def build(self, cursor):
switch_to_db(cursor, get_compara_name(cursor))
for leaf in self.leafs:
leaf.tax_id = species2taxid(cursor, leaf.name)
leaf.is_leaf = True
self.node[leaf.tax_id] = leaf
# build the tree using ncbi_tax.nodes
# fill in the names using ncbi_tax.names
switch_to_db(cursor, get_ncbi_tax_name(cursor))
for leaf in self.leafs:
parent_id = taxid2parentid(cursor, leaf.tax_id)
leaf.parent_id = parent_id
current_id = leaf.tax_id
# move to the root
while current_id:
current_node = self.node[current_id]
parent_id = taxid2parentid(cursor, parent_id)
if (not parent_id or current_id == parent_id):
current_node.is_root = True
self.root = self.node[current_id]
current_id = None
else:
# does parent exist by any chance
if self.node.has_key(parent_id):
parent_node = self.node[parent_id]
parent_node.children.append(current_node)
# we are done here
current_id = None
else: # make new node
parent_name = taxid2name(cursor, parent_id)
parent_node = Node(parent_name)
parent_node.tax_id = parent_id
# grampa:
parent_node.parent_id = taxid2parentid(
cursor, parent_id)
parent_node.children.append(current_node)
self.node[parent_id] = parent_node
# attach the current node to the parent
current_id = parent_id
# shortcircuit nodes with a single child
new_root = self.root.__cleanup__()
if (new_root):
new_root.is_root = True
self.root = new_root
del_ids = []
for node_id, node in self.node.iteritems():
if node.is_leaf:
continue
if (not node.children):
del_ids.append(node_id)
for node_id in del_ids:
del self.node[node_id]
self.__set_parent_ids__(self.root)
def find_mammals(cursor, trivial_name_list):
mammals = []
for trivial_name in trivial_name_list:
switch_to_db(cursor, get_compara_name(cursor))
tax_id = trivial2taxid (cursor, trivial_name)
parent_id = taxid2parentid (cursor, tax_id)
tax_id = parent_id
is_mammal = False
while tax_id:
qry = "select name_txt from names where tax_id= %d " % int(tax_id)
qry += " and name_class = 'scientific name'";
rows = search_db (cursor, qry)
if rows and rows[0][0]:
if 'mammal' in rows[0][0].lower():
is_mammal = True
break
elif 'vertebrat' in rows[0][0].lower():
# if the thing wasa mammal, we would have found it by now
is_mammal = False
break
parent_id = taxid2parentid (cursor, tax_id)
if parent_id and parent_id>1:
tax_id = parent_id
else:
tax_id = None
if is_mammal:
mammals.append(trivial_name)
return mammals
def find_mammals(cursor, trivial_name_list):
mammals = []
for trivial_name in trivial_name_list:
switch_to_db(cursor, get_compara_name(cursor))
tax_id = trivial2taxid(cursor, trivial_name)
parent_id = taxid2parentid(cursor, tax_id)
tax_id = parent_id
is_mammal = False
while tax_id:
qry = "select name_txt from names where tax_id= %d " % int(tax_id)
qry += " and name_class = 'scientific name'"
rows = search_db(cursor, qry)
if rows and rows[0][0]:
if 'mammal' in rows[0][0].lower():
is_mammal = True
break
elif 'vertebrat' in rows[0][0].lower():
# if the thing wasa mammal, we would have found it by now
is_mammal = False
break
parent_id = taxid2parentid(cursor, tax_id)
if parent_id and parent_id > 1:
tax_id = parent_id
else:
tax_id = None
if is_mammal:
mammals.append(trivial_name)
return mammals
def build (self, cursor):
switch_to_db(cursor, get_compara_name(cursor))
for leaf in self.leafs:
leaf.tax_id = species2taxid (cursor, leaf.name)
leaf.is_leaf = True
self.node[leaf.tax_id] = leaf
