def join_tables(* args, **kwargs):
"""Join together tables into one table.
Each argument is a tuple (table_i, key_i, cols_i)
key_i is either a column name or a function that maps a
table row to a unique key
"""
if len(args) == 0:
return Table()
# determine common keys
tab, key, cols = args[0]
if isinstance(key, str):
keys = tab.cget(key)
lookups = [tab.lookup(key)]
else:
keys = map(key, tab)
lookup = {}
for row in tab:
lookup[key(row)] = row
lookups = [lookup]
keyset = set(keys)
for tab, key, cols in args[1:]:
if isinstance(key, str):
keyset = keyset & set(tab.cget(key))
lookups.append(tab.lookup(key))
else:
keyset = keyset & set(map(key, tab))
lookup = {}
for row in tab:
lookup[key(row)] = row
lookups.append(lookup)
keys = filter(lambda x: x in keyset, keys)
# build new table
if "headers" not in kwargs:
headers = util.concat(*util.cget(args, 2))
else:
headers = kwargs["headers"]
tab = Table(headers=headers)
for key in keys:
row = {}
for (tab2, key2, cols), lookup in zip(args, lookups):
row.update(util.subdict(lookup[key], cols))
tab.append(row)
return tab
def join_tables(* args, **kwargs):
"""Join together tables into one table.
Each argument is a tuple (table_i, key_i, cols_i)
key_i is either a column name or a function that maps a
table row to a unique key
"""
if len(args) == 0:
return Table()
# determine common keys
tab, key, cols = args[0]
if isinstance(key, str):
keys = tab.cget(key)
lookups = [tab.lookup(key)]
else:
keys = map(key, tab)
lookup = {}
for row in tab:
lookup[key(row)] = row
lookups = [lookup]
keyset = set(keys)
for tab, key, cols in args[1:]:
if isinstance(key, str):
keyset = keyset & set(tab.cget(key))
lookups.append(tab.lookup(key))
else:
keyset = keyset & set(map(key, tab))
lookup = {}
for row in tab:
lookup[key(row)] = row
lookups.append(lookup)
keys = filter(lambda x: x in keyset, keys)
# build new table
if "headers" not in kwargs:
headers = util.concat(*util.cget(args, 2))
else:
headers = kwargs["headers"]
tab = Table(headers=headers)
for key in keys:
row = {}
for (tab2, key2, cols), lookup in zip(args, lookups):
row.update(util.subdict(lookup[key], cols))
tab.append(row)
return tab
def align(self, * names):
if len(names) == 0:
print "nothing to align"
# get names from genes if they are not strings
if type(names[0]) != str:
names = [i.name for i in names]
seqs2 = util.subdict(self.seqs, names)
aln = muscle.muscle(seqs2)
muscle.printAlign(aln)
def align(self, *names):
if len(names) == 0:
print "nothing to align"
# get names from genes if they are not strings
if type(names[0]) != str:
names = [i.name for i in names]
seqs2 = util.subdict(self.seqs, names)
aln = muscle.muscle(seqs2)
muscle.printAlign(aln)
def new(self, headers=None):
"""
return a new table with the same info but no data
headers - if specified, only a subset of the headers will be copied
"""
if headers == None:
headers = self.headers
tab = type(self)(headers=headers)
tab.types = util.subdict(self.types, headers)
tab.defaults = util.subdict(self.defaults, headers)
tab.comments = copy.copy(self.comments)
tab.delim = self.delim
tab.nheaders = self.nheaders
tab._type_lookup = copy.copy(self._type_lookup)
return tab
def new(self, headers=None):
"""
return a new table with the same info but no data
headers - if specified, only a subset of the headers will be copied
"""
if headers == None:
headers = self.headers
tab = type(self)(headers=headers)
tab.types = util.subdict(self.types, headers)
tab.defaults = util.subdict(self.defaults, headers)
tab.comments = copy.copy(self.comments)
tab.delim = self.delim
tab.nheaders = self.nheaders
tab._type_lookup = copy.copy(self._type_lookup)
return tab
def new(self, headers=None):
"""
Return a new table with the same info but no data.
headers: if specified, only a subset of the headers will be copied.
"""
if headers is None:
headers = self.headers
tab = type(self)(headers=headers)
tab.types = util.subdict(self.types, headers)
tab.comments = copy.copy(self.comments)
tab.delim = self.delim
tab.nheaders = self.nheaders
return tab