def solve(par):
N, K = par
results = []
for i in range(1 << N):
b = Bits(int=i, length=N + 1)
if b.count(1) == K:
results.append(b.bin[1:])
return '\n'.join(results)
def solve(par):
N, K = par
results = []
for i in range(1 << N):
b = Bits(int=i, length=N + 1)
if b.count(1) == K:
results.append(b.bin[1:])
return '\n'.join(results)
def calc_hash(mu, num):
bits = None
if isinstance(mu, str):
bits = Bits(bytes=mu.encode())
elif isinstance(mu, Bits):
bits = mu
bits += bits_from_uint(num)
ones = bits.count(1)
return bits_from_uint(ones), ones
def make_exon_alignment(cursor, ensembl_db_name, human_exon_id, human_exon_known, mitochondrial,
min_similarity, flank_length, first_human_exon = True):
sequence_pep = {}
sequence_dna = {}
shortest_l = -1 # Uninitialized leading padding length
shortest_r = -1 # Uninitialized trailing padding length
pep_aln_length = 0
dna_aln_length = 0
# find all other exons that map to the human exon
maps = get_maps(cursor, ensembl_db_name, human_exon_id, human_exon_known)
maps = filter (lambda m: not m.exon_id_2 is None, maps)
maps_sw = filter (lambda m: m.source=='sw_sharp' or m.source=='usearch', maps)
for map in maps:
if map.similarity < min_similarity: continue
# get the raw (unaligned) sequence for the exon that maps onto human
exon_seqs = get_exon_seqs(cursor, map.exon_id_2, map.exon_known_2, ensembl_db_name[map.species_2])
if (not exon_seqs):
#print " exon_seqs for" , map.source
continue
[pepseq, pepseq_transl_start,
pepseq_transl_end, left_flank, right_flank, dna_seq] = exon_seqs[1:]
# rpl11 starts with an exon that translates into 2 aa's,
# rpl10A has a single methionine (or so they say) followed by a split codon
# *supposedly there is evidence at the protein level
# but will this give me tons of junk elsewhere? ...
pepseq_noX = pepseq.replace ('X','')
if len(pepseq_noX)<3:
# if this is the first exon, and if it starts with M, we'll let it off the hook
# abd then if it's human, we'll also salvage it at any price
if first_human_exon and pepseq_noX[0] == 'M' or map.species_2=='homo_sapiens':
pass
else:
continue
# check
dnaseq = Seq (dna_seq[pepseq_transl_start:pepseq_transl_end], generic_dna)
if (mitochondrial):
pepseq2 = dnaseq.translate(table="Vertebrate Mitochondrial").tostring()
else:
pepseq2 = dnaseq.translate().tostring()
if (not pepseq == pepseq2):
continue
# inflate the compressed sequence
if not map.bitmap:
continue
bs = Bits(bytes=map.bitmap)
if (not bs.count(1) == len(pepseq)): continue # check bitmap has correct number of 1s
usi = iter(pepseq)
#reconst_pepseq = "".join(('-' if c=='0' else next(usi) for c in bs.bin))
reconst_pepseq = ''
for c in bs.bin:
if c == '0': reconst_pepseq += '-'
else: reconst_pepseq += next(usi)
# come up with a unique name for this sequence
species = map.species_2
# let's also have the start in gene here - might make our lives easier later
exon2 = get_exon (cursor, map.exon_id_2, map.exon_known_2, ensembl_db_name[species])
sequence_name = species + "_" + str(map.exon_id_2)+"_"+str(map.exon_known_2)+"_"+str(exon2.start_in_gene)
if reconst_pepseq:
sequence_pep[sequence_name] = reconst_pepseq
pep_aln_length = len(reconst_pepseq)
reconst_ntseq = expand_pepseq (reconst_pepseq, exon_seqs[1:], flank_length)
if reconst_ntseq:
sequence_dna[sequence_name] = reconst_ntseq
dna_aln_length = len(reconst_ntseq)
# strip common gaps
sequence_stripped_pep = strip_gaps (sequence_pep)
if not sequence_stripped_pep:
c=inspect.currentframe()
#print " in %s:%d" % ( c.f_code.co_filename, c.f_lineno)
return ['','']
# strip common gaps
sequence_stripped_dna = strip_gaps (sequence_dna)
if not sequence_stripped_dna:
c=inspect.currentframe()
#print " in %s:%d" % ( c.f_code.co_filename, c.f_lineno)
return ['', '']
return [sequence_stripped_pep, sequence_stripped_dna]
def make_exon_alignment(cursor, ensembl_db_name, human_exon_id, human_exon_known, mitochondrial,
min_similarity, flank_length):
sequence_pep = {}
sequence_dna = {}