# build the tree using ncbi_tax.nodes
# fill in the names using ncbi_tax.names
switch_to_db(cursor, get_ncbi_tax_name(cursor))
for leaf in self.leafs:
parent_id = taxid2parentid (cursor, leaf.tax_id)
leaf.parent_id = parent_id
current_id = leaf.tax_id
# move to the root
while current_id:
current_node = self.node[current_id]
parent_id = taxid2parentid (cursor, parent_id)
if (not parent_id or current_id == parent_id):
current_node.is_root = True
self.root = self.node[current_id]
current_id = None
else:
# does parent exist by any chance
if self.node.has_key(parent_id):
parent_node = self.node[parent_id]
parent_node.children.append(current_node)
# we are done here
current_id = None
else: # make new node
parent_name = taxid2name(cursor, parent_id)
parent_node = Node(parent_name)
parent_node.tax_id = parent_id
# grampa:
parent_node.parent_id = taxid2parentid (cursor, parent_id)
parent_node.children.append(current_node)
self.node[parent_id] = parent_node
# attach the current node to the parent
current_id = parent_id
# shortcircuit nodes with a single child
new_root = self.root.__cleanup__()
if (new_root):
new_root.is_root = True
self.root = new_root
del_ids = []
for node_id, node in self.node.iteritems():
if node.is_leaf:
continue
if (not node.children):
del_ids.append(node_id)
for node_id in del_ids:
del self.node[node_id]
self.__set_parent_ids__ (self.root)
def add(self, cursor, name):
leaf = Node(name)
# get tax ids from the gene_db table in compara database
switch_to_db(cursor, get_compara_name(cursor))
leaf.tax_id = species2taxid (cursor, leaf.name)
leaf.is_leaf = True
self.leafs.append(leaf)
self.node[leaf.tax_id] = leaf
def add(self, cursor, name):
leaf = Node(name)
# get tax ids from the gene_db table in compara database
switch_to_db(cursor, get_compara_name(cursor))
leaf.tax_id = species2taxid(cursor, leaf.name)
leaf.is_leaf = True
self.leafs.append(leaf)
self.node[leaf.tax_id] = leaf
def taxid2name(cursor, tax_id):
switch_to_db(cursor, get_ncbi_tax_name(cursor))
qry = "select name_txt from names where tax_id= %d " % int(tax_id)
qry += " and name_class = 'scientific name'"
rows = search_db(cursor, qry)
if (not rows):
rows = search_db(cursor, qry, verbose=True)
return ""
return rows[0][0]
def trivial2taxid(cursor, trivial_name):
switch_to_db(cursor, get_ncbi_tax_name(cursor))
qry = "select tax_id from names where name_txt= '%s' " % trivial_name
qry += " and name_class = 'trivial'"
rows = search_db(cursor, qry)
if (not rows or 'ERROR' in rows[0]):
rows = search_db(cursor, qry, verbose=True)
return ""
return int(rows[0][0])
def taxid2trivial (cursor, tax_id):
switch_to_db (cursor, get_ncbi_tax_name (cursor))
qry = "select name_txt from names where tax_id= %d " % int(tax_id)
qry += " and name_class = 'trivial'";
rows = search_db (cursor, qry)
if (not rows or 'ERROR' in rows[0]):
rows = search_db (cursor, qry, verbose = True)
return ""
return rows[0][0]
def trivial2taxid (cursor, trivial_name):
switch_to_db (cursor, get_ncbi_tax_name (cursor))
qry = "select tax_id from names where name_txt= '%s' " % trivial_name
qry += " and name_class = 'trivial'";
rows = search_db (cursor, qry)
if (not rows or 'ERROR' in rows[0]):
rows = search_db (cursor, qry, verbose = True)
return ""
return int(rows[0][0])
def taxid2name (cursor, tax_id):
switch_to_db (cursor, get_ncbi_tax_name (cursor))
qry = "select name_txt from names where tax_id= %d " % int(tax_id)
