def test_require_parents_equal_except_location(self):
with pytest.raises(MismatchedParentException):
ObjectValidation.require_parents_equal_except_location(
Parent(id="parent1"), Parent(id="parent2"))
ObjectValidation.require_parents_equal_except_location(
Parent(id="parent", location=SingleInterval(0, 5, Strand.PLUS)),
Parent(id="parent", location=SingleInterval(10, 20, Strand.MINUS)),
)
def test_require_location_nonempty(self):
with pytest.raises(LocationException):
ObjectValidation.require_location_nonempty(
SingleInterval(5, 5, Strand.PLUS))
with pytest.raises(LocationException):
ObjectValidation.require_location_nonempty(
CompoundInterval([5, 10], [5, 10], Strand.PLUS))
ObjectValidation.require_location_nonempty(
SingleInterval(5, 6, Strand.PLUS))
def test_require_object_has_type(self):
with pytest.raises(TypeError):
ObjectValidation.require_object_has_type(
CompoundInterval.from_single_intervals(
[SingleInterval(5, 10, Strand.PLUS)]),
SingleInterval,
)
ObjectValidation.require_object_has_type(
CompoundInterval.from_single_intervals(
[SingleInterval(5, 10, Strand.PLUS)]),
CompoundInterval,
)
def test_init(self):
sequence = Sequence(
"ACTG",
Alphabet.NT_STRICT,
id="id",
type="seqtype_1",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
location=SingleInterval(5, 9, Strand.MINUS, parent="parent"),
),
)
# Sequence data
assert sequence.sequence == Seq("ACTG")
assert str(sequence) == "ACTG"
# Alphabet
assert sequence.alphabet == Alphabet.NT_STRICT
# ID
assert sequence.id == "id"
# Sequence type
assert sequence.sequence_type == "seqtype_1"
# Parent ID
assert sequence.parent_id == "parent"
assert Sequence("A", Alphabet.NT_STRICT,
parent="parent").parent_id == "parent"
assert (Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(
location=SingleInterval(5, 6, Strand.MINUS, parent="parent")),
).parent_id == "parent")
assert Sequence("A", Alphabet.NT_STRICT).parent_id is None
# Parent type
assert sequence.parent_type == "seqtype_2"
# Parent strand
assert sequence.parent_strand == Strand.MINUS
assert Sequence(
"A", Alphabet.NT_STRICT,
parent=Strand.UNSTRANDED).parent_strand == Strand.UNSTRANDED
# Location on parent
assert (Sequence(
"A",
Alphabet.NT_STRICT,
parent=SingleInterval(3, 4, Strand.UNSTRANDED),
).parent_strand == Strand.UNSTRANDED)
assert Sequence("A", Alphabet.NT_STRICT).parent_strand is None
assert sequence.location_on_parent == SingleInterval(5,
9,
Strand.MINUS,
parent="parent")
# No alphabet validation
Sequence("xxx", Alphabet.NT_STRICT, validate_alphabet=False)
def test_require_location_has_parent_with_sequence(self):
with pytest.raises(NullParentException):
ObjectValidation.require_location_has_parent_with_sequence(
SingleInterval(0, 5, Strand.PLUS))
with pytest.raises(NullSequenceException):
ObjectValidation.require_location_has_parent_with_sequence(
SingleInterval(0, 5, Strand.PLUS, parent="parent"))
ObjectValidation.require_location_has_parent_with_sequence(
SingleInterval(
0,
5,
Strand.PLUS,
parent=Parent(id="parent",
sequence=Sequence("AAAAA", Alphabet.NT_STRICT)),
))
def test_intersection_error(self):
with pytest.raises(MismatchedParentException):
EmptyLocation().intersection(SingleInterval(0,
1,
Strand.PLUS,
parent="seq"),
strict_parent_compare=True)
def chunk_relative_interval_to_cds(self, chr_start: int, chr_end: int, chr_strand: Strand) -> Location:
"""Converts a contiguous interval on the chunk-relative sequence to a relative location within the CDS."""
if not self.is_coding:
raise NoncodingTranscriptError("No CDS positions on non-coding transcript")
return self.cds.chunk_relative_location.parent_to_relative_location(
SingleInterval(chr_start, chr_end, chr_strand, parent=self.cds.chunk_relative_location.parent)
)
def test_intersection_exception(self):
schema = FeatureIntervalModel.Schema().load(
dict(interval_starts=[10], interval_ends=[15], strand=Strand.MINUS.name)
)
feat = schema.to_feature_interval()
s = SingleInterval(0, 5, Strand.PLUS)
with pytest.raises(EmptyLocationException):
_ = feat.intersect(s)
def chromosome_location(self) -> Location:
"""Returns the Location of this in *chromosome coordinates*.
If the coordinate system is unknown, this will return the same coordinate system as
``chunk_relative_location``, that is the true underlying ``_location`` member.
This Location object will always have the full span of the Interval in chromosome coordinates,
even if this feature exists in chunk relative coordinates. As a result of this, if this
Interval was built on chunk relative coordinates, the sequence information will not be present.
"""
if self._parent_or_seq_chunk_parent and self._parent_or_seq_chunk_parent.has_ancestor_of_type(
SequenceType.CHROMOSOME
):
parent = self._parent_or_seq_chunk_parent.first_ancestor_of_type(SequenceType.CHROMOSOME)
return SingleInterval(self.start, self.end, Strand.PLUS, parent)
else:
return SingleInterval(self.start, self.end, Strand.PLUS)
def test_seq_chunk_to_parent(strand):
obs = seq_chunk_to_parent("ATGCATGC", "TestSeq", 200, 208, strand=strand)
assert obs == Parent(
id="TestSeq:200-208",
sequence_type=SequenceType.SEQUENCE_CHUNK,
strand=None,
location=None,
sequence=Sequence(
data="ATGCATGC",
id="TestSeq:200-208",
alphabet=Alphabet.NT_EXTENDED_GAPPED,
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(
id="TestSeq",
sequence_type=SequenceType.CHROMOSOME,
strand=strand,
location=SingleInterval(
200,
208,
strand,
parent=Parent(
id="TestSeq",
sequence_type=SequenceType.CHROMOSOME,
strand=strand,
location=SingleInterval(200, 208, strand),
sequence=None,
parent=None,
),
),
sequence=None,
parent=Parent(
id="TestSeq",
sequence_type=SequenceType.CHROMOSOME,
strand=strand,
location=SingleInterval(200, 208, strand, parent=None),
sequence=None,
parent=None,
),
),
),
)
assert obs.has_ancestor_of_type(SequenceType.CHROMOSOME)
assert obs.has_ancestor_of_type(SequenceType.SEQUENCE_CHUNK)
def test_seq_to_parent():
seq = "ATGCATGC"
seq_id = "TestSeq"
obs = seq_to_parent(seq, seq_id=seq_id)
assert obs == Parent(
sequence=Sequence(seq,
Alphabet.NT_EXTENDED_GAPPED,
type=SequenceType.CHROMOSOME,
id=seq_id),
location=SingleInterval(0, len(seq), Strand.PLUS),
)
def test_require_parent_has_parent_with_location(self):
with pytest.raises(NullParentException):
ObjectValidation.require_parent_has_parent_with_location(
Parent(id="parent"))
with pytest.raises(NullParentException):
ObjectValidation.require_parent_has_parent_with_location(
Parent(id="parent", parent="grandparent"))
ObjectValidation.require_parent_has_parent_with_location(
Parent(
id="parent",
parent=Parent(id="grandparent",
location=SingleInterval(0, 5, Strand.PLUS)),
))
def seq_chunk_to_parent(
seq: str,
sequence_name: Union[UUID, str],
start: int,
end: int,
strand: Optional[Strand] = Strand.PLUS,
alphabet: Optional[Alphabet] = Alphabet.NT_EXTENDED_GAPPED,
) -> Parent:
"""Construct a sequence chunk parent from a sequence. This is used when an annotation collection is being
instantiated with a subset of a genome sequence.
NOTE: This sequence is assumed to be a subset of a chromosome. There is no way to validate that within this
function.
