ens_pro['length'] = pro['sequence']['@length']
ens_pro['noX'] = prd.checkUncomm(pro['sequence']['#text'])
ens_pro['inlenRange'] = prd.checkLengthRange(int(
pro['sequence']['@length']),
min=50,
max=2000)
ens_pro['subcellularLocations'] = [] if 'comment' not in pro.keys(
) else getSubcelllist(pro['comment'])
ens_pro['seq'] = pro['sequence']['#text']
return ens_pro
""" @author: [email protected] @time: 2020/4/17 17:50 @desc: """ import os from Bio import SeqIO import numpy as np from Bio.Seq import Seq from Bio.SeqRecord import SeqRecord from ProteinDealer import Protein from common import check_path, getPairs, readIDlist p = Protein() class FastaDealer: 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)
# encoding: utf-8 """ @author: [email protected] @time: 2020/4/17 17:52 @desc: """ import os import pandas as pd from ProteinDealer import Protein import numpy as np from common import getPairs, check_path myprotein = Protein() class ComposeData: def save(self, dirout, flist, ratios, limit, labels=None, sep='\t', filename='all.txt', groupcount=1, repeate=True): """ same length of flist,ratios,labels if not None