qry += " and name_class = 'scientific name'";
rows = search_db (cursor, qry)
if (not rows):
rows = search_db (cursor, qry, verbose = True)
return ""
return rows[0][0]
def taxid2trivial(cursor, tax_id):
switch_to_db(cursor, get_ncbi_tax_name(cursor))
qry = "select name_txt from names where tax_id= %d " % int(tax_id)
qry += " and name_class = 'trivial'"
rows = search_db(cursor, qry)
if (not rows or 'ERROR' in rows[0]):
rows = search_db(cursor, qry, verbose=True)
return ""
return rows[0][0]
def species2genome_db_id(cursor, species):
switch_to_db(cursor, get_compara_name(cursor))
qry = "select genome_db_id from genome_db where name = '%s'" % species
rows = search_db(cursor, qry)
if not rows:
search_db(cursor, qry, verbose=True)
return 0
return int(rows[0][0])
def genome_db_id2species(cursor, genome_db_id):
switch_to_db(cursor, get_compara_name(cursor))
qry = "select name from genome_db where genome_db_id = %d" % int(genome_db_id)
rows = search_db(cursor, qry)
if not rows:
search_db(cursor, qry, verbose=True)
return ""
return rows[0][0]
def taxid2parentid (cursor, tax_id):
switch_to_db (cursor, get_ncbi_tax_name (cursor))
qry = "select parent_tax_id from nodes where tax_id= %d " % int(tax_id)
rows = search_db (cursor, qry)
if (not rows):
rows = search_db (cursor, qry, verbose = True)
return ""
try:
retval = int(rows[0][0])
except:
retval = ""
return retval
def taxid2parentid(cursor, tax_id):
switch_to_db(cursor, get_ncbi_tax_name(cursor))
qry = "select parent_tax_id from nodes where tax_id= %d " % int(tax_id)
rows = search_db(cursor, qry)
if (not rows):
rows = search_db(cursor, qry, verbose=True)
return ""
try:
retval = int(rows[0][0])
except:
retval = ""
return retval
def genome_db_id2species (cursor, genome_db_id):
switch_to_db (cursor, get_compara_name (cursor))
qry = "select name from genome_db where genome_db_id = %d" % int(genome_db_id)
rows = search_db (cursor, qry)
if (not rows):
search_db (cursor, qry, verbose = True)
return ""
return rows[0][0]
def species2genome_db_id (cursor, species):
switch_to_db (cursor, get_compara_name (cursor))
qry = "select genome_db_id from genome_db where name = '%s'" % species
rows = search_db (cursor, qry)
if (not rows):
search_db (cursor, qry, verbose = True)
return 0
return int(rows[0][0])
def species2taxid(cursor, species):
switch_to_db(cursor, get_compara_name(cursor))
qry = "select taxon_id from genome_db where name = '%s'" % species
rows = search_db(cursor, qry)
if not rows:
search_db(cursor, qry, verbose=True)
return ""
try:
retval = int(rows[0][0])
except:
retval = ""
return retval
def species2taxid (cursor, species):
switch_to_db (cursor, get_compara_name (cursor))
qry = "select taxon_id from genome_db where name = '%s'" % species
rows = search_db (cursor, qry)
if (not rows):
search_db (cursor, qry, verbose = True)
return ""
try:
retval = int(rows[0][0])
except:
retval = ""
return retval
def get_canonical_coding_exons (cursor, gene_id, db_name=None):
if db_name and not switch_to_db(cursor, db_name):
return []
all_exons = gene2exon_list (cursor, gene_id)
if not all_exons: return []
exons = filter (lambda x: x.is_coding and x.is_canonical, all_exons)
if not exons: return []
# now, the problem is that an exon can be coding,
# but not in the canonical version of the transcript
exons.sort(key=lambda exon: exon.start_in_gene)
if not exons: return []