Args:
seq: Sequence subset to use.
sequence_name: The name of the sequence.
start: The genomic start position of this sequence.
end: The genomic end position of this sequence.
strand: The strand this chunk is relative to the genome.
alphabet: The alphabet the sequence is in.
Returns:
An instantiated Parent object ready to be passed to a constructor.
"""
chunk_id = f"{sequence_name}:{start}-{end}"
return Parent(
id=chunk_id,
sequence=Sequence(
seq,
alphabet,
id=chunk_id,
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(location=SingleInterval(
start,
end,
strand,
parent=Parent(id=sequence_name,
sequence_type=SequenceType.CHROMOSOME),
)),
),
)
def test_require_locations_overlap(self):
with pytest.raises(LocationOverlapException):
ObjectValidation.require_locations_overlap(
SingleInterval(0, 5, Strand.PLUS),
SingleInterval(5, 10, Strand.PLUS))
with pytest.raises(LocationOverlapException):
ObjectValidation.require_locations_overlap(
SingleInterval(0, 5, Strand.PLUS),
SingleInterval(0, 5, Strand.MINUS),
match_strand=True,
)
ObjectValidation.require_locations_overlap(
SingleInterval(0, 5, Strand.PLUS),
SingleInterval(3, 6, Strand.PLUS),
match_strand=True,
)
def initialize_location(
starts: List[int],
ends: List[int],
strand: Strand,
parent_or_seq_chunk_parent: Optional[Parent] = None,
) -> Location:
"""
Initialize the :class:`Location` object for this interval.
Args:
starts: Start positions relative to the chromosome.
ends: End positions relative to the chromosome.
strand: Strand relative to the chromosome.
parent_or_seq_chunk_parent: An optional parent, either as a full chromosome or as a sequence chunk.
"""
if len(starts) != len(ends):
raise ValidationException("Number of interval starts does not match number of interval ends")
elif len(starts) == len(ends) == 1:
location = SingleInterval(starts[0], ends[0], strand)
else:
location = CompoundInterval(starts, ends, strand)
return AbstractInterval.liftover_location_to_seq_chunk_parent(location, parent_or_seq_chunk_parent)
def seq_to_parent(
seq: str,
alphabet: Optional[Alphabet] = Alphabet.NT_EXTENDED_GAPPED,
seq_id: Optional[str] = None,
seq_type: Optional[str] = SequenceType.CHROMOSOME,
) -> Parent:
"""Convert a string into a Parent object. This is the intermediate that transfers a BioPython sequence object to
a BioCantor sequence object.
NOTE: This sequence is assumed to be the entire chromosome.
Args:
seq: String of sequence.
alphabet: Alphabet this sequence is in.
seq_id: ID to attach to the Parent.
seq_type: Sequence type to attach to the Parent.
Returns:
A :class:`Parent` object.
"""
return Parent(sequence=Sequence(seq, alphabet, type=seq_type, id=seq_id),
location=SingleInterval(0, len(seq), Strand.PLUS))
def test_getitem_error(self):
with pytest.raises(InvalidStrandException):
Sequence("actgactg",
Alphabet.NT_STRICT,
parent=SingleInterval(0, 8, Strand.UNSTRANDED))[3:6]
class TestEmptyLocation:
def test_is_singleton(self):
assert EmptyLocation() is EmptyLocation()
def test_len(self):
assert len(EmptyLocation()) == 0
def test_length(self):
assert EmptyLocation().length == 0
def test_parent(self):
assert EmptyLocation().parent is None
def test_strand(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().strand
def test_start(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().start
def test_end(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().end
def test_is_contiguous(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().is_contiguous
def test_blocks(self):
assert EmptyLocation().blocks == []
def test_scan_blocks(self):
assert not EmptyLocation().scan_blocks()
def test_num_blocks(self):
assert EmptyLocation().num_blocks == 0
def test_optimize_blocks(self):
assert EmptyLocation().optimize_blocks() == EmptyLocation()
def test_gap_list(self):
assert EmptyLocation().gap_list() == []
def test_gaps_location(self):
assert EmptyLocation().gaps_location() == EmptyLocation()
def test_extract_sequence(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().extract_sequence()
def test_parent_to_relative_pos(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().parent_to_relative_pos(0)
def test_relative_to_parent_pos(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().relative_to_parent_pos(0)
def test_parent_to_relative_location(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().parent_to_relative_location(
SingleInterval(0, 0, Strand.PLUS))
@pytest.mark.parametrize(
"other",
[
EmptyLocation(),
SingleInterval(5, 10, Strand.PLUS),
CompoundInterval([5], [10], Strand.PLUS),
],
)
def test_intersection(self, other):
assert EmptyLocation().intersection(other) == EmptyLocation()
def test_intersection_error(self):
with pytest.raises(MismatchedParentException):
EmptyLocation().intersection(SingleInterval(0,
1,
Strand.PLUS,
parent="seq"),
strict_parent_compare=True)
@pytest.mark.parametrize(
"other",
[
EmptyLocation(),
SingleInterval(5, 10, Strand.PLUS),
CompoundInterval([5], [10], Strand.PLUS),
],
)
def test_minus(self, other):
assert EmptyLocation().minus(other) == EmptyLocation()
def test_minus_error(self):
with pytest.raises(MismatchedParentException):
EmptyLocation().minus(SingleInterval(0,
1,
Strand.PLUS,
parent="seq"),
strict_parent_compare=True)
@pytest.mark.parametrize(
"other",
[
EmptyLocation(),
SingleInterval(5, 10, Strand.PLUS),
CompoundInterval([5], [10], Strand.PLUS),
],
)
def test_location_relative_to(self, other):
assert EmptyLocation().location_relative_to(other) == EmptyLocation()
def test_relative_interval_to_parent_location(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().relative_interval_to_parent_location(
0, 1, Strand.PLUS)
def test_has_overlap(self):
assert EmptyLocation().has_overlap(EmptyLocation()) is False
def test_has_overlap_error(self):
with pytest.raises(MismatchedParentException):
EmptyLocation().has_overlap(SingleInterval(0,
1,
Strand.PLUS,
parent="seq"),
strict_parent_compare=True)
def test_contains(self):
assert not EmptyLocation().contains(EmptyLocation())
assert not EmptyLocation().contains(SingleInterval(0, 1, Strand.PLUS))
def test_contains_error(self):
with pytest.raises(MismatchedParentException):
EmptyLocation().contains(SingleInterval(0,
1,
Strand.PLUS,
parent="seq"),
strict_parent_compare=True)
def test_reverse(self):
assert EmptyLocation().reverse() == EmptyLocation()
def test_reverse_strand(self):
assert EmptyLocation().reverse_strand() == EmptyLocation()
def test_reset_strand(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().reset_strand(Strand.PLUS)
def test_reset_parent(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().reset_parent(None)
def test_extend_absolute(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().extend_absolute(0, 0)
def test_extend_relative(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().extend_relative(0, 0)
