def convert(motifs_list, alphabet):
"""
Returns a new list of motifs converting using given alphabet.
"""
new_list = []
for motif in motifs_list:
tot_inst = []
# Recreate instances using given alphabet
for i in motif.instances:
inst = meme.Instance(i.tostring(), alphabet)
inst.motif_name = i.motif_name
inst.sequence_name = i.sequence_name
inst.start = int(i.start)
inst.length = int(i.length)
tot_inst.append(inst)
tot_inst = motifs.Instances(tot_inst, alphabet)
# Converting instances and motifs
mot = meme.Motif(alphabet=alphabet, instances=tot_inst)
mot.name = motif.name
mot.evalue = motif.evalue
new_list.append(mot)
return new_list
def __read_database_and_motifs(record, handle):
for line in handle:
if line.startswith('DATABASE AND MOTIFS'):
break
line = next(handle)
if not line.startswith('****'):
raise ValueError("Line does not start with '****':\n%s" % line)
line = next(handle)
if 'DATABASE' not in line:
raise ValueError("Line does not contain 'DATABASE':\n%s" % line)
words = line.strip().split()
record.database = words[1]
if words[2] == '(nucleotide)':
record.alphabet = IUPAC.unambiguous_dna
elif words[2] == '(peptide)':
record.alphabet = IUPAC.protein
for line in handle:
if 'MOTIF WIDTH' in line:
break
line = next(handle)
if '----' not in line:
raise ValueError("Line does not contain '----':\n%s" % line)
for line in handle:
if not line.strip():
break
words = line.strip().split()
motif = meme.Motif(record.alphabet)
motif.name = words[0]
motif.length = int(words[1])
# words[2] contains the best possible match
record.append(motif)
def __read_database_and_motifs(record, handle):
for line in handle:
if line.startswith('DATABASE AND MOTIFS'):
break
line = next(handle)
if not line.startswith('****'):
raise ValueError("Line does not start with '****':\n%s" % line)
line = next(handle)
if 'DATABASE' not in line:
raise ValueError("Line does not contain 'DATABASE':\n%s" % line)
words = line.strip().split()
record.database = words[1]
if words[2] == '(nucleotide)':
record.alphabet = "ACGT"
elif words[2] == '(peptide)':
record.alphabet = "ACDEFGHIKLMNPQRSTVWY"
for line in handle:
if 'WIDTH BEST POSSIBLE MATCH' in line:
break
line = next(handle)
if '----' not in line:
raise ValueError("Line does not contain '----':\n%s" % line)
has_motif_ids = (len(line.strip().split()) == 5)
for line in handle:
if not line.strip():
break
words = line.strip().split()
motif = meme.Motif(record.alphabet)
motif.name = words[0]
if has_motif_ids:
motif.id = words[1]
motif.alt_id = words[2]
motif.length = int(words[-2])
# words[-1] contains the best possible match
record.append(motif)
def __read_metadata(record, xml_tree):
record.version = xml_tree.getroot().get('version')
record.database = xml_tree.find('sequence_dbs').find('sequence_db').get('source')
record.alphabet = xml_tree.find('alphabet').get('name')
record.strand_handling = xml_tree.find('settings').get('strand_handling')
# TODO - read other metadata
for i, motif_tree in enumerate(xml_tree.find('motifs').findall('motif')):
motif = meme.Motif(record.alphabet)
# TODO - motif.name not in XML - always index?
motif.name = str(i + 1)
motif.id = motif_tree.get('id')
motif.alt_id = motif_tree.get('alt')
motif.length = int(motif_tree.get('length'))
# TODO - add nsites, evalue
record.append(motif)
def __read_metadata(record, xml_tree):
record.version = xml_tree.getroot().get("version")
record.database = xml_tree.find("sequence_dbs").find("sequence_db").get(
"source")
record.alphabet = xml_tree.find("alphabet").get("name")
record.strand_handling = xml_tree.find("settings").get("strand_handling")
# TODO - read other metadata
for i, motif_tree in enumerate(xml_tree.find("motifs").findall("motif")):
motif = meme.Motif(record.alphabet)
# TODO - motif.name not in XML - always index?
motif.name = str(i + 1)
motif.id = motif_tree.get("id")
motif.alt_id = motif_tree.get("alt")
motif.length = int(motif_tree.get("length"))
# TODO - add nsites, evalue
record.append(motif)