# is there info about the beginning and the end of canonical translation?
canonical_transcript_id = get_canonical_transcript_id (cursor, gene_id, db_name=None)
if not canonical_transcript_id: return []
ret = get_canonical_coordinates (cursor, canonical_transcript_id)
if not ret or not len(ret) == 4: return []
[canonical_start_in_exon, canonical_start_exon_id,
canonical_end_in_exon, canonical_end_exon_id] = ret
if canonical_start_exon_id is None or canonical_end_exon_id is None: return []
# filter the exons that are within the start and end bracket
canonical_exons = []
reading = 0
for exon in exons:
if exon.exon_id == canonical_start_exon_id or exon.exon_id == canonical_end_exon_id:
reading = 1 - reading
canonical_exons.append(exon)
elif reading:
canonical_exons.append(exon)
return canonical_exons
def get_common_name(cursor, species):
switch_to_db(cursor, get_compara_name(cursor))
tax_id = species2taxid(cursor, species)
switch_to_db(cursor, get_ncbi_tax_name(cursor))
qry = "select name_txt from names where "
qry += "tax_id = %d and " % tax_id
qry += "name_class = 'genbank common name'"
rows = search_db(cursor, qry)
if rows:
if ('ERROR' in rows[0]):
search_db(cursor, qry, verbose=True)
return ""
else:
return rows[0][0]
return ""
def get_common_name (cursor, species):
switch_to_db(cursor, get_compara_name(cursor))
tax_id = species2taxid (cursor, species)
switch_to_db(cursor,get_ncbi_tax_name (cursor))
qry = "select name_txt from names where "
qry += "tax_id = %d and " % tax_id
qry += "name_class = 'genbank common name'"
rows = search_db (cursor, qry)
if rows:
if ('ERROR' in rows[0]):
search_db (cursor, qry, verbose = True)
return ""
else:
return rows[0][0]
return ""
def get_exon_pepseq(cursor, exon, db_name=None, verbose=False):
if db_name:
if not switch_to_db(cursor, db_name):
return False
if exon.exon_seq_id:
exon_seq_id = exon.exon_seq_id
qry = "select protein_seq "
qry += " from exon_seq where exon_seq_id = %d" % exon_seq_id
else:
exon_id = exon.exon_id
is_known = exon.is_known
qry = "select protein_seq "
qry += " from exon_seq where exon_id = %d and is_known = %d" % (exon_id, is_known)
rows = search_db(cursor, qry)
if not rows:
if verbose:
rows = search_db(cursor, qry, verbose=True)
print rows
return ""
protein_seq = rows[0][0]
if protein_seq is None:
protein_seq = ""
return protein_seq
def gene2exon_list(cursor, gene_id, db_name=None, verbose=False):
exons = []
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "select * from gene2exon where gene_id = %d " % gene_id
rows = search_db(cursor, qry)
if not rows:
rows = search_db(cursor, "select database()")
if verbose:
print "database ", rows[0][0]
rows = search_db(cursor, qry, verbose=True)
print rows
return []
for row in rows:
exon = Exon()
if not exon.load_from_gene2exon(row):
continue
exons.append(exon)
return exons
def get_exon_pepseq (cursor, exon, db_name=None, verbose=False):
if (db_name):
if not switch_to_db(cursor, db_name):
return False
if exon.exon_seq_id:
exon_seq_id = exon.exon_seq_id
qry = "select protein_seq "
qry += " from exon_seq where exon_seq_id = %d" % exon_seq_id
else:
exon_id = exon.exon_id
is_known = exon.is_known
qry = "select protein_seq "
qry += " from exon_seq where exon_id = %d and is_known = %d" % (exon_id, is_known)
rows = search_db(cursor, qry)
if (not rows):
if verbose:
rows = search_db(cursor, qry, verbose = True)
print rows
return ""
protein_seq = rows[0][0]
if (protein_seq is None):
protein_seq = ""
return protein_seq
def get_exon (cursor, exon_id, is_known=None, db_name=None):
exon = Exon ()
if (db_name):
if not switch_to_db(cursor, db_name):
return exon
if is_known==2:
# sw# exon
qry = "select * from sw_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry, verbose=False)
if (not rows):
return exon
exon.load_from_novel_exon (rows[0], "sw_exon")
elif is_known==3:
# sw# exon
qry = "select * from usearch_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry, verbose=False)
if (not rows):
return exon
