def test_keep_chars(self):
"""keep_chars returns a string containing only chars in keep"""
f = keep_chars('ab c3*[')
self.assertEqual(f(''), '') #empty
self.assertRaises(AttributeError, f, None) #None
#one character, case sensitive
self.assertEqual(f('b'), 'b')
self.assertEqual(f('g'), '')
self.assertEqual(f('xyz123'), '3')
self.assertEqual(f('xyz 123'), ' 3')
#more characters, case sensitive
self.assertEqual(f('kjbwherzcagebcujrkcs'), 'bcabcc')
self.assertEqual(f('f[ffff*ff*fff3fff'), '[**3')
# case insensitive
f = keep_chars('AbC', False)
self.assertEqual(f('abcdef'), 'abc')
self.assertEqual(f('ABCDEF'), 'ABC')
self.assertEqual(f('aBcDeF'), 'aBc')
def test_keep_chars(self):
"""keep_chars returns a string containing only chars in keep"""
f = keep_chars('ab c3*[')
self.assertEqual(f(''),'') #empty
self.assertRaises(AttributeError,f,None) #None
#one character, case sensitive
self.assertEqual(f('b'),'b')
self.assertEqual(f('g'),'')
self.assertEqual(f('xyz123'),'3')
self.assertEqual(f('xyz 123'),' 3')
#more characters, case sensitive
self.assertEqual(f('kjbwherzcagebcujrkcs'),'bcabcc')
self.assertEqual(f('f[ffff*ff*fff3fff'),'[**3')
# case insensitive
f = keep_chars('AbC',False)
self.assertEqual(f('abcdef'),'abc')
self.assertEqual(f('ABCDEF'),'ABC')
self.assertEqual(f('aBcDeF'),'aBc')
def test_keep_chars(self):
"""keep_chars returns a string containing only chars in keep"""
f = keep_chars("ab c3*[")
self.assertEqual(f(""), "") # empty
self.assertRaises(AttributeError, f, None) # None
# one character, case sensitive
self.assertEqual(f("b"), "b")
self.assertEqual(f("g"), "")
self.assertEqual(f("xyz123"), "3")
self.assertEqual(f("xyz 123"), " 3")
# more characters, case sensitive
self.assertEqual(f("kjbwherzcagebcujrkcs"), "bcabcc")
self.assertEqual(f("f[ffff*ff*fff3fff"), "[**3")
# case insensitive
f = keep_chars("AbC", False)
self.assertEqual(f("abcdef"), "abc")
self.assertEqual(f("ABCDEF"), "ABC")
self.assertEqual(f("aBcDeF"), "aBc")
def __init__(self, chars, name=None, invert_charset=False, strip_f=strip, default_char=None):
"""Returns new CharFilter object."""
self.Chars = chars
self.Name = name
self.Invert = invert_charset
if invert_charset:
if default_char:
trans_table = trans_all(chars, default_char)
self.Filter = lambda s: s.translate(trans_table)
else:
self.Filter = exclude_chars(chars)
else:
if default_char:
trans_table = trans_except(chars, default_char)
self.Filter = lambda s: s.translate(trans_table)
else:
self.Filter = keep_chars(chars)
self.stripF = strip_f
def __init__(self,
chars,
name=None,
invert_charset=False,
strip_f=strip,
default_char=None):
"""Returns new CharFilter object."""
self.Chars = chars
self.Name = name
self.Invert = invert_charset
if invert_charset:
if default_char:
trans_table = trans_all(chars, default_char)
self.Filter = lambda s: s.translate(trans_table)
else:
self.Filter = exclude_chars(chars)
else:
if default_char:
trans_table = trans_except(chars, default_char)
self.Filter = lambda s: s.translate(trans_table)
else:
self.Filter = keep_chars(chars)
self.stripF = strip_f
def __init__(self, motifset, Gap=IUPAC_gap, Missing=IUPAC_missing,\
Gaps=None,
Sequence=None, Ambiguities=None,
label=None, Complements=None, Pairs=None, MWCalculator=None, \
add_lower=False, preserve_existing_moltypes=False, \
make_alphabet_group=False, ModelSeq=None):
"""Returns a new MolType object. Note that the parameters are in flux.
Currently:
motifset: Alphabet or sequence of items in the default
alphabet. Does not include degenerates.
Gap: default gap symbol
Missing: symbol for missing data
Gaps: any other symbols that should be treated as gaps (doesn't have
to include Gap or Missing; they will be silently added)
Sequence: Class for constructing sequences.
Ambiguities: dict of char:tuple, doesn't include gaps (these are
hard-coded as - and ?, and added later.
label: text label, don't know what this is used for. Unnecessary?
Complements: dict of symbol:symbol showing how the non-degenerate
single characters complement each other. Used for constructing
on the fly the complement table, incl. support for mustPair and
canPair.