def test_shift_position(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().shift_position(0)
def test_distance_to(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().distance_to(EmptyLocation())
def test_merge_overlapping(self):
assert EmptyLocation().merge_overlapping() == EmptyLocation()
def test_first_ancestor_of_type(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().first_ancestor_of_type("seqtype")
def test_has_ancestor_of_type(self):
assert EmptyLocation().has_ancestor_of_type("seqtype") is False
def test_union(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().union(SingleInterval(0, 1, Strand.PLUS))
def test_union_preserve_overlaps(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().union_preserve_overlaps(
SingleInterval(0, 1, Strand.PLUS))
class TestSequence:
def test_init(self):
sequence = Sequence(
"ACTG",
Alphabet.NT_STRICT,
id="id",
type="seqtype_1",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
location=SingleInterval(5, 9, Strand.MINUS, parent="parent"),
),
)
# Sequence data
assert sequence.sequence == Seq("ACTG")
assert str(sequence) == "ACTG"
# Alphabet
assert sequence.alphabet == Alphabet.NT_STRICT
# ID
assert sequence.id == "id"
# Sequence type
assert sequence.sequence_type == "seqtype_1"
# Parent ID
assert sequence.parent_id == "parent"
assert Sequence("A", Alphabet.NT_STRICT,
parent="parent").parent_id == "parent"
assert (Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(
location=SingleInterval(5, 6, Strand.MINUS, parent="parent")),
).parent_id == "parent")
assert Sequence("A", Alphabet.NT_STRICT).parent_id is None
# Parent type
assert sequence.parent_type == "seqtype_2"
# Parent strand
assert sequence.parent_strand == Strand.MINUS
assert Sequence(
"A", Alphabet.NT_STRICT,
parent=Strand.UNSTRANDED).parent_strand == Strand.UNSTRANDED
# Location on parent
assert (Sequence(
"A",
Alphabet.NT_STRICT,
parent=SingleInterval(3, 4, Strand.UNSTRANDED),
).parent_strand == Strand.UNSTRANDED)
assert Sequence("A", Alphabet.NT_STRICT).parent_strand is None
assert sequence.location_on_parent == SingleInterval(5,
9,
Strand.MINUS,
parent="parent")
# No alphabet validation
Sequence("xxx", Alphabet.NT_STRICT, validate_alphabet=False)
@pytest.mark.parametrize(
"data,alphabet,parent_id,parent_type,parent_strand,location_on_parent,expected_exception",
[
("A-C", Alphabet.NT_STRICT, None, None, None, None, AlphabetError),
(
"ACG",
Alphabet.NT_STRICT,
None,
None,
None,
SingleInterval(0, 4, Strand.PLUS),
ParentException,
),
(
"ATT",
Alphabet.NT_STRICT,
"parent1",
None,
None,
SingleInterval(0, 3, Strand.PLUS, parent="parent2"),
ParentException,
),
(
"GGG",
Alphabet.NT_STRICT,
None,
None,
Strand.MINUS,
SingleInterval(0, 3, Strand.PLUS),
InvalidStrandException,
),
(
"GGG",
Alphabet.NT_STRICT,
None,
"seqtype_2",
None,
SingleInterval(
0,
3,
Strand.PLUS,
parent=Parent(sequence_type="seqtype_3"),
),
ParentException,
),
],
)
def test_init_invalid_params(
self,
data,
alphabet,
parent_id,
parent_type,
parent_strand,
location_on_parent,
expected_exception,
):
with pytest.raises(expected_exception):
Sequence(
data,
alphabet,
parent=Parent(
id=parent_id,
sequence_type=parent_type,
strand=parent_strand,
location=location_on_parent,
),
)
@pytest.mark.parametrize(
"sequence,other,expected",
[
(
Sequence("AAAA", Alphabet.NT_STRICT, validate_alphabet=False),
"AAAA",
False,
),
(
Sequence("AAAA", Alphabet.NT_STRICT, validate_alphabet=False),
Sequence("AAAA", Alphabet.NT_STRICT, validate_alphabet=True),
True,
),
(
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
True,
),
(
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
Sequence(
"AAAa",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
False,
),
(
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
Sequence(
"AAAA",
Alphabet.NT_EXTENDED,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
False,
),
(
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq2",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
False,
),
(
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype_1",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype_2",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
False,
),
(
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.PLUS, None),
validate_alphabet=False,
),
Sequence(
"AAAA",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype",
parent=SingleInterval(0, 4, Strand.UNSTRANDED, None),
validate_alphabet=False,
),
False,
),
(
Sequence("AAAA", Alphabet.NT_STRICT, parent="parent1"),
Sequence("AAAA", Alphabet.NT_STRICT, parent="parent2"),
False,
),
(
Sequence("AAAA", Alphabet.NT_STRICT, parent="parent"),
Sequence("AAAA", Alphabet.NT_STRICT),
False,
),
(
Sequence("AAAA", Alphabet.NT_STRICT, parent=Strand.UNSTRANDED),
Sequence("AAAA", Alphabet.NT_STRICT, parent=Strand.PLUS),
False,
),
(
Sequence("AAAA", Alphabet.NT_STRICT, parent=Strand.UNSTRANDED),
Sequence("AAAA", Alphabet.NT_STRICT),
False,
),
(
Sequence(
"AAAA",
Alphabet.NT_STRICT,
parent="seqtype",
),
Sequence("AAAA", Alphabet.NT_STRICT),
False,
),
],
)
def test_equals(self, sequence, other, expected):
assert (sequence == other) is expected
assert (other == sequence) is expected
def test_str(self):
assert str(Sequence("AAAAt", Alphabet.NT_EXTENDED_GAPPED)) == "AAAAt"
def test_len(self):
assert len(Sequence("", Alphabet.NT_EXTENDED_GAPPED)) == 0
assert len(Sequence("AAAAt", Alphabet.NT_EXTENDED_GAPPED)) == 5
@pytest.mark.parametrize(
"seq,key,exp",
[
# No parent
(Sequence("acgtacgt", Alphabet.NT_STRICT), 3,
Sequence("t", Alphabet.NT_STRICT)),
(Sequence("acgtacgt", Alphabet.NT_STRICT), slice(
3, 6), Sequence("tac", Alphabet.NT_STRICT)),
(Sequence("acgtacgt", Alphabet.NT_STRICT), slice(
3, 10), Sequence("tacgt", Alphabet.NT_STRICT)),
# Parent with location; slice
(
Sequence("actgactg",
Alphabet.NT_STRICT,
parent=SingleInterval(0, 8, Strand.PLUS)),
slice(3, 6),
Sequence("gac",
Alphabet.NT_STRICT,
parent=SingleInterval(3, 6, Strand.PLUS)),
),
(
Sequence("actgactg",
Alphabet.NT_STRICT,
parent=SingleInterval(0, 8, Strand.MINUS)),
slice(3, 6),
Sequence("gac",
Alphabet.NT_STRICT,
parent=SingleInterval(2, 5, Strand.MINUS)),
),
# Parent with location; single position
(
Sequence("actgactg",
Alphabet.NT_STRICT,
parent=SingleInterval(0, 8, Strand.PLUS)),
3,
Sequence("g",
Alphabet.NT_STRICT,
parent=SingleInterval(3, 4, Strand.PLUS)),
),
(
Sequence("actgactg",
Alphabet.NT_STRICT,
parent=SingleInterval(0, 8, Strand.MINUS)),
3,
Sequence("g",
Alphabet.NT_STRICT,
parent=SingleInterval(4, 5, Strand.MINUS)),
),
# Parent without full location
(
Sequence("actgactg", Alphabet.NT_STRICT, parent="parent"),
slice(3, 6),
Sequence("gac", Alphabet.NT_STRICT, parent="parent"),
),
(
Sequence(
"actgactg", Alphabet.NT_STRICT, parent=Strand.UNSTRANDED),
slice(3, 6),
Sequence("gac", Alphabet.NT_STRICT, parent=Strand.UNSTRANDED),