exon.load_from_novel_exon (rows[0], "usearch_exon")
else:
qry = "select * from gene2exon where exon_id = %d" % exon_id
if is_known: qry += " and is_known = %s " % is_known
rows = search_db(cursor, qry, verbose=False)
if (not rows):
return exon
exon.load_from_gene2exon (rows[0])
return exon
def gene2exon_list (cursor, gene_id, db_name=None, verbose=False):
exons = []
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "select * from gene2exon where gene_id = %d " % gene_id
rows = search_db(cursor, qry)
if (not rows):
rows = search_db(cursor, 'select database()')
if verbose:
print "database ", rows[0][0]
rows = search_db(cursor, qry, verbose = True)
print rows
return []
for row in rows:
exon = Exon()
if (not exon.load_from_gene2exon(row)):
continue
exons.append(exon)
return exons
def get_exon_seq_by_db_id (cursor, exon_seq_id, db_name=None):
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "select exon_seq_id, protein_seq, pepseq_transl_start, pepseq_transl_end, "
qry += " left_flank, right_flank, dna_seq "
qry += " from exon_seq where exon_seq_id = %d" % exon_seq_id
rows = search_db(cursor, qry)
if (not rows):
#rows = search_db(cursor, qry, verbose = True)
return []
[exon_seq_id, protein_seq, pepseq_transl_start,
pepseq_transl_end, left_flank, right_flank, dna_seq] = rows[0]
if (protein_seq is None):
protein_seq = ""
if (left_flank is None):
left_flank = ""
if (right_flank is None):
right_flank = ""
if (dna_seq is None):
dna_seq = ""
return [exon_seq_id, protein_seq, pepseq_transl_start, pepseq_transl_end, left_flank, right_flank, dna_seq]
def get_exon_seq_by_db_id(cursor, exon_seq_id, db_name=None):
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "select exon_seq_id, protein_seq, pepseq_transl_start, pepseq_transl_end, "
qry += " left_flank, right_flank, dna_seq "
qry += " from exon_seq where exon_seq_id = %d" % exon_seq_id
rows = search_db(cursor, qry)
if not rows:
# rows = search_db(cursor, qry, verbose = True)
return []
[exon_seq_id, protein_seq, pepseq_transl_start, pepseq_transl_end, left_flank, right_flank, dna_seq] = rows[0]
if protein_seq is None:
protein_seq = ""
if left_flank is None:
left_flank = ""
if right_flank is None:
right_flank = ""
if dna_seq is None:
dna_seq = ""
return [exon_seq_id, protein_seq, pepseq_transl_start, pepseq_transl_end, left_flank, right_flank, dna_seq]
def get_exon(cursor, exon_id, is_known=None, db_name=None):
exon = Exon()
if db_name:
if not switch_to_db(cursor, db_name):
return exon
if is_known == 2:
# sw# exon
qry = "select * from sw_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry, verbose=False)
if not rows:
return exon
exon.load_from_novel_exon(rows[0], "sw_exon")
elif is_known == 3:
# sw# exon
qry = "select * from usearch_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry, verbose=False)
if not rows:
return exon
exon.load_from_novel_exon(rows[0], "usearch_exon")
else:
qry = "select * from gene2exon where exon_id = %d" % exon_id
if is_known:
qry += " and is_known = %s " % is_known
rows = search_db(cursor, qry, verbose=False)
if not rows:
return exon
exon.load_from_gene2exon(rows[0])
return exon
def get_maps(cursor, ensembl_db_name, exon_id, is_known):
maps = []
switch_to_db (cursor, ensembl_db_name['homo_sapiens'])
qry = "select * from exon_map where exon_id = %d " % exon_id
qry += " and exon_known = %d " % is_known
rows = search_db (cursor, qry)
if not rows or "ERROR" in rows[0]:
return []
for row in rows:
map = Map()
map.load_from_db(row, cursor)
maps.append(map)
return maps
def get_pepseq_transl_range (cursor, exon_id, exon_known, db_name=None):
if db_name and not switch_to_db(cursor, db_name):
return []
qry = "select pepseq_transl_start, pepseq_transl_end from exon_seq "
qry += "where exon_id = %d " % int(exon_id)
qry += "and is_known = %d " % int(exon_known)
if not rows:
return []
return rows[0]
def get_pepseq_transl_range(cursor, exon_id, exon_known, db_name=None):
if db_name and not switch_to_db(cursor, db_name):
return []