shortest_l = -1 # Uninitialized leading padding length
shortest_r = -1 # Uninitialized trailing padding length
pep_aln_length = 0
dna_aln_length = 0
# find all other exons that map to the human exon
maps = get_maps(cursor, ensembl_db_name, human_exon_id, human_exon_known)
maps = filter (lambda m: not m.exon_id_2 is None, maps)
maps_sw = filter (lambda m: m.source=='sw_sharp' or m.source=='usearch', maps)
for map in maps:
if map.similarity < min_similarity: continue
# get the raw (unaligned) sequence for the exon that maps onto human
exon_seqs = get_exon_seqs(cursor, map.exon_id_2, map.exon_known_2, ensembl_db_name[map.species_2])
if (not exon_seqs):
print " exon_seqs for" , map.source
exit(1)
continue
[pepseq, pepseq_transl_start,
pepseq_transl_end, left_flank, right_flank, dna_seq] = exon_seqs[1:]
if len(pepseq)<3: continue
pepseq_noX = pepseq.replace ('X','')
if len(pepseq_noX)<3: continue
# check
dnaseq = Seq (dna_seq[pepseq_transl_start:pepseq_transl_end], generic_dna)
if (mitochondrial):
pepseq2 = dnaseq.translate(table="Vertebrate Mitochondrial").tostring()
else:
pepseq2 = dnaseq.translate().tostring()
if (not pepseq == pepseq2):
continue
# inflate the compressed sequence
if not map.bitmap:
continue
bs = Bits(bytes=map.bitmap)
if (not bs.count(1) == len(pepseq)): continue # check bitmap has correct number of 1s
usi = iter(pepseq)
#reconst_pepseq = "".join(('-' if c=='0' else next(usi) for c in bs.bin))
reconst_pepseq = ''
for c in bs.bin:
if c == '0': reconst_pepseq += '-'
else: reconst_pepseq += next(usi)
# come up with a unique name for this sequence
species = map.species_2
sequence_name = species + "_" + str(map.exon_id_2)+"_"+str(map.exon_known_2)
if reconst_pepseq:
sequence_pep[sequence_name] = reconst_pepseq
pep_aln_length = len(reconst_pepseq)
reconst_ntseq = expand_pepseq (reconst_pepseq, exon_seqs[1:], flank_length)
if reconst_ntseq:
sequence_dna[sequence_name] = reconst_ntseq
dna_aln_length = len(reconst_ntseq)
# strip common gaps
sequence_stripped_pep = strip_gaps (sequence_pep)
if not sequence_stripped_pep:
c=inspect.currentframe()
print " in %s:%d" % ( c.f_code.co_filename, c.f_lineno)
exit(1)
# strip common gaps
sequence_stripped_dna = strip_gaps (sequence_dna)
if not sequence_stripped_dna:
c=inspect.currentframe()
print " in %s:%d" % ( c.f_code.co_filename, c.f_lineno)
exit(1)
return [sequence_stripped_pep, sequence_stripped_dna]
def main():
no_threads = 1
special = None
if len(sys.argv) > 1 and len(sys.argv) < 3:
print "usage: %s <set name> <number of threads> " % sys.argv[0]
exit(1)
elif len(sys.argv) == 3:
special = sys.argv[1]
special = special.lower()
if special == 'none': special = None
no_threads = int(sys.argv[2])
db = connect_to_mysql()
cfg = ConfigurationReader()
cursor = db.cursor()
# find db ids adn common names for each species db
[all_species, ensembl_db_name] = get_species(cursor)
species = 'homo_sapiens'
switch_to_db(cursor, ensembl_db_name[species])
if special:
print "using", special, "set"
gene_list = get_theme_ids(cursor, ensembl_db_name, cfg, special)
else:
print "using all protein coding genes"
switch_to_db(cursor, ensembl_db_name['homo_sapiens'])
gene_list = get_gene_ids(cursor, biotype='protein_coding', is_known=1)
incomplete = 0
genes_checked = 0
#for gene_id in gene_list:
#for gene_id in [743609]:
for sampling_count in range(1000):
gene_id = choice(gene_list)
genes_checked += 1
with_map = 0
tot = 0
switch_to_db(cursor, ensembl_db_name['homo_sapiens'])
print gene2stable(cursor, gene_id), get_description(cursor, gene_id)
# find all exons we are tracking in the database
human_exons = gene2exon_list(cursor, gene_id)
human_exons.sort(key=lambda exon: exon.start_in_gene)
has_a_map = False
for human_exon in human_exons:
if (not human_exon.is_canonical or not human_exon.is_coding):
continue
if verbose:
print
print "\t human", human_exon.exon_id, human_exon.is_known
print "\t ", get_exon_pepseq(cursor, human_exon,
ensembl_db_name['homo_sapiens'])
print "\t checking maps ..."