),
],
)
def test_getitem(self, seq, key, exp):
assert seq[key] == exp
def test_getitem_error(self):
with pytest.raises(InvalidStrandException):
Sequence("actgactg",
Alphabet.NT_STRICT,
parent=SingleInterval(0, 8, Strand.UNSTRANDED))[3:6]
@pytest.mark.parametrize(
"sequence,alphabet,validate_alphabet",
[
("", Alphabet.NT_STRICT, True),
("acgtACGT", Alphabet.NT_STRICT, True),
("N", Alphabet.NT_STRICT, False),
("acNNNw", Alphabet.NT_EXTENDED, True),
("AN-", Alphabet.NT_EXTENDED, False),
("GG--AAA", Alphabet.NT_STRICT_GAPPED, True),
("AN-", Alphabet.NT_STRICT_GAPPED, False),
("nnAAw-cg", Alphabet.NT_EXTENDED_GAPPED, True),
("xxx", Alphabet.NT_EXTENDED_GAPPED, False),
("MWT*", Alphabet.AA, True),
("T*-", Alphabet.AA, False),
("ABCDE-", Alphabet.GENERIC, True),
("*", Alphabet.GENERIC, False),
],
)
def test_validate_alphabet(self, sequence, alphabet, validate_alphabet):
Sequence(sequence, alphabet, validate_alphabet=validate_alphabet)
@pytest.mark.parametrize(
"sequence,alphabet",
[
("N", Alphabet.NT_STRICT),
("A-", Alphabet.NT_EXTENDED),
("AN-", Alphabet.NT_STRICT_GAPPED),
("E", Alphabet.NT_EXTENDED_GAPPED),
("R-", Alphabet.AA),
("?", Alphabet.GENERIC),
],
)
def test_validate_alphabet_error(self, sequence, alphabet):
with pytest.raises(AlphabetError):
Sequence(sequence, alphabet, validate_alphabet=True)
@pytest.mark.parametrize(
"sequence,expected",
[
(Sequence("A", Alphabet.NT_STRICT, parent="parent"), "parent"),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(id="parent",
location=SingleInterval(
10, 11, Strand.UNSTRANDED)),
),
"parent",
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
location=SingleInterval(
10, 11, Strand.UNSTRANDED, parent="parent"),
),
),
"parent",
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(location=SingleInterval(
10, 11, Strand.UNSTRANDED, parent="parent")),
),
"parent",
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=SingleInterval(10, 11, Strand.UNSTRANDED),
),
None,
),
(Sequence("A", Alphabet.NT_STRICT), None),
],
)
def test_parent_id(self, sequence, expected):
assert sequence.parent_id == expected
@pytest.mark.parametrize(
"sequence,expected",
[
(Sequence("A", Alphabet.NT_STRICT), None),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(sequence_type="seqtype"),
),
"seqtype",
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(location=SingleInterval(
0,
1,
Strand.PLUS,
parent=Parent(sequence_type="seqtype"),
)),
),
"seqtype",
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(location=SingleInterval(
0,
1,
Strand.PLUS,
parent=Sequence(
"AA", Alphabet.NT_STRICT, type="seqtype"),
)),
),
"seqtype",
),
],
)
def test_parent_type(self, sequence, expected):
assert sequence.parent_type == expected
@pytest.mark.parametrize(
"sequence,expected",
[
(Sequence("A", Alphabet.NT_STRICT), None),
(
Sequence("A", Alphabet.NT_STRICT, parent=Strand.MINUS),
Strand.MINUS,
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=SingleInterval(10, 11, Strand.MINUS),
),
Strand.MINUS,
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=SingleInterval(10, 11, Strand.MINUS),
),
Strand.MINUS,
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
parent=Parent(
strand=Strand.MINUS,
location=SingleInterval(10, 11, Strand.MINUS),
),
),
Strand.MINUS,
),
],
)
def test_parent_strand(self, sequence, expected):
assert sequence.parent_strand == expected
@pytest.mark.parametrize(
"sequence,new_id,new_type,expected",
[
(
Sequence("", Alphabet.NT_STRICT),
None,
None,
Sequence("", Alphabet.NT_STRICT),
),
(
Sequence("ACGtacgT", Alphabet.NT_STRICT),
None,
None,
Sequence("AcgtaCGT", Alphabet.NT_STRICT),
),
(
Sequence("ATUGCYRSWKMBdhvnNVHDbmkwsrycguta",
Alphabet.NT_EXTENDED),
None,
None,
Sequence("taacgryswmkvHDBNnbdhVKMWSYRGCAAT",
Alphabet.NT_EXTENDED),
),
(
Sequence("--A-CGta", Alphabet.NT_STRICT_GAPPED),
None,
None,
Sequence("taCG-T--", Alphabet.NT_STRICT_GAPPED),
),
(
Sequence("AtUC-N-", Alphabet.NT_EXTENDED_GAPPED),
None,
None,
Sequence("-N-GAaT", Alphabet.NT_EXTENDED_GAPPED),
),
(
Sequence("ACGta", Alphabet.NT_STRICT),
"new_id",
"seqtype",
Sequence(
"taCGT",
Alphabet.NT_STRICT,
id="new_id",
type="seqtype",
),
),
(
Sequence("ACGta", Alphabet.NT_STRICT, parent=Strand.PLUS),
None,
None,
Sequence("taCGT", Alphabet.NT_STRICT, parent=Strand.MINUS),
),
(
Sequence(
"ACGta",
Alphabet.NT_STRICT,
parent=SingleInterval(5, 10, Strand.PLUS),
),
None,
None,
Sequence(
"taCGT",
Alphabet.NT_STRICT,
parent=SingleInterval(5, 10, Strand.MINUS),
),
),
],
)
def test_reverse_complement(self, sequence, new_id, new_type, expected):
assert sequence.reverse_complement(new_id=new_id,
new_type=new_type) == expected
@pytest.mark.parametrize(
"sequence",
[
Sequence("AAA", Alphabet.AA),
Sequence("AAA", Alphabet.GENERIC),
Sequence("xxx", Alphabet.NT_STRICT, validate_alphabet=False),
],
)
def test_reverse_complement_error(self, sequence):
with pytest.raises(AlphabetError):
sequence.reverse_complement()
@pytest.mark.parametrize(
"seq1,seq2,new_id,data_only,expected",
[
(
Sequence("", Alphabet.NT_STRICT, parent="parent1"),
Sequence("", Alphabet.NT_STRICT, parent="parent2"),
"new_id",
True,
Sequence("", Alphabet.NT_STRICT, id="new_id"),
),
(
Sequence("AA", Alphabet.NT_STRICT, parent="parent1"),
Sequence("TT", Alphabet.NT_STRICT, parent="parent2"),
"new_id",
True,
Sequence("AATT", Alphabet.NT_STRICT, id="new_id"),
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
type="seqtype_1",
parent=Parent(
id="parent1",
strand=Strand.PLUS,
location=SingleInterval(5, 7, Strand.PLUS),
),
),
Sequence(
"TT",
Alphabet.NT_STRICT,
type="seqtype_2",
parent=Parent(
id="parent1",
strand=Strand.MINUS,
location=SingleInterval(20, 22, Strand.MINUS),
),
),
None,
True,
Sequence("AATT", Alphabet.NT_STRICT),
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(
id="parent1",
strand=Strand.PLUS,
location=SingleInterval(5, 7, Strand.PLUS),
),
),
Sequence(
"TT",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(id="parent1", strand=Strand.PLUS),
),
None,
False,
Sequence(
"AATT",
Alphabet.NT_STRICT,
type="seqtype",
parent="parent1",
),
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(id="parent1", strand=Strand.PLUS),
),
Sequence(
"TT",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(
id="parent1",
strand=Strand.PLUS,
location=SingleInterval(5, 7, Strand.PLUS),
),
),
None,
False,
Sequence(
"AATT",
Alphabet.NT_STRICT,
type="seqtype",
parent="parent1",
),
),
(
Sequence(
"CC",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(id="parent",
location=SingleInterval(3, 5, Strand.PLUS)),
),
Sequence(
"TT",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(id="parent",
location=SingleInterval(10, 12,
Strand.PLUS)),
),
"new_id",
True,
Sequence("CCTT", Alphabet.NT_STRICT, id="new_id"),
),
(
Sequence(
"CC",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(id="parent",