qry = "select pepseq_transl_start, pepseq_transl_end from exon_seq "
qry += "where exon_id = %d " % int(exon_id)
qry += "and is_known = %d " % int(exon_known)
if not rows:
return []
return rows[0]
def exon_id2gene_id (cursor, ensembl_db_name, exon_id, is_known):
switch_to_db(cursor, ensembl_db_name)
if is_known==3: # sw_sharp exon
qry = "select gene_id from usearch_exon where "
qry += "exon_id = %d " % exon_id
elif is_known==2: # sw_sharp exon
qry = "select gene_id from sw_exon where "
qry += "exon_id = %d " % exon_id
else:
qry = "select gene_id from gene2exon where "
qry += "exon_id = %s and is_known = %s " % (exon_id, is_known)
rows = search_db (cursor, qry)
if (not rows or 'ERROR' in rows[0]):
rows = search_db (cursor, qry, verbose = True)
return ""
return rows[0][0]
def exon_id2gene_id(cursor, ensembl_db_name, exon_id, is_known):
switch_to_db(cursor, ensembl_db_name)
if is_known == 3: # sw_sharp exon
qry = "select gene_id from usearch_exon where "
qry += "exon_id = %d " % exon_id
elif is_known == 2: # sw_sharp exon
qry = "select gene_id from sw_exon where "
qry += "exon_id = %d " % exon_id
else:
qry = "select gene_id from gene2exon where "
qry += "exon_id = %s and is_known = %s " % (exon_id, is_known)
rows = search_db(cursor, qry)
if not rows or "ERROR" in rows[0]:
rows = search_db(cursor, qry, verbose=True)
return ""
return rows[0][0]
def is_coding_exon (cursor, exon_id, is_known, db_name=None):
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "select is_coding from gene2exon where exon_id = %d and is_known = %d" % (exon_id, is_known)
rows = search_db (cursor, qry)
if ( not rows):
return False
return rows[0][0]>0
def exon_seq_id2exon_id(cursor, exon_seq_id, db_name=None):
if db_name:
if not switch_to_db(cursor, db_name):
return ""
qry = "select exon_id, is_known from exon_seq where exon_seq_id = %d " % int(exon_seq_id)
rows = search_db(cursor, qry)
if not rows:
return ""
return rows[0]
def is_coding_exon(cursor, exon_id, is_known, db_name=None):
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "select is_coding from gene2exon where exon_id = %d and is_known = %d" % (exon_id, is_known)
rows = search_db(cursor, qry)
if not rows:
return False
return rows[0][0] > 0
def get_description(cursor, gene_id, db_name=None):
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "select description from gene where gene_id = %d " % int(gene_id)
rows = search_db(cursor, qry)
if rows:
return rows[0][0]
return ""
def get_description (cursor, gene_id, db_name = None):
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "select description from gene where gene_id = %d " % int(gene_id)
rows = search_db(cursor, qry)
if rows:
return rows[0][0]
return ""
def exon_seq_id2exon_id (cursor, exon_seq_id, db_name=None):
if (db_name):
if not switch_to_db(cursor, db_name):
return ""
qry = "select exon_id, is_known from exon_seq where exon_seq_id = %d " % int(exon_seq_id)
rows = search_db(cursor, qry)
if (not rows):
return ""
return rows[0]
def get_sw_seq_id (cursor, exon_id, db_name=None):
if (db_name):
if not switch_to_db(cursor, db_name):
return -1
qry = "select exon_seq_id "
qry += " from sw_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry)
if not rows or not rows[0][0]:
return -1
return int(rows[0][0])
def get_sw_seq_id(cursor, exon_id, db_name=None):
if db_name:
if not switch_to_db(cursor, db_name):
return -1
qry = "select exon_seq_id "
qry += " from sw_exon where exon_id = %d" % exon_id
rows = search_db(cursor, qry)
if not rows or not rows[0][0]:
return -1
return int(rows[0][0])
def get_status (cursor, exon_id, db_name=None):
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "select status from gene where gene_id = %d " % int(exon_id)
rows = search_db (cursor, qry)
if ( not rows):
return False
return rows[0][0]
def get_status(cursor, exon_id, db_name=None):
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "select status from gene where gene_id = %d " % int(exon_id)
rows = search_db(cursor, qry)
if not rows:
