def _main(args):
if len(args) != 3:
print("usage: xls_get_region_from_fasta.py <fasta> <xls> <window>")
sys.exit(1)
win = int(args[2])
fasta = fasta_subseq_2.FastaDB()
fasta.openFastaFile(args[0])
seqs = []
for ln in open(args[1]):
sp = ln[:-1].split()
print sp
pk = int(sp[1]) + int(sp[4])
seq = fasta[sp[0]]['sequence'][(pk - win):(pk + win)]
get_in = sp[
-1] #raw_input(">%s:%d..%d\n\'k\'=keep; \'r\' = reverse comp; \'<anything else>\' = discard: " % (sp[0],pk-win,pk+win))
if get_in == 'k':
pass
elif get_in == 'r':
seq = fasta_subseq_2.revcomp(seq)
else:
continue
seqs.append(">%s:%d..%d\n%s" % (sp[0], pk - win, pk + win, seq))
outfile = raw_input("name of output file: ")
outfh = open(outfile, "w")
for s in seqs:
print >> outfh, s
def get_enrich(xls_regs, sgr, winsize, fasta):
fasta_db = fasta_subseq_2.FastaDB()
fasta_db.openFastaFile(fasta)
for reg in xls_regs:
for x in ('mtx1_hits', 'mtx2_hits'):
hit_info = []
for h in reg[x]:
hit = {'hit_obj': h}
width = abs(h['start'] - h['end'])
if h['strand'] == "+":
seq = fasta_db[h['chr']]['sequence'][h['start']:(
h['start'] + width)]
hit['loc'] = h['start']
else:
### !!!!! CHANGE IF FIX HIT DATABASE!!!!
seq = fasta_subseq_2.revcomp(
fasta_db[h['chr']]['sequence'][h['end']:(h['end'] +
width)])
hit['loc'] = h['end']
hit['nearest'] = (0, 0)
hit['vals'] = []
hit['seq'] = seq
hit_info.append(hit)
reg[x + '_info'] = hit_info
for y in open(sgr):
(chr, loc, val) = y.split()
loc = int(loc)
val = int(val)
#print chr
for x in xls_regs:
for hit_info in ('mtx1_hits_info', 'mtx2_hits_info'):
for d in x[hit_info]:
#print (loc,target_loc)
target_loc = d['loc']
if (chr == d['hit_obj']['chr']) and (
abs(loc - target_loc) <
abs(loc - d['nearest'][0])):
d['nearest'] = (loc, val)
if (chr == d['hit_obj']['chr']) and (abs(loc - target_loc)
< (winsize / 2)):
d['vals'].append(val)
print >> sys.stderr, d
for x in xls_regs:
for hit_info in ('mtx1_hits_info', 'mtx2_hits_info'):
for h in x[hit_info]:
h['win_mean'] = np.mean(h['vals'])
h['win_median'] = np.median(h['vals'])
h['enrich_md'] = h['nearest'][1] / h['win_median']
h['enrich_mn'] = h['nearest'][1] / h['win_mean']
print >> sys.stderr, h
return xls_regs
def _main(args):
if len(args) < 1:
print "usage: revcomseq.py <fasta>"
sys.exit(0)
sq = ""
header = ""
for l in open(args[0]):
if re.search("^>",l):
header = l[:-1] + "_rc"
sq=""
if (sq):
tr_seq = re.sub("\n","",sq)
print "%s\n%s" % (header,revcomp(tr_seq))
else:
sq += l
tr_seq = re.sub("\n","",sq)
print "%s\n%s" % (header,revcomp(tr_seq))
def _main(args):
if len(args) < 4:
print >> sys.stderr, "usage: xls_motif_window.py <xls> <fasta> <matrix_file> <window>"
sys.exit(1)
fasta = fasta_subseq_2.FastaDB()
fasta.openFastaFile(args[1])
xls_regions = []
for x in open(args[0]):
spl = x[:-1].split()
region = {
'chr': spl[0],
'start': int(spl[1]),
'end': int(spl[2]),
'enrich': spl[7]
}
region['seq'] = fasta[
region['chr']]['sequence'][region['start']:region['end']]
xls_regions.append(region)
for r in xls_regions:
try:
annot = patser_tools.makePatserAnnotation(sequence=r['seq'],
matrix=args[2])
except IOError:
print >> sys.stderr, "Error in seq %s:%d..%d:" % (
r['chr'], r['start'], r['end'])
continue
if len(annot.getAllFeatures()) < 1:
continue
maxhit = annot.getMaxFeature("score")
winstart = None
winend = None
winseq = None
if maxhit.tags["strand"] == '+':
winstart = r['start'] + (maxhit.start - int(args[3]) / 2)
winend = r['start'] + (maxhit.start + int(args[3]) / 2)
win_seq = fasta[r['chr']]['sequence'][winstart:winend]
else:
winstart = r['start'] + ((maxhit.end - 3) - int(args[3]) / 2)
winend = r['start'] + ((maxhit.end - 3) + int(args[3]) / 2)
win_seq = fasta_subseq_2.revcomp(
fasta[r['chr']]['sequence'][winstart:winend])
print ">%s:%d..%d:%s enr=%s mtx=%s" % (
r['chr'], winstart, winend, maxhit.tags['strand'], r['enrich'],
maxhit.tags['score'])