location=SingleInterval(3, 5, Strand.PLUS)),
),
Sequence(
"TT",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(id="parent",
location=SingleInterval(0, 2, Strand.PLUS)),
),
"new_id",
True,
Sequence("CCTT", Alphabet.NT_STRICT, id="new_id"),
),
(
Sequence("AA", Alphabet.NT_STRICT, id="seq1", parent="parent"),
Sequence("", Alphabet.NT_STRICT, id="seq2", parent="parent"),
None,
False,
Sequence("AA", Alphabet.NT_STRICT, parent="parent"),
),
(
Sequence(
"ACT",
Alphabet.NT_STRICT,
parent=Parent(id="parent",
location=SingleInterval(2, 5, Strand.PLUS)),
),
Sequence(
"GGA",
Alphabet.NT_STRICT,
parent=Parent(id="parent",
location=SingleInterval(8, 11, Strand.PLUS)),
),
"new_id",
False,
Sequence(
"ACTGGA",
Alphabet.NT_STRICT,
id="new_id",
parent=Parent(
id="parent",
location=CompoundInterval.from_single_intervals([
SingleInterval(2, 5, Strand.PLUS),
SingleInterval(8, 11, Strand.PLUS),
]),
),
),
),
(
Sequence(
"ACT",
Alphabet.NT_STRICT,
parent=Parent(id="parent",
location=SingleInterval(8, 11,
Strand.MINUS)),
),
Sequence(
"GGA",
Alphabet.NT_STRICT,
parent=Parent(id="parent",
location=SingleInterval(2, 5, Strand.MINUS)),
),
"new_id",
False,
Sequence(
"ACTGGA",
Alphabet.NT_STRICT,
id="new_id",
parent=Parent(
id="parent",
location=CompoundInterval.from_single_intervals([
SingleInterval(2, 5, Strand.MINUS),
SingleInterval(8, 11, Strand.MINUS),
]),
),
),
),
(
Sequence(
"ACT",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype_1",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
strand=Strand.PLUS,
location=SingleInterval(5, 8, Strand.PLUS),
),
),
Sequence(
"GCG",
Alphabet.NT_STRICT,
id="seq1",
type="seqtype_1",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
strand=Strand.PLUS,
location=SingleInterval(15, 18, Strand.PLUS),
),
),
None,
False,
Sequence(
"ACTGCG",
Alphabet.NT_STRICT,
type="seqtype_1",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
strand=Strand.PLUS,
location=CompoundInterval.from_single_intervals([
SingleInterval(5, 8, Strand.PLUS),
SingleInterval(15, 18, Strand.PLUS),
]),
),
),
),
],
)
def test_append(self, seq1, seq2, new_id, data_only, expected):
assert seq1.append(seq2, new_id, data_only) == expected
@pytest.mark.parametrize(
"seq1,seq2,new_id,data_only",
[
(
Sequence("AA", Alphabet.NT_STRICT),
Sequence("TT", Alphabet.NT_EXTENDED),
None,
True,
),
(
Sequence("AA", Alphabet.NT_STRICT, type="seqtype_1"),
Sequence("AA", Alphabet.NT_STRICT, type="seqtype_2"),
None,
False,
),
(
Sequence("AA", Alphabet.NT_STRICT, parent="parent1"),
Sequence("AA", Alphabet.NT_STRICT, parent="parent2"),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=SingleInterval(10, 12, Strand.PLUS),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=SingleInterval(5, 7, Strand.PLUS),
),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=SingleInterval(2, 4, Strand.MINUS),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=SingleInterval(5, 7, Strand.MINUS),
),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=SingleInterval(10, 12, Strand.PLUS),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=SingleInterval(11, 13, Strand.PLUS),
),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=SingleInterval(10, 12, Strand.UNSTRANDED),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=SingleInterval(15, 17, Strand.UNSTRANDED),
),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent1",
sequence_type="seqtype",
strand=Strand.PLUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent2",
sequence_type="seqtype",
strand=Strand.PLUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
sequence_type="seqtype_1",
strand=Strand.PLUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
sequence_type="seqtype_2",
strand=Strand.PLUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
sequence_type="seqtype",
strand=Strand.PLUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
sequence_type="seqtype",
strand=Strand.MINUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
sequence_type="seqtype",
strand=Strand.PLUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
sequence_type="seqtype",
strand=Strand.PLUS,
sequence=Sequence("AAAT", Alphabet.NT_STRICT),
parent="grandparent",
),
),
None,
False,
),
(
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
sequence_type="seqtype",
strand=Strand.PLUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent1",
),
),
Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(
id="parent",
sequence_type="seqtype",
strand=Strand.PLUS,
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent2",
),
),
None,
False,
),
],
)
def test_append_error(self, seq1, seq2, new_id, data_only):
with pytest.raises(ValueError):
seq1.append(seq2, new_id, data_only)
@pytest.mark.parametrize(
"sequence,sequence_type,include_self,expected",
[
(
Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
"seqtype",
True,
Parent(sequence=Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
)),
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
"seqtype",
False,
Parent(id="parent", sequence_type="seqtype"),
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype_1",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
parent=Parent(id="grandparent",
sequence_type="seqtype_2"),
),
),
"seqtype_2",
True,
Parent(
id="parent",
sequence_type="seqtype_2",
parent=Parent(id="grandparent", sequence_type="seqtype_2"),
),
),
],
)
def test_first_ancestor_of_type(self, sequence, sequence_type,
include_self, expected):
assert sequence.first_ancestor_of_type(
sequence_type, include_self=include_self) == expected
@pytest.mark.parametrize(
"sequence,sequence_type,include_self",
[
(
Sequence("A", Alphabet.NT_STRICT, id="self"),
"seqtype_1",
True,
),
(
Sequence("A", Alphabet.NT_STRICT, id="self", parent="parent"),
"seqtype_1",
True,
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype_2",
parent=Parent(
id="parent",
sequence_type="seqtype_1",
parent=Parent(id="grandparent",
sequence_type="seqtype_1"),
),
),
"seqtype_3",
True,
),
],
)
def test_first_ancestor_of_type_error(self, sequence, sequence_type,
include_self):
with pytest.raises(NoSuchAncestorException):
sequence.first_ancestor_of_type(sequence_type,
include_self=include_self)
@pytest.mark.parametrize(
"sequence,sequence_type,include_self,expected",
[
(
Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
"seqtype",
True,
True,
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
"seqtype",
False,
True,
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype_1",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
parent=Parent(id="grandparent",
sequence_type="seqtype_2"),
),
),
"seqtype_2",
True,
True,
),
(
Sequence("A", Alphabet.NT_STRICT, id="self"),
"seqtype_1",
True,
False,
),
(
Sequence("A", Alphabet.NT_STRICT, id="self", parent="parent"),
"seqtype_1",
True,
False,
),
(
Sequence(
"A",