return False
return rows[0][0]
def get_gene_biotype(cursor, gene_id, db_name=None):
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "select biotype from gene where gene_id = %d " % int(gene_id)
rows = search_db(cursor, qry)
if not rows:
return ""
return rows[0][0]
def get_gene_biotype (cursor, gene_id, db_name=None):
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "select biotype from gene where gene_id = %d " % int(gene_id)
rows = search_db (cursor, qry)
if ( not rows):
return ""
return rows[0][0]
def is_coding (cursor, gene_id, db_name=None):
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "select is_coding from gene where gene_id = %d " % int(gene_id)
rows = search_db (cursor, qry)
if ( not rows):
return False
return rows[0][0]>0
def trivial2scientific(cursor, trivial):
switch_to_db(cursor, get_ncbi_tax_name(cursor))
qry = "select tax_id from names where "
qry += "name_txt = '%s' and " % trivial
qry += "name_class = 'trivial'"
rows = search_db(cursor, qry)
if rows:
if ('ERROR' in rows[0]):
search_db(cursor, qry, verbose=True)
return ""
else:
try:
tax_id = int(rows[0][0])
except:
return ""
sciname = taxid2name(cursor, tax_id).lower().replace(" ", "_")
#canis_lupus_familiaris - don't know what to do with it
sciname = sciname.replace("_familiaris", "")
return sciname
return ""
def trivial2scientific (cursor, trivial):
switch_to_db(cursor,get_ncbi_tax_name (cursor))
qry = "select tax_id from names where "
qry += "name_txt = '%s' and " % trivial
qry += "name_class = 'trivial'"
rows = search_db (cursor, qry)
if rows:
if ('ERROR' in rows[0]):
search_db (cursor, qry, verbose = True)
return ""
else:
try:
tax_id = int(rows[0][0])
except:
return ""
sciname = taxid2name(cursor, tax_id).lower().replace(" ", "_")
#canis_lupus_familiaris - don't know what to do with it
sciname = sciname.replace("_familiaris", "")
return sciname
return ""
def is_coding(cursor, gene_id, db_name=None):
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "select is_coding from gene where gene_id = %d " % int(gene_id)
rows = search_db(cursor, qry)
if not rows:
return False
return rows[0][0] > 0
def get_orthologues_from_species(cursor, ensembl_db_name, ortho_type, gene_member_id, species):
# the ortho_type is one of the following: 'ortholog_one2one',
# 'ortholog_one2many', 'ortholog_many2many', 'possible_ortholog', 'apparent_ortholog_one2one'
orthos = []
# find genome db_id
genome_db_id = species2genome_db_id (cursor, species)
# make the cursor point to compara database - should be the responsibility of each function
switch_to_db (cursor, get_compara_name (cursor))
qry = "select homology.homology_id from homology_member, homology "
qry += " where homology_member.gene_member_id =%d " % gene_member_id
qry += " and homology.homology_id = homology_member.homology_id "
qry += " and homology.description = '%s' " % ortho_type
rows = search_db (cursor, qry)
if (not rows):
return [] # no orthologs here
# for each homology id find the other member id
#print qry
#print member_id, ortho_type, species, genome_db_id
#print rows
for row in rows:
homology_id = row[0]
#print "\t homology id:", homology_id
switch_to_db (cursor, get_compara_name (cursor))
qry = "select gene_member_id from homology_member "
qry += " where homology_id = %d" % int(homology_id)
qry += " and not gene_member_id = %d" % gene_member_id
rows2 = search_db (cursor, qry, verbose = False)
if (not rows2):
#print "\t ",
#rows2 = search_db (cursor, qry, verbose = True)
continue
for row2 in rows2:
ortho_id = row2[0]
#print "\t\t ortho id:", ortho_id
qry = "select stable_id from gene_member "
qry += " where gene_member_id = %d " % ortho_id
qry += " and genome_db_id = %d " % genome_db_id
rows3 = search_db (cursor, qry, verbose = False)
if (not rows3):
#print "\t\t ",
#rows3 = search_db (cursor, qry, verbose = True)
continue
ortho_stable = rows3[0][0]
#print "\t\t ortho stable:", ortho_stable
orthos.append(ortho_stable)
if orthos:
switch_to_db (cursor, ensembl_db_name [species])