maps = get_maps(cursor, ensembl_db_name, human_exon.exon_id,
human_exon.is_known)
tot += 1
if maps:
has_a_map = True
with_map += 1
#print "ok"
else:
print "no maps for exon", human_exon.exon_id
continue
if verbose:
for map in maps:
species = map.species_2
exon = map2exon(cursor, ensembl_db_name, map)
unaligned_sequence = get_exon_pepseq(
cursor, exon, ensembl_db_name[species])
if (map.similarity):
print "\t", species, map.source, map.exon_id_2, map.exon_known_2
print "\tmaps to ", map.exon_id_1, map.exon_known_1
print "\tsim", map.similarity,
print "\tsource", map.source
print "\t", unaligned_sequence
if not map.bitmap:
print "\t bitmap not assigned"
else:
bs = Bits(bytes=map.bitmap)
reconst_pepseq = ''
if (not bs.count(1) == len(unaligned_sequence)):
print "\talnd seq mismatch"
else:
usi = iter(unaligned_sequence)
for c in bs.bin:
if c == '0': reconst_pepseq += '-'
else: reconst_pepseq += next(usi)
print "\tbinary : ", bs.bin
print "\talnd seq: ", reconst_pepseq
print
if not tot == with_map:
print "#### gene id: %d total exons: %d with map: %d ( = %d%%) " % \
(gene_id, tot, with_map, int(float(with_map)/tot*100) )
incomplete += 1
print "genes checked: %d, incomplete: %d" % (genes_checked, incomplete)
cursor.close()
db.close()
print tot, with_map
def main():
no_threads = 1
special = None
if len(sys.argv) > 1 and len(sys.argv)<3:
print "usage: %s <set name> <number of threads> " % sys.argv[0]
exit(1)
elif len(sys.argv)==3:
special = sys.argv[1]
special = special.lower()
if special == 'none': special = None
no_threads = int(sys.argv[2])
db = connect_to_mysql()
cfg = ConfigurationReader()
cursor = db.cursor()
# find db ids adn common names for each species db
[all_species, ensembl_db_name] = get_species (cursor)
species = 'homo_sapiens'
switch_to_db (cursor, ensembl_db_name[species])
if special:
print "using", special, "set"
gene_list = get_theme_ids (cursor, ensembl_db_name, cfg, special )
else:
print "using all protein coding genes"
switch_to_db (cursor, ensembl_db_name['homo_sapiens'])
gene_list = get_gene_ids (cursor, biotype='protein_coding', is_known=1)
incomplete = 0
genes_checked = 0
#for gene_id in gene_list:
#for gene_id in [743609]:
for sampling_count in range(1000):
gene_id = choice(gene_list)
genes_checked += 1
with_map = 0
tot = 0
switch_to_db (cursor, ensembl_db_name['homo_sapiens'])
print gene2stable(cursor, gene_id), get_description (cursor, gene_id)
# find all exons we are tracking in the database
human_exons = gene2exon_list(cursor, gene_id)
human_exons.sort(key=lambda exon: exon.start_in_gene)
has_a_map = False
for human_exon in human_exons:
if (not human_exon.is_canonical or not human_exon.is_coding): continue
if verbose:
print
print "\t human", human_exon.exon_id, human_exon.is_known
print "\t ", get_exon_pepseq(cursor, human_exon, ensembl_db_name['homo_sapiens'])
print "\t checking maps ..."
maps = get_maps(cursor, ensembl_db_name, human_exon.exon_id, human_exon.is_known)
tot += 1
if maps:
has_a_map = True
with_map += 1
#print "ok"
else:
print"no maps for exon", human_exon.exon_id
continue
if verbose:
for map in maps:
species = map.species_2
exon = map2exon(cursor, ensembl_db_name, map)
unaligned_sequence = get_exon_pepseq(cursor, exon, ensembl_db_name[species])
if ( map.similarity):
print "\t", species, map.source, map.exon_id_2, map.exon_known_2
print "\tmaps to ", map.exon_id_1, map.exon_known_1
print "\tsim", map.similarity,
print "\tsource", map.source
print "\t", unaligned_sequence
if not map.bitmap:
print "\t bitmap not assigned"
else:
bs = Bits(bytes=map.bitmap)
reconst_pepseq = ''
if (not bs.count(1) == len(unaligned_sequence)):
print "\talnd seq mismatch"
else:
usi = iter(unaligned_sequence)
for c in bs.bin:
if c == '0': reconst_pepseq += '-'
else: reconst_pepseq += next(usi)
print "\tbinary : ", bs.bin
print "\talnd seq: ", reconst_pepseq
print
if not tot== with_map:
print "#### gene id: %d total exons: %d with map: %d ( = %d%%) " % \
(gene_id, tot, with_map, int(float(with_map)/tot*100) )
incomplete += 1
print "genes checked: %d, incomplete: %d" % (genes_checked, incomplete)
cursor.close()
db.close()
print tot, with_map