Alphabet.NT_STRICT,
id="self",
type="seqtype_2",
parent=Parent(
id="parent",
sequence_type="seqtype_1",
parent=Parent(id="grandparent",
sequence_type="seqtype_1"),
),
),
"seqtype_3",
True,
False,
),
],
)
def test_has_ancestor_of_type(self, sequence, sequence_type, include_self,
expected):
assert sequence.has_ancestor_of_type(
sequence_type, include_self=include_self) is expected
@pytest.mark.parametrize(
"sequence,expected",
[
(Sequence("ATGCATATTTGGAAACCAA", Alphabet.NT_STRICT,
id="test"), ">test\nATGCATATTT\nGGAAACCAA"),
(Sequence("ATGCATATTTGGAAACCAA",
Alphabet.NT_STRICT), ">None\nATGCATATTT\nGGAAACCAA"),
(Sequence("GGAAACCAA", Alphabet.NT_STRICT,
id="test"), ">test\nGGAAACCAA"),
(
Sequence("ATGCATATTTGGAAACCAAGGAAACCAA",
Alphabet.NT_STRICT,
id="test"),
">test\nATGCATATTT\nGGAAACCAAG\nGAAACCAA",
),
(
Sequence(
data="AAAAAAA",
alphabet=Alphabet.NT_STRICT,
id="test",
parent=Parent(
location=SingleInterval(33, 40, Strand.MINUS)),
),
">test\nAAAAAAA",
),
],
)
def test_to_fasta(self, sequence, expected):
s_fa = sequence.to_fasta(num_chars=10)
assert s_fa == expected
def test_empty_to_fasta(self):
s = Sequence("", Alphabet.NT_STRICT)
with pytest.raises(EmptySequenceFastaError):
s.to_fasta()
def test_contains(self):
assert not EmptyLocation().contains(EmptyLocation())
assert not EmptyLocation().contains(SingleInterval(0, 1, Strand.PLUS))
def test_union_preserve_overlaps(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().union_preserve_overlaps(
SingleInterval(0, 1, Strand.PLUS))
def test_parent_to_relative_location(self):
with pytest.raises(EmptyLocationException):
EmptyLocation().parent_to_relative_location(
SingleInterval(0, 0, Strand.PLUS))
class TestAbstractInterval:
@pytest.mark.parametrize(
"location,parent,exp",
[
# no-op
(
SingleInterval(0, 10, Strand.PLUS),
None,
SingleInterval(0, 10, Strand.PLUS),
),
# non-specified parent
(SingleInterval(0, 10, Strand.PLUS), Parent(),
SingleInterval(0, 10, Strand.PLUS, parent=Parent())),
# specified parent that is non-standard
(
SingleInterval(0, 10, Strand.PLUS),
Parent(sequence_type="nonstandard"),
SingleInterval(0,
10,
Strand.PLUS,
parent=Parent(sequence_type="nonstandard")),
),
# chromosome parent
(
SingleInterval(0, 10, Strand.PLUS),
Parent(sequence_type=SequenceType.CHROMOSOME),
SingleInterval(
0,
10,
Strand.PLUS,
parent=Parent(sequence_type=SequenceType.CHROMOSOME)),
),
# chromosome parent with real sequence
(
SingleInterval(0, 10, Strand.PLUS),
Parent(sequence_type=SequenceType.CHROMOSOME,
sequence=Sequence("ATACGATCAAA",
Alphabet.NT_EXTENDED_GAPPED)),
SingleInterval(
0,
10,
Strand.PLUS,
parent=Parent(
sequence_type=SequenceType.CHROMOSOME,
sequence=Sequence("ATACGATCAAA",
Alphabet.NT_EXTENDED_GAPPED),
),
),
),
# chromosome parent as would be built by seq_to_parent
(
SingleInterval(0, 10, Strand.PLUS),
Parent(
sequence=Sequence("ATACGATCAAA",
Alphabet.NT_EXTENDED_GAPPED,
type=SequenceType.CHROMOSOME,
id="test"),
location=SingleInterval(0, 11, Strand.PLUS),
),
SingleInterval(
0,
10,
Strand.PLUS,
Parent(
sequence=Sequence("ATACGATCAAA",
Alphabet.NT_EXTENDED_GAPPED,
type=SequenceType.CHROMOSOME,
id="test"),
location=SingleInterval(0, 11, Strand.PLUS),
),
),
),
# chunk parent as would be built by seq_chunk_to_parent
(
SingleInterval(2, 10, Strand.PLUS),
Parent(
id="test:1-11",
sequence=Sequence(
"TACGATCAAA",
Alphabet.NT_EXTENDED_GAPPED,
id="test:1-11",
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(location=SingleInterval(
1,
11,
Strand.PLUS,
parent=Parent(
id="test",
sequence_type=SequenceType.CHROMOSOME),
)),
),
),
# chunk-relative location is 1-9
SingleInterval(
1,
9,
Strand.PLUS,
Parent(
id="test:1-11",
sequence=Sequence(
"TACGATCAAA",
Alphabet.NT_EXTENDED_GAPPED,
id="test:1-11",
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(location=SingleInterval(
1,
11,
Strand.PLUS,
parent=Parent(
id="test",
sequence_type=SequenceType.CHROMOSOME),
)),
),
),
),
),
],
)
def test_liftover_location_to_seq_chunk_parent(self, location, parent,
exp):
obs = AbstractInterval.liftover_location_to_seq_chunk_parent(
location, parent)
assert obs == exp
@pytest.mark.parametrize(
"location,parent,exception",
[
# cannot have a naked chunk parent
(
SingleInterval(0, 10, Strand.PLUS),
Parent(sequence_type=SequenceType.SEQUENCE_CHUNK),
NoSuchAncestorException,
),
# cannot have a sequence-less chunk parent
(
SingleInterval(0, 10, Strand.PLUS),
Parent(
id="test:1-11",
sequence_type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(location=SingleInterval(
1,
11,
Strand.PLUS,
parent=Parent(id="test",
sequence_type=SequenceType.CHROMOSOME),
)),
),
NullSequenceException,
),
# if location is on a chunk, must be a proper chunk with a chromosome
(
SingleInterval(
0,
10,
Strand.PLUS,
parent=Parent(sequence_type=SequenceType.SEQUENCE_CHUNK)),
Parent(),
NoSuchAncestorException,
),
# if location is on a chunk, that chunk's chromosome must match the new parent
(
SingleInterval(
5,
8,
Strand.PLUS,
parent=Parent(
id="test:0-9",
sequence=Sequence(
"ATACGATCA",
Alphabet.NT_EXTENDED_GAPPED,
id="test:0-9",
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(location=SingleInterval(
0,
9,
Strand.PLUS,
parent=Parent(
id="test",
sequence_type=SequenceType.CHROMOSOME),
)),
),
),
),
Parent(
id="wrongchrom:0-9",
sequence=Sequence(
"ATACGATCA",
Alphabet.NT_EXTENDED_GAPPED,
id="wrongchrom:0-9",
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(location=SingleInterval(
0,
9,
Strand.PLUS,
parent=Parent(
id="wrongchrom",
sequence_type=SequenceType.CHROMOSOME),
)),
),
),
MismatchedParentException,
),
# if location is on a chunk, and the chromosome parent has sequence, sequence must match
(
SingleInterval(
5,
8,
Strand.PLUS,
parent=Parent(
id="test:0-9",
sequence=Sequence(
"ATACGATCA",
Alphabet.NT_EXTENDED_GAPPED,
id="test:0-9",
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(
sequence=Sequence(
"ATACGATCAAAA",
Alphabet.NT_EXTENDED_GAPPED,
type=SequenceType.CHROMOSOME),
location=SingleInterval(
0,
9,
Strand.PLUS,
parent=Parent(
id="test",
sequence_type=SequenceType.CHROMOSOME),
),
),
),
),
),
Parent(
id="test:0-9",
sequence=Sequence(
"ATACGATCA",
Alphabet.NT_EXTENDED_GAPPED,
id="test:0-9",
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(
sequence=Sequence("ATACGATCAATA",
Alphabet.NT_EXTENDED_GAPPED,
type=SequenceType.CHROMOSOME),
location=SingleInterval(
0,
9,
Strand.PLUS,
parent=Parent(
id="test",
sequence_type=SequenceType.CHROMOSOME),
),
),
),
),
MismatchedParentException,
),
],
)
def test_liftover_location_to_seq_chunk_parent_exceptions(
self, location, parent, exception):
with pytest.raises(exception):
_ = AbstractInterval.liftover_location_to_seq_chunk_parent(
location, parent)
class TestFeatureIntervalSequenceSubset:
"""Test the ability to slice the genomic sequence of a feature interval and still get useful results."""