orthos = map (lambda gene_id: stable2gene(cursor, gene_id), orthos)
#print 'orthos:', orthos
return orthos
def get_orthologues_from_species(cursor, ensembl_db_name, ortho_type, gene_member_id, species):
# the ortho_type is one of the following: 'ortholog_one2one',
# 'ortholog_one2many', 'ortholog_many2many', 'possible_ortholog', 'apparent_ortholog_one2one'
orthos = []
# find genome db_id
genome_db_id = species2genome_db_id(cursor, species)
# make the cursor point to compara database - should be the responsibility of each function
switch_to_db(cursor, get_compara_name(cursor))
qry = "select homology.homology_id from homology_member, homology "
qry += " where homology_member.gene_member_id =%d " % gene_member_id
qry += " and homology.homology_id = homology_member.homology_id "
qry += " and homology.description = '%s' " % ortho_type
rows = search_db(cursor, qry)
if not rows:
return [] # no orthologs here
# for each homology id find the other member id
# print qry
# print member_id, ortho_type, species, genome_db_id
# print rows
for row in rows:
homology_id = row[0]
# print "\t homology id:", homology_id
switch_to_db(cursor, get_compara_name(cursor))
qry = "select gene_member_id from homology_member "
qry += " where homology_id = %d" % int(homology_id)
qry += " and not gene_member_id = %d" % gene_member_id
rows2 = search_db(cursor, qry, verbose=False)
if not rows2:
# print "\t ",
# rows2 = search_db (cursor, qry, verbose = True)
continue
for row2 in rows2:
ortho_id = row2[0]
# print "\t\t ortho id:", ortho_id
qry = "select stable_id from gene_member "
qry += " where gene_member_id = %d " % ortho_id
qry += " and genome_db_id = %d " % genome_db_id
rows3 = search_db(cursor, qry, verbose=False)
if not rows3:
# print "\t\t ",
# rows3 = search_db (cursor, qry, verbose = True)
continue
ortho_stable = rows3[0][0]
# print "\t\t ortho stable:", ortho_stable
orthos.append(ortho_stable)
if orthos:
switch_to_db(cursor, ensembl_db_name[species])
orthos = map(lambda gene_id: stable2gene(cursor, gene_id), orthos)
# print 'orthos:', orthos
return orthos
def get_logic_name(cursor, analysis_id, db_name = None):
if analysis_id < 0:
return ''
if (db_name):
if not switch_to_db(cursor, db_name):
return False
qry = "SELECT logic_name FROM analysis WHERE analysis_id = %d" % analysis_id
rows = search_db (cursor, qry)
if (not rows):
logic_name = ''
else:
logic_name = rows[0][0]
return logic_name
def gene2canon_transl(cursor, gene_id, db_name=None,):
if (db_name and not switch_to_db(cursor, db_name)):
return False
qry = "select translation.translation_id from translation, gene "
qry += " where gene.canonical_transcript_id = translation.transcript_id "
qry += " and gene.gene_id = %d " % gene_id
rows = search_db (cursor, qry, verbose = False)
if (not rows):
rows = search_db (cursor, qry, verbose = True)
return ""
return rows[0][0]
def get_logic_name(cursor, analysis_id, db_name=None):
if analysis_id < 0:
return ""
if db_name:
if not switch_to_db(cursor, db_name):
return False
qry = "SELECT logic_name FROM analysis WHERE analysis_id = %d" % analysis_id
rows = search_db(cursor, qry)
if not rows:
logic_name = ""
else:
logic_name = rows[0][0]
return logic_name
def gene2stable_canon_transl(cursor, gene_id, db_name=None):
if db_name and not switch_to_db(cursor, db_name):
return False
qry = "select translation.stable_id from translation, gene "
qry += " where gene.canonical_transcript_id = translation.transcript_id "
qry += " and gene.gene_id = %d " % gene_id
rows = search_db(cursor, qry, verbose=False)
if not rows:
rows = search_db(cursor, qry, verbose=True)
return ""
return rows[0][0]
def stable2gene (cursor, stable_id=None, db_name=None):
if (not stable_id):
return 0
if (db_name and not switch_to_db(cursor, db_name)):
return False
qry = "select gene_id from gene where stable_id='%s'" % stable_id
rows = search_db (cursor, qry, verbose = False)
if (not rows):
rows = search_db (cursor, qry, verbose = True)
return 0
return int(rows[0][0])