print win_seq
def run_aln_mapline(map_line):
files = []
for (sp, fasta) in map_line['fastas']:
#print >> sys.stderr, (map_line['map_idx'],'a')
sp_map = map_line['map_dict'][sp]
#print >> sys.stderr, (map_line['map_idx'],'b')
outname = str(map_line['map_idx']) + "_" + sp + ".fa"
out = open(outname, "w")
#print >> sys.stderr, (map_line['map_idx'],'c')
print >> sys.stderr, sp_map
print >> sys.stderr, sp
seq = fasta[sp_map['chr']][sp_map['start']:sp_map['end'] - 1]
#print >> sys.stderr, (map_line['map_idx'],'d')
if sp_map['strand'] == "-":
print sp + " revcomp"
seq = fasta_subseq_2.revcomp(seq)
#print >> sys.stderr, (map_line['map_idx'],'e')
print >> out, ">%s %s-%d:%d" % (sp, sp_map['chr'], sp_map['start'],
sp_map['end'])
print >> out, seq
print "Wrote fasta %s" % (outname, )
#print (map_line['map_idx'],'f')
out.close()
files.append(outname)
#print >> sys.stderr, (map_line['map_idx'],'g')
call_pecan = [
"java", "-classpath", PECANPATH, "-Xmx2000m", "bp.pecan.Pecan", "-E",
map_line['tree'] + ";", "-F"
]
call_pecan.extend(files)
call_pecan.extend(["-G", str(map_line['map_idx']) + ".mfa"])
print "Running pecan with command: %s" % (" ".join(call_pecan))
print "Starting alignment %d..." % (map_line['map_idx'], )
sub.check_call(call_pecan)
print "Alignment %d finished" % (map_line['map_idx'], )
map(os.remove, files)
def _main(args):
if len(args) != 3:
print "usage: <bed_file> <seq_file> <matrix>"
sys.exit(0)
fasta = fasta_subseq_2.FastaDB()
fasta.openFastaFile(args[1])
bed_annots = []
bed_in = open(args[0])
for line in bed_in:
spl = line[:-1].split()
fseq = fasta[spl[0]]["sequence"][int(spl[1]):int(spl[2])]
if spl[5] == "-":
fseq = fasta_subseq_2.revcomp(fseq)
#print spl
try:
patannot = patser_tools.makePatserAnnotation(sequence=fseq,
matrix=args[2])
except:
continue
#print "-" * 30
#print spl
#print pp(patannot.getAllFeatures())
bed_annots.append({
"seq": spl[0] + "_" + spl[1] + "_" + spl[2],
"annotation": patannot
})
for ann in bed_annots:
for feat in ann["annotation"].getAllFeatures():
print "%s\t%i\t%i\t%f\t%f\t%s" % (
ann["seq"], feat.st, feat.en, feat.tagset["score"],
feat.tagset["pval"], feat.tagset["strand"])
def _main(args):
if len(args) < 1:
print "usage: patser_list.py [mtx1][mtx2] ... < seqs.fa"
sys.exit(1)
matrices = args
seqs = sys.stdin
hits = {}
name = ""
seq = ""
pssms = convertFreqMtx(matrices)
for s in seqs:
nameres = re.search(">(\S+)", s)
if nameres and not (name == ""):
hits[name] = {'seq': seq}
name = nameres.group(1)
seq = ""
elif nameres and (name == ""):
#print "1"
name = nameres.group(1)
seq = ""
else:
seq += s[:-1]
if not (name == ""):
hits[name] = {'seq': seq}
#print hits
mtx_names = []
for (name, d) in hits.iteritems():
for mtx in matrices:
hit_annot = patser_tools.makePatserAnnotation(sequence=d['seq'],
matrix=mtx,
seqname=name,
scorecut=-100)
features = hit_annot.getAllFeatures()
hit = None
if len(features) > 0:
max = features[0]
for x in features:
if x.tags['score'] > max.tags['score']:
max = x
hit = max
else:
print >> sys.stderr, "Sequence %s: No hit for matrix %s in %s" % (
name, mtx, d['seq'])
continue
#print hit
d[hit.tags['motif_name']] = hit
if hit.tags['motif_name'] not in mtx_names:
mtx_names.append(hit.tags['motif_name'])
#print hits
print "name\tsequence\t",
for x in mtx_names:
print "%s_score\t%s_pval\t%s_PSSM_score" % (x, x, x),
print ""
for (name, h) in hits.iteritems():
matrices = [x for x in h.keys() if not x == 'seq']
print "%s\t%s\t" % (name, h['seq']),
for x in matrices:
print str(h[x].tags['score']) + "\t",
if 'pval' in h[x].tags.keys():
print str(h[x].tags['pval']) + "\t",
else:
print "-",
pssm_scores = [
scoreSeq(pssms[x], h['seq']),
scoreSeq(pssms[x], fasta_subseq_2.revcomp(h['seq']))
]
score = None
if pssm_scores[1] > pssm_scores[0]:
score = pssm_scores[1]
else:
score = pssm_scores[0]
print str(score) + "\t",
print ""