@pytest.mark.parametrize(
"schema,absolute_value,relative_value,expected,parent",
[
(e3_spliced, 2, 1, 0, parent_genome2_1_15),
(e3_spliced, 7, 6, 4, parent_genome2_1_15),
(e3_spliced, 14, 13, 9, parent_genome2_1_15),
(e3_spliced_minus, 2, 1, 9, parent_genome2_1_15),
(e3_spliced_minus, 7, 6, 5, parent_genome2_1_15),
(e3_spliced_minus, 14, 13, 0, parent_genome2_1_15),
],
)
def test_sequence_pos_to_feature(self, schema, absolute_value, relative_value, expected, parent):
feat = schema.to_feature_interval(parent)
assert feat.sequence_pos_to_feature(absolute_value) == expected
assert feat.chunk_relative_pos_to_feature(relative_value) == expected
@pytest.mark.parametrize(
"schema,absolute_value,relative_value,expected,parent",
[
(
e3_spliced,
(7, 13, Strand.PLUS),
(6, 12, Strand.PLUS),
SingleInterval(4, 8, Strand.PLUS),
parent_genome2_1_15,
),
(
e3_spliced_minus,
(7, 13, Strand.PLUS),
(6, 12, Strand.PLUS),
(SingleInterval(2, 6, Strand.MINUS)),
parent_genome2_1_15,
),
],
)
def test_sequence_interval_to_feature(self, schema, absolute_value, relative_value, expected, parent):
feat = schema.to_feature_interval(parent)
assert feat.sequence_interval_to_feature(*absolute_value) == expected
assert feat.chunk_relative_interval_to_feature(*relative_value) == expected
@pytest.mark.parametrize(
"schema,value,absolute_expected,relative_expected,parent",
[
(e3_spliced, 0, 2, 1, parent_genome2_1_15),
(e3_spliced, 9, 14, 13, parent_genome2_1_15),
(e3_spliced, 4, 7, 6, parent_genome2_1_15),
(e3_spliced_minus, 0, 14, 13, parent_genome2_1_15),
(e3_spliced_minus, 9, 2, 1, parent_genome2_1_15),
(e3_spliced_minus, 5, 7, 6, parent_genome2_1_15),
],
)
def test_feature_pos_to_sequence(self, schema, value, absolute_expected, relative_expected, parent):
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
assert feat.feature_pos_to_sequence(value) == absolute_expected
assert feat.feature_pos_to_chunk_relative(value) == relative_expected
@pytest.mark.parametrize(
"schema,value,absolute_expected,relative_expected,parent",
[
(
e3_spliced,
(0, 5, Strand.PLUS),
CompoundInterval([2, 7], [6, 8], Strand.PLUS),
CompoundInterval([1, 6], [5, 7], Strand.PLUS),
parent_genome2_1_15,
),
(
e3_spliced_minus,
(0, 5, Strand.PLUS),
CompoundInterval([8, 12], [10, 15], Strand.MINUS),
CompoundInterval([7, 11], [9, 14], Strand.MINUS),
parent_genome2_1_15,
),
],
)
def test_feature_interval_to_sequence(self, schema, value, absolute_expected, relative_expected, parent):
feat = schema.to_feature_interval(parent)
assert feat.feature_interval_to_sequence(*value).reset_parent(None) == absolute_expected
assert feat.feature_interval_to_chunk_relative(*value).reset_parent(None) == relative_expected
@pytest.mark.parametrize(
"schema,parent,expected_spliced",
[
(e3_spliced, parent_genome2_1_15, "GTATCTTACC"),
(
e3_spliced,
parent_genome2_1_15,
"GTATCTTACC",
),
(e3_spliced_minus, parent_genome2_1_15, "GGTAAGATAC"),
(
e3_spliced_minus,
parent_genome2_1_15,
"GGTAAGATAC",
),
],
)
def test_spliced_sequence(self, schema, parent, expected_spliced):
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
assert str(feat.get_spliced_sequence()) == expected_spliced
@pytest.mark.parametrize(
"schema,parent,expected_genomic,expected_stranded_genomic",
[
(e3_spliced, parent_genome2_1_15, "GTATTCTTGGACC", "GTATTCTTGGACC"),
(
e3_spliced,
parent_genome2_1_15,
"GTATTCTTGGACC",
"GTATTCTTGGACC",
),
(e3_spliced_minus, parent_genome2_1_15, "GTATTCTTGGACC", "GGTCCAAGAATAC"),
],
)
def test_genomic_sequence(self, schema, parent, expected_genomic, expected_stranded_genomic):
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
assert str(feat.get_reference_sequence()) == expected_genomic
assert str(feat.get_genomic_sequence()) == expected_stranded_genomic
def test_sequence_exceptions(self):
"""All sequence accessors should raise good errors when attempted without sequence info"""
feat = e3_spliced.to_feature_interval(parent_or_seq_chunk_parent=parent_no_seq)
with pytest.raises(NullSequenceException):
_ = feat.get_reference_sequence()
with pytest.raises(NullSequenceException):
_ = feat.get_spliced_sequence()
with pytest.raises(NullSequenceException):
_ = feat.get_genomic_sequence()
def test_start_end(self):
feat = e3_spliced.to_feature_interval(parent_or_seq_chunk_parent=parent_genome2_1_15)
assert feat.chunk_relative_start + 1 == feat.start
assert feat.chunk_relative_end + 1 == feat.end
def test_dict(self):
feat = e3_spliced.to_feature_interval(parent_genome2_1_15)
feat2 = e3_spliced.to_feature_interval()
d = feat.to_dict()
d2 = feat2.to_dict()
del d["feature_interval_guid"]
del d2["feature_interval_guid"]
assert d == d2
rel_d = feat.to_dict(chromosome_relative_coordinates=False)
del rel_d["feature_interval_guid"]
assert rel_d != d
assert [x + 1 for x in rel_d["interval_starts"]] == list(d["interval_starts"])
def test_nonstandard_parents(self):
feat0 = e3_spliced.to_feature_interval(parent)
seq0 = feat0.get_spliced_sequence()
feat1 = e3_spliced.to_feature_interval(parent_nonstandard_type)
with pytest.raises(NullSequenceException):
_ = feat1.get_spliced_sequence()
feat2 = e3_spliced.to_feature_interval(parent_no_seq)
with pytest.raises(NullSequenceException):
_ = feat2.get_spliced_sequence()
feat3 = e3_spliced.to_feature_interval(parent_nonstandard_type_with_sequence)
seq = feat3.get_spliced_sequence()
assert seq == seq0
assert feat0.chromosome_location == feat0.chunk_relative_location
assert feat1.chromosome_location != feat1.chunk_relative_location
assert feat1._chunk_relative_bounded_chromosome_location == feat1.chunk_relative_location
assert feat2.chromosome_location == feat2.chunk_relative_location
assert feat3.chromosome_location != feat3.chunk_relative_location
assert feat3._chunk_relative_bounded_chromosome_location == feat3.chunk_relative_location
# OTOH, this is not the same
feat4 = e3_spliced.to_feature_interval(parent_genome2_1_15)
assert feat4.chromosome_location != feat4.chunk_relative_location
def test_liftover_to_parent_or_seq_chunk_parent(self):
feat0 = e3_spliced.to_feature_interval(parent_genome2)
feat1 = feat0.liftover_to_parent_or_seq_chunk_parent(parent_genome2_1_15)
assert str(feat0.get_spliced_sequence()) == str(feat1.get_spliced_sequence())
assert feat0.chromosome_location.reset_parent(None) == feat1.chromosome_location.reset_parent(None)
# bringing in a null parent means no sequence anymore
feat2 = feat0.liftover_to_parent_or_seq_chunk_parent(parent_no_seq_with_id)
with pytest.raises(NullSequenceException):
_ = feat2.get_spliced_sequence()
# we can also start in chunk coordinates, then lift to different chunk coordinates
feat_chunk = e3_spliced.to_feature_interval(parent_genome2_1_15)
feat_subchunk = feat_chunk.liftover_to_parent_or_seq_chunk_parent(parent_genome2_2_8)
# this is now a subset
assert str(feat_subchunk.get_spliced_sequence()) in str(feat_chunk.get_spliced_sequence())
def test_liftover_to_parent_or_seq_chunk_parent_exception(self):
feat0 = e3_spliced.to_feature_interval(parent_genome2_1_15)
with pytest.raises(MismatchedParentException):
_ = feat0.liftover_to_parent_or_seq_chunk_parent(parent_named)
feat1 = e3_spliced.to_feature_interval(parent_genome2)
with pytest.raises(MismatchedParentException):
_ = feat1.liftover_to_parent_or_seq_chunk_parent(parent_named)
feat2 = e3_spliced.to_feature_interval(parent_genome2_1_15)
with pytest.raises(MismatchedParentException):
_ = feat2.liftover_to_parent_or_seq_chunk_parent(parent_no_seq)
with pytest.raises(MismatchedParentException):
