def test_KeepChars(self):
"""KeepChars returns a string containing only chars in keep"""
f = KeepChars("ab c3*[")
self.assertEqual(f(""), "") # empty
self.assertRaises(TypeError, 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 = KeepChars("AbC", False)
self.assertEqual(f("abcdef"), "abc")
self.assertEqual(f("ABCDEF"), "ABC")
self.assertEqual(f("aBcDeF"), "aBc")
def __init__(
self,
motifset,
gap=IUPAC_gap,
missing=IUPAC_missing,
gaps=None,
seq_constructor=None,
ambiguities=None,
label=None,
complements=None,
pairs=None,
mw_calculator=None,
add_lower=False,
preserve_existing_moltypes=False,
make_alphabet_group=False,
array_seq_constructor=None,
colors=None,
):
"""Returns a new MolType object. Note that the parameters are in flux.
Parameters
----------
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)
seq_constructor
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 must_pair and
can_pair.
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.
mw_calculator
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.
array_seq_constructor
sequence type for array sequence
colors
dict mapping moltype characters to colors for display
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._serialisable = {k: v for k, v in locals().items() if k != "self"}
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 seq_constructor is None:
seq_constructor = "".join # safe default string constructor
elif not preserve_existing_moltypes:
seq_constructor.moltype = self
self._make_seq = seq_constructor
# 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.mw_calculator = mw_calculator
# 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.gap_string = "".join(self.gaps)
strict_gap = "".join(set(self.gap_string) - set(self.degenerates))
self.strip_degenerate = FunctionWrapper(
KeepChars(strict_gap + "".join(self.alphabet)))
self.strip_bad = FunctionWrapper(KeepChars("".join(self.All)))
to_keep = set(self.alphabet) ^ set(self.degenerates) - set(self.gaps)
self.strip_bad_and_gaps = FunctionWrapper(KeepChars("".join(to_keep)))
# make inverse degenerates from degenerates
# ensure that lowercase versions also exist if appropriate
inv_degens = {}
for key, val in list(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.inverse_degenerates = inv_degens
# set array type for modeling alphabets
try:
self.array_type = self.alphabet.array_type
except AttributeError:
self.array_type = None
# set modeling sequence
self._make_array_seq = array_seq_constructor
self._colors = colors or defaultdict(_DefaultValue("black"))