def hit_filter(self, func=None):
"""Creates a new QueryResult object whose Hit objects pass the filter
function.
Arguments:
func -- Callback function that accepts a Hit object as its parameter,
does a boolean check, and returns True or False
Here is an example of using `hit_filter` to select Hits whose
description begins with the string 'H**o sapiens', case sensitive:
>>> from SAP.Bio import SearchIO
>>> qresult = next(SearchIO.parse('Blast/mirna.xml', 'blast-xml'))
>>> def desc_filter(hit):
... return hit.description.startswith('H**o sapiens')
...
>>> len(qresult)
100
>>> filtered = qresult.hit_filter(desc_filter)
>>> len(filtered)
39
>>> print(filtered[:4])
Program: blastn (2.2.27+)
Query: 33211 (61)
mir_1
Target: refseq_rna
Hits: ---- ----- ----------------------------------------------------------
# # HSP ID + description
---- ----- ----------------------------------------------------------
0 1 gi|262205317|ref|NR_030195.1| H**o sapiens microRNA 52...
1 2 gi|262205330|ref|NR_030198.1| H**o sapiens microRNA 52...
2 1 gi|262205302|ref|NR_030191.1| H**o sapiens microRNA 51...
3 1 gi|262205451|ref|NR_030222.1| H**o sapiens microRNA 51...
Note that instance attributes (other than the hits) from the unfiltered
QueryResult are retained in the filtered object.
>>> qresult.program == filtered.program
True
>>> qresult.target == filtered.target
True
"""
hits = list(filter(func, self.hits))
obj = self.__class__(hits, self.id, self._hit_key_function)
self._transfer_attrs(obj)
return obj
def _filter_search(self, filter_func, order, follow_attrs):
"""Perform a BFS or DFS traversal through all elements in this tree.
:returns: generator of all elements for which `filter_func` is True.
"""
order_opts = {'preorder': _preorder_traverse,
'postorder': _postorder_traverse,
'level': _level_traverse}
try:
order_func = order_opts[order]
except KeyError:
raise ValueError("Invalid order '%s'; must be one of: %s"
% (order, tuple(order_opts)))
if follow_attrs:
get_children = _sorted_attrs
root = self
else:
get_children = lambda elem: elem.clades
root = self.root
return filter(filter_func, order_func(root, get_children))
def _filter_search(self, filter_func, order, follow_attrs):
"""Perform a BFS or DFS traversal through all elements in this tree.
:returns: generator of all elements for which `filter_func` is True.
"""
order_opts = {
'preorder': _preorder_traverse,
'postorder': _postorder_traverse,
'level': _level_traverse
}
try:
order_func = order_opts[order]
except KeyError:
raise ValueError("Invalid order '%s'; must be one of: %s" %
(order, tuple(order_opts)))
if follow_attrs:
get_children = _sorted_attrs
root = self
else:
get_children = lambda elem: elem.clades
root = self.root
return filter(filter_func, order_func(root, get_children))
def filter(self, func=None):
"""Creates a new Hit object whose HSP objects pass the filter
function.
Arguments:
func -- Callback function that accepts a HSP object as its parameter,
does a boolean check, and returns True or False.
`filter` is analogous to Python's built-in `filter` function, except
that instead of returning a list it returns a `Hit` object. Here is an
example of using `filter` to select for HSPs having bitscores bigger
than 60:
>>> from SAP.Bio import SearchIO
>>> qresult = next(SearchIO.parse('Blast/mirna.xml', 'blast-xml'))
>>> hit = qresult[3]
>>> evalue_filter = lambda hsp: hsp.bitscore > 60
>>> filtered_hit = hit.filter(evalue_filter)
>>> len(hit)
2
>>> len(filtered_hit)
1
>>> print(filtered_hit)
Query: 33211
mir_1
Hit: gi|301171322|ref|NR_035857.1| (86)
Pan troglodytes microRNA mir-520c (MIR520C), microRNA
HSPs: ---- -------- --------- ------ --------------- ---------------------
# E-value Bit score Span Query range Hit range
---- -------- --------- ------ --------------- ---------------------
0 8.9e-20 100.47 60 [1:61] [13:73]
"""
hsps = list(filter(func, self.hsps))
if hsps:
obj = self.__class__(hsps)
self._transfer_attrs(obj)
return obj
def filter(self, func=None):
"""Creates a new Hit object whose HSP objects pass the filter
function.
Arguments:
func -- Callback function that accepts a HSP object as its parameter,
does a boolean check, and returns True or False.
`filter` is analogous to Python's built-in `filter` function, except
that instead of returning a list it returns a `Hit` object. Here is an
example of using `filter` to select for HSPs having bitscores bigger
than 60:
>>> from SAP.Bio import SearchIO
>>> qresult = next(SearchIO.parse('Blast/mirna.xml', 'blast-xml'))
>>> hit = qresult[3]
>>> evalue_filter = lambda hsp: hsp.bitscore > 60
>>> filtered_hit = hit.filter(evalue_filter)
>>> len(hit)
2
>>> len(filtered_hit)
1
>>> print(filtered_hit)
Query: 33211
mir_1
Hit: gi|301171322|ref|NR_035857.1| (86)
Pan troglodytes microRNA mir-520c (MIR520C), microRNA
HSPs: ---- -------- --------- ------ --------------- ---------------------
# E-value Bit score Span Query range Hit range
---- -------- --------- ------ --------------- ---------------------
0 8.9e-20 100.47 60 [1:61] [13:73]
"""
hsps = list(filter(func, self.hsps))
if hsps:
obj = self.__class__(hsps)
self._transfer_attrs(obj)
return obj