Pairs: dict in which keys are pairs of symbols that can pair
with each other, values are True (must pair) or False (might
pair). Currently, the meaning of GU pairs as 'weak' is conflated
with the meaning of degenerate symbol pairs (which might pair
with each other but don't necessarily, depending on how the
symbol is resolved). This should be refactored.
MWCalculator: f(seq) -> molecular weight.
add_lower: if True (default: False) adds the lowercase versions of
everything into the alphabet. Slated for deletion.
preserve_existing_moltypes: if True (default: False), does not
set the MolType of the things added in **kwargs to self.
make_alphabet_group: if True, makes an AlphabetGroup relating
the various alphabets to one another.
ModelSeq: sequence type for modeling
Note on "Degenerates" versus "Ambiguities": self.Degenerates contains
_only_ mappings for degenerate symbols, whereas self.Ambiguities
contains mappings for both degenerate and non-degenerate symbols.
Sometimes you want one, sometimes the other, so both are provided.
"""
self.Gap = Gap
self.Missing = Missing
self.Gaps = frozenset([Gap, Missing])
if Gaps:
self.Gaps = self.Gaps.union(frozenset(Gaps))
self.label = label
#set the sequence constructor
if Sequence is None:
Sequence = ''.join #safe default string constructor
elif not preserve_existing_moltypes:
Sequence.MolType = self
self.Sequence = Sequence
#set the ambiguities
ambigs = {self.Missing:tuple(motifset)+(self.Gap,),self.Gap:(self.Gap,)}
if Ambiguities:
ambigs.update(Ambiguities)
for c in motifset:
ambigs[c] = (c,)
self.Ambiguities = ambigs
#set Complements -- must set before we make the alphabet group
self.Complements = Complements or {}
if make_alphabet_group: #note: must use _original_ ambiguities here
self.Alphabets = AlphabetGroup(motifset, Ambiguities, \
MolType=self)
self.Alphabet = self.Alphabets.Base
else:
if isinstance(motifset, Enumeration):
self.Alphabet = motifset
elif max(len(motif) for motif in motifset) == 1:
self.Alphabet = CharAlphabet(motifset, MolType=self)
else:
self.Alphabet = Alphabet(motifset, MolType=self)
#set the other properties
self.Degenerates = Ambiguities and Ambiguities.copy() or {}
self.Degenerates[self.Missing] = ''.join(motifset)+self.Gap
self.Matches = make_matches(motifset, self.Gaps, self.Degenerates)
self.Pairs = Pairs and Pairs.copy() or {}
self.Pairs.update(make_pairs(Pairs, motifset, self.Gaps, \
self.Degenerates))
self.MWCalculator = MWCalculator
#add lowercase characters, if we're doing that
if add_lower:
self._add_lowercase()
#cache various other data that make the calculations faster
self._make_all()
self._make_comp_table()
# a gap can be a true gap char or a degenerate character, typically '?'
# we therefore want to ensure consistent treatment across the definition
# of characters as either gap or degenerate
self.GapString = ''.join(self.Gaps)
strict_gap = "".join(set(self.GapString) - set(self.Degenerates))
self.stripDegenerate = FunctionWrapper(
keep_chars(strict_gap+''.join(self.Alphabet)))
self.stripBad = FunctionWrapper(keep_chars(''.join(self.All)))
to_keep = set(self.Alphabet) ^ set(self.Degenerates) - set(self.Gaps)
self.stripBadAndGaps = FunctionWrapper(keep_chars(''.join(to_keep)))
#make inverse degenerates from degenerates
#ensure that lowercase versions also exist if appropriate
inv_degens = {}
for key, val in self.Degenerates.items():
inv_degens[frozenset(val)] = key.upper()
if add_lower:
inv_degens[frozenset(''.join(val).lower())] = key.lower()
for m in self.Alphabet:
inv_degens[frozenset(m)] = m
if add_lower:
inv_degens[frozenset(''.join(m).lower())] = m.lower()
for m in self.Gaps:
inv_degens[frozenset(m)] = m
self.InverseDegenerates = inv_degens
#set array type for modeling alphabets
try:
self.ArrayType = self.Alphabet.ArrayType
except AttributeError:
self.ArrayType = None
#set modeling sequence
self.ModelSeq = ModelSeq
def __init__(self, motifset, Gap=IUPAC_gap, Missing=IUPAC_missing,\
Gaps=None,
Sequence=None, Ambiguities=None,
label=None, Complements=None, Pairs=None, MWCalculator=None, \
add_lower=False, preserve_existing_moltypes=False, \
make_alphabet_group=False, ModelSeq=None):
"""Returns a new MolType object. Note that the parameters are in flux.
Currently:
motifset: Alphabet or sequence of items in the default
alphabet. Does not include degenerates.