_ = feat2.liftover_to_parent_or_seq_chunk_parent(parent)
@pytest.mark.parametrize(
"feature,parent,expected",
[
(
e3_spliced,
parent_genome2_2_8,
CompoundInterval(
[2, 7, 12],
[6, 10, 15],
Strand.PLUS,
parent=parent_genome2_2_8.first_ancestor_of_type(SequenceType.CHROMOSOME),
),
),
(
e3_spliced,
parent_genome2_1_15,
CompoundInterval(
[2, 7, 12],
[6, 10, 15],
Strand.PLUS,
parent=parent_genome2_1_15.first_ancestor_of_type(SequenceType.CHROMOSOME),
),
),
],
)
def test_chromosome_location(self, feature, parent, expected):
feat = feature.to_feature_interval(parent)
assert feat.chromosome_location == expected
@pytest.mark.parametrize(
"feature,parent,expected",
[
# parent_genome2_2_8 slices off the first exon
(
e3_spliced,
parent_genome2_2_8,
CompoundInterval(
[2, 7],
[6, 8],
Strand.PLUS,
parent=parent_genome2_2_8.first_ancestor_of_type(SequenceType.CHROMOSOME),
),
),
# parent_genome2_1_15 does not slice off anything
(
e3_spliced,
parent_genome2_1_15,
CompoundInterval(
[2, 7, 12],
[6, 10, 15],
Strand.PLUS,
parent=parent_genome2_1_15.first_ancestor_of_type(SequenceType.CHROMOSOME),
),
),
],
)
def test__chunk_relative_bounded_chromosome_location(self, feature, parent, expected):
feat = feature.to_feature_interval(parent)
assert feat._chunk_relative_bounded_chromosome_location == expected
@pytest.mark.parametrize(
"feature,parent,expected_gaps",
[
(
e3_spliced,
parent_genome2_1_15,
CompoundInterval([5, 9], [6, 11], Strand.PLUS, parent=parent_genome2_1_15),
),
(
e3_spliced_minus,
parent_genome2_1_15,
CompoundInterval([5, 9], [6, 11], Strand.MINUS, parent=parent_genome2_1_15),
),
(e3_spliced, parent_genome2_2_8, SingleInterval(4, 5, Strand.PLUS, parent=parent_genome2_2_8)),
(e3_spliced_minus, parent_genome2_2_8, SingleInterval(4, 5, Strand.MINUS, parent=parent_genome2_2_8)),
],
)
def test_chunk_relative_gaps_location(self, feature, parent, expected_gaps):
feat = feature.to_feature_interval(parent)
ObjectValidation.require_locations_have_same_nonempty_parent(feat.chunk_relative_gaps_location, expected_gaps)
@pytest.mark.parametrize(
"feature,parent,expected_span",
[
(e3_spliced, parent_genome2_1_15, SingleInterval(2, 14, Strand.PLUS, parent=parent_genome2_1_15)),
(e3_spliced_minus, parent_genome2_1_15, SingleInterval(2, 14, Strand.MINUS, parent=parent_genome2_1_15)),
(e3_spliced, parent_genome2_2_8, SingleInterval(2, 6, Strand.PLUS, parent=parent_genome2_2_8)),
(e3_spliced_minus, parent_genome2_2_8, SingleInterval(2, 6, Strand.MINUS, parent=parent_genome2_2_8)),
],
)
def test_chunk_relative_span(self, feature, parent, expected_span):
feat = feature.to_feature_interval(parent)
ObjectValidation.require_locations_have_same_nonempty_parent(expected_span, feat.chunk_relative_span)
def test_require_location_has_parent(self):
with pytest.raises(NullParentException):
ObjectValidation.require_location_has_parent(
SingleInterval(0, 5, Strand.PLUS))
ObjectValidation.require_location_has_parent(
SingleInterval(0, 5, Strand.PLUS, parent="parent"))
class TestFeatureInterval:
"""Test constructing features of various types"""
@pytest.mark.parametrize(
"schema,expected",
[
(se_unspliced, se_unspliced_repr),
(e3_spliced, e3_spliced_repr),
(e3_spliced, e3_spliced_repr),
(se_unspliced_minus, se_unspliced_repr_minus),
(e3_spliced_minus, e3_spliced_repr_minus),
],
)
def test_feature_constructor(self, schema, expected):
assert str(schema.to_feature_interval()) == expected
assert str(schema.to_feature_interval(parent_or_seq_chunk_parent=parent)) == expected
@pytest.mark.parametrize(
"schema,expected_exception",
[
(
FeatureIntervalModel.Schema().load(
dict(interval_starts=[0], interval_ends=[], strand=Strand.PLUS.name)
),
ValidationException,
),
(
FeatureIntervalModel.Schema().load(
dict(interval_starts=[0], interval_ends=[2, 5], strand=Strand.PLUS.name)
),
ValidationException,
),
(
FeatureIntervalModel.Schema().load(
dict(interval_starts=[], interval_ends=[2, 5], strand=Strand.PLUS.name)
),
ValidationException,
),
],
)
def test_feature_excecptions(self, schema, expected_exception):
with pytest.raises(expected_exception):
_ = schema.to_feature_interval()
@pytest.mark.parametrize(
"schema,value,expected",
[
(e3_spliced, 2, 0),
(e3_spliced, 7, 4),
(e3_spliced, 14, 9),
(e3_spliced_minus, 2, 9),
(e3_spliced_minus, 7, 5),
(e3_spliced_minus, 14, 0),
],
)
def test_sequence_pos_to_feature(self, schema, value, expected):
feat = schema.to_feature_interval()
assert feat.sequence_pos_to_feature(value) == expected
assert feat.chunk_relative_pos_to_feature(value) == expected
@pytest.mark.parametrize(
"schema,value,expected",
[
(e3_spliced, (7, 13, Strand.PLUS), SingleInterval(4, 8, Strand.PLUS)),
(e3_spliced_minus, (7, 13, Strand.PLUS), (SingleInterval(2, 6, Strand.MINUS))),
],
)
def test_sequence_interval_to_feature(self, schema, value, expected):
feat = schema.to_feature_interval()
assert feat.sequence_interval_to_feature(*value) == expected
assert feat.chunk_relative_interval_to_feature(*value) == expected
@pytest.mark.parametrize(
"schema,value,expected",
[
(e3_spliced, 0, 2),
(e3_spliced, 9, 14),
(e3_spliced, 4, 7),
(e3_spliced_minus, 0, 14),
(e3_spliced_minus, 9, 2),
(e3_spliced_minus, 5, 7),
],
)
def test_feature_pos_to_sequence(self, schema, value, expected):
feat = schema.to_feature_interval()
assert feat.feature_pos_to_sequence(value) == expected
assert feat.feature_pos_to_chunk_relative(value) == expected
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
assert feat.feature_pos_to_sequence(value) == expected
assert feat.feature_pos_to_chunk_relative(value) == expected
@pytest.mark.parametrize(
"schema,value,expected",
[
(e3_spliced, (0, 5, Strand.PLUS), CompoundInterval([2, 7], [6, 8], Strand.PLUS)),
(e3_spliced_minus, (0, 5, Strand.PLUS), CompoundInterval([8, 12], [10, 15], Strand.MINUS)),
],
)
def test_feature_interval_to_sequence(self, schema, value, expected):
feat = schema.to_feature_interval()
assert feat.feature_interval_to_sequence(*value).reset_parent(None) == expected
assert feat.feature_interval_to_chunk_relative(*value).reset_parent(None) == expected
@pytest.mark.parametrize(
"schema,value,expected",
[
(e3_spliced, (0, 5, Strand.PLUS), CompoundInterval([2, 7], [6, 8], Strand.PLUS, parent=parent)),
(e3_spliced_minus, (0, 5, Strand.PLUS), CompoundInterval([8, 12], [10, 15], Strand.MINUS, parent=parent)),
],
)
def test_feature_interval_to_sequence_parent(self, schema, value, expected):
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
assert feat.feature_interval_to_sequence(*value) == expected
assert feat.feature_interval_to_chunk_relative(*value) == expected
@pytest.mark.parametrize(
"schema,value,expected",
[
(
e3_spliced,
SingleInterval(0, 10, Strand.PLUS, parent=parent),
dict(interval_starts=(2, 7), interval_ends=(6, 10), strand=Strand.PLUS.name),
),
(
e3_spliced_minus,
SingleInterval(0, 10, Strand.PLUS, parent=parent),
dict(interval_starts=(2, 7), interval_ends=(6, 10), strand=Strand.MINUS.name),
),
],
)
def test_intersection(self, schema, value, expected):
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
val = feat.intersect(value)
expected = FeatureIntervalModel.Schema().load(expected).to_feature_interval(parent_or_seq_chunk_parent=parent)
assert str(expected) == str(val)
@pytest.mark.parametrize(
"schema,value,expected",
[
(
e3_spliced,
SingleInterval(0, 10, Strand.PLUS),
dict(interval_starts=(2, 7), interval_ends=(6, 10), strand=Strand.PLUS.name),
),
(
e3_spliced_minus,
SingleInterval(0, 10, Strand.PLUS),
dict(interval