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):
usage = "pecan_WGA_runner.py <genome_fastas_file> <treefile> <mercator_map> <mercator_genome_order> <outdir>"
if len(args) != 5:
print usage
sys.exit(0)
genome_dict = {}
genome_order = []
mercator_genome_order = []
map_lines = []
mercator_order_file = open(args[3])
mercator_genome_order = mercator_order_file.readline().split()
for ln in open(args[0]):
(sp, fasta) = ln[:-1].split()
genome_order.append(sp)
genome_dict[sp] = fasta_subseq_2.FastaDB()
genome_dict[sp].openFastaFile(fasta)
tree_dict = generate_trees(genome_dict, args[1])
for ln in open(args[2]):
map_ln = ln[:-1].split()
(species, map_dict) = make_map_dict(map_ln[1:], genome_order,
mercator_genome_order)
tree_obj = tree_dict[species]
tree_strIO = StringIO()
Phylo.write(tree_obj, tree_strIO, "newick")
fastas = [(sp, genome_dict[sp]) for sp in genome_order
if (sp in species)]
map_entry = {
'map_dict': map_dict,
'tree': tree_strIO.getvalue(),
'map_idx': int(map_ln[0]),
'fastas': fastas
}
map_lines.append(map_entry)
os.chdir(args[4])
pool = mp.Pool(2)
#pool.map(run_aln_mapline, map_lines)
pool.map(run_aln_mapline, map_lines)
print "ALL DONE!"
def _buildAlnTables(self, subAlnObj, alnScore):
species = subAlnObj.getSeqDict().keys()
dt_init = [] # description for alignment table - build as we read spp
exp_size = 0
self.species = species
for (j, sp) in enumerate(species):
# open fasta to get unaligned chromosome lengths & seqs
sp_fasta = fs2.FastaDB()
sp_fasta.openFastaFile(SPP_FASTAS[sp])
chrs = sorted(sp_fasta.keys())
# group for each species' chromosome arrays
sp_grp = self.h5.createGroup(self.h5.root, sp,
"%s Chromosomes" % (sp, ))
self.species_chrs[sp] = {}
# use 1 byte index as chr identifier
self.chr_key_arrays[sp] = [None] * len(chrs)
for (i, ch) in enumerate(chrs):
# build 2 x len(chr) array of (base,aligned_coord) pairs
# print sp_fasta[ch]
bases = np.matrix(list(sp_fasta[ch].getFullSeq()),
dtype=np.dtype("a1")).T
maps = np.matrix(np.zeros(len(sp_fasta[ch]) - 1),
dtype=np.dtype("u8")).T
flat_chr_arr = np.ndarray(shape=(len(maps), 1),
dtype=np.dtype([('base', 'a1'),
('aln_map', 'u8')]))
flat_chr_arr['base'] = bases
flat_chr_arr['aln_map'] = maps
self.species_chrs[sp][ch] = self.h5.createTable(
sp_grp, "chr" + ch,
np.dtype([('base', 'a1'), ('aln_map', 'u8')]))
self.species_chrs[sp][ch].append(flat_chr_arr)
self.species_chrs[sp][ch].flush()
self.chr_key_arrays[sp][i] = [ch, self.species_chrs[sp][ch]]
print "%s %s length = %d added to %s" % (sp, ch, len(bases),
sp_grp)
exp_size += len(sp_fasta[ch])
# add a column to the align table description
dt_init.append((sp, [('base', 'a1'), ('chr_key', 'u4'),
('position', 'u8')]))
self.aln_tbl_dtype = np.dtype(dt_init)
self.aln_table = self.h5.createTable(
self.h5.root,
'aln_table',
self.aln_tbl_dtype,
expectedrows=exp_size) # make the alignment table
self.aln_table.flush()
self.built_chr_tabs = True
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 _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) != 3:
print "usage: patser_annotate_genome_noxgrid.py <genome_seq> <matrix_file> <matrix_name>"
sys.exit(0)
# open fasta
fasta = fasta_subseq_2.FastaDB()
fasta.openFastaFile(args[0])
jobs = []
for (name, chr) in fasta.items():
srch = searchObj(chrObj=chr,
seq_name=name,
matrix=args[1],
matrix_name=args[2])
print srch
jobs.append(srch)
print jobs
pool = mp.Pool()
results = pool.map(search, jobs)
def _fillWGADBFromFile(self):
#print self.h5.root
#print self.h5.root._v_children.items()
self.aln_table = self.h5.root.aln_table
for sp in self.h5.root._v_groups.keys():
self.species_chrs[sp] = {}
self.species.append(sp)
sp_fasta = fs2.FastaDB()
sp_fasta.openFastaFile(SPP_FASTAS[sp])
chrs = sorted(sp_fasta.keys()) #.sorted()
self.chr_key_arrays[sp] = [None] * len(chrs)
for chrom in self.h5.root._v_groups[sp]._v_children.keys():
ch = chrom.replace("chr", "")
#print (ch,chrom)
self.species_chrs[sp][ch] = self.h5.root._v_groups[
sp]._v_children[chrom]
for (i, chrom) in enumerate(chrs):
#print self.species_chrs
try:
self.chr_key_arrays[sp][i] = self.species_chrs[sp][chrom]
except:
print >> sys.stderr, "WARNING: chromsome %s not found" % (
chrom, )
self.built_chr_tabs = True