Gap: default gap symbol
Missing: symbol for missing data
Gaps: any other symbols that should be treated as gaps (doesn't have
to include Gap or Missing; they will be silently added)
Sequence: Class for constructing sequences.
Ambiguities: dict of char:tuple, doesn't include gaps (these are
hard-coded as - and ?, and added later.
label: text label, don't know what this is used for. Unnecessary?
Complements: dict of symbol:symbol showing how the non-degenerate
single characters complement each other. Used for constructing
on the fly the complement table, incl. support for mustPair and
canPair.
Pairs: dict in which keys are pairs of symbols that can pair
with each other, values are True (must pair) or False (might
pair). Currently, the meaning of GU pairs as 'weak' is conflated
with the meaning of degenerate symbol pairs (which might pair
with each other but don't necessarily, depending on how the
symbol is resolved). This should be refactored.
MWCalculator: f(seq) -> molecular weight.
add_lower: if True (default: False) adds the lowercase versions of
everything into the alphabet. Slated for deletion.
preserve_existing_moltypes: if True (default: False), does not
set the MolType of the things added in **kwargs to self.
make_alphabet_group: if True, makes an AlphabetGroup relating
the various alphabets to one another.
ModelSeq: sequence type for modeling
Note on "Degenerates" versus "Ambiguities": self.Degenerates contains
_only_ mappings for degenerate symbols, whereas self.Ambiguities
contains mappings for both degenerate and non-degenerate symbols.
Sometimes you want one, sometimes the other, so both are provided.
"""
self.Gap = Gap
self.Missing = Missing
self.Gaps = frozenset([Gap, Missing])
if Gaps:
self.Gaps = self.Gaps.union(frozenset(Gaps))
self.label = label
#set the sequence constructor
if Sequence is None:
Sequence = ''.join #safe default string constructor
elif not preserve_existing_moltypes:
Sequence.MolType = self
self.Sequence = Sequence
#set the ambiguities
ambigs = {self.Missing:tuple(motifset)+(self.Gap,),self.Gap:(self.Gap,)}
if Ambiguities:
ambigs.update(Ambiguities)
for c in motifset:
ambigs[c] = (c,)
self.Ambiguities = ambigs
#set Complements -- must set before we make the alphabet group
self.Complements = Complements or {}
if make_alphabet_group: #note: must use _original_ ambiguities here
self.Alphabets = AlphabetGroup(motifset, Ambiguities, \
MolType=self)
self.Alphabet = self.Alphabets.Base
else:
if isinstance(motifset, Enumeration):
self.Alphabet = motifset
elif max(len(motif) for motif in motifset) == 1:
self.Alphabet = CharAlphabet(motifset, MolType=self)
else:
self.Alphabet = Alphabet(motifset, MolType=self)
#set the other properties
self.Degenerates = Ambiguities and Ambiguities.copy() or {}
self.Degenerates[self.Missing] = ''.join(motifset)+self.Gap
self.Matches = make_matches(motifset, self.Gaps, self.Degenerates)
self.Pairs = Pairs and Pairs.copy() or {}
self.Pairs.update(make_pairs(Pairs, motifset, self.Gaps, \
self.Degenerates))
self.MWCalculator = MWCalculator
#add lowercase characters, if we're doing that
if add_lower:
self._add_lowercase()
#cache various other data that make the calculations faster
self._make_all()
self._make_comp_table()
# a gap can be a true gap char or a degenerate character, typically '?'
# we therefore want to ensure consistent treatment across the definition
# of characters as either gap or degenerate
self.GapString = ''.join(self.Gaps)
strict_gap = "".join(set(self.GapString) - set(self.Degenerates))
self.stripDegenerate = FunctionWrapper(
keep_chars(strict_gap+''.join(self.Alphabet)))
self.stripBad = FunctionWrapper(keep_chars(''.join(self.All)))
to_keep = set(self.Alphabet) ^ set(self.Degenerates) - set(self.Gaps)
self.stripBadAndGaps = FunctionWrapper(keep_chars(''.join(to_keep)))
#make inverse degenerates from degenerates
#ensure that lowercase versions also exist if appropriate
inv_degens = {}
for key, val in self.Degenerates.items():
inv_degens[frozenset(val)] = key.upper()
if add_lower:
inv_degens[frozenset(''.join(val).lower())] = key.lower()
for m in self.Alphabet:
inv_degens[frozenset(m)] = m
if add_lower:
inv_degens[frozenset(''.join(m).lower())] = m.lower()
for m in self.Gaps:
inv_degens[frozenset(m)] = m
self.InverseDegenerates = inv_degens
#set array type for modeling alphabets
try:
self.ArrayType = self.Alphabet.ArrayType
except AttributeError:
self.ArrayType = None
#set modeling sequence
self.ModelSeq = ModelSeq
