def SearchCARE(filename, dbname, bgfile):
"""
在序列中搜索顺式调控元件motif
"""
basename = filename.split(".")[0]
if not os.path.exists(dbname):
CARE = CAREdb(dbname)
CARE.importMotifs("CARE.txt")
else:
CARE = CAREdb(dbname)
Motif_lst = loadMotifs(CARE)
scanned = checkScannedLog("%s.log" % basename)
scanlog = open("%s.log" % basename, "a")
scanned_path = "%s/Scan" % basename
if not os.path.exists(scanned_path):
os.makedirs(scanned_path)
for entry in Motif_lst:
(motif_name, Sequence) = entry
print motif_name, Sequence
if Sequence in scanned:
print "skip"
continue
CARE.addScanned(ScanMotif(Sequence, filename, bgfile))
scanned.append(Sequence)
scanlog.write(Sequence + "\n")
def dumpCoMotif(co_Occur,Merged_REF_SeqName_dist,output,zip=False):
care=CAREdb(dbname)
Seqnames=care.loadDB2dict('fa_file',0,1)
if not os.path.exists(output):
os.makedirs(output)
if not os.path.exists("%s/gene_list" % output):
os.makedirs("%s/gene_list" % output)
#Co-occur
for motifa,motifb in co_Occur.keys():
print motifa,motifb
filename="%s&%s.lst" % (motifa,motifb)
outfile=open("%s/gene_list/%s" % (output,filename),'w')
outfile.write("\n".join([Seqnames[k] for k in co_Occur[(motifa,motifb)].keys()]))
outfile.close()
#single motif
motifs=[]
[motifs.extend(k) for k in co_Occur.keys()]
motifs=set(motifs)
for motif in motifs:
filename="%s.lst" % motif
outfile=open("%s/gene_list/%s" % (output,filename),'w')
outfile.write("\n".join([Seqnames[k] for k in Merged_REF_SeqName_dist[motif].keys()]))
outfile.close()
"""
select count(distinct REF_SeqName) from cache.Instance,cache.Scanned
where cache.Scanned.REF_MotifSeq=cache.Instance.REF_MotifSeq and REF_Motif=1
"""
]
SQL3=[
"""
SELECT REF_SeqName,REF_Motif,REF_Organism,Description,start,stop,strand,pValue
FROM cache.Instance,cache.Scanned
WHERE cache.Instance.REF_MotifSeq=cache.Scanned.REF_MotifSeq
AND REF_SeqName in ('40747','10669','1620','5828','30256','10402')
""",
]
benchmark=CAREdb('RAP_3kbp.db')
benchmark.cache('Instance')
benchmark.cache('Scanned',['REF_SeqName','REF_MotifSeq','start','stop','strand','pValue'],SeqName=['40747','10669','1620','5828','30256','10402'])
for sql in SQL3:
print sql
timestamp=time.time()
benchmark.cur.execute(sql)
print [rs[0] for rs in benchmark.cur.fetchall()]
print time.time()-timestamp
'''
print "------------"
tables=['Scanned','cache.Scanned']
for tname in tables:
timestamp=time.time()
#benchmark.select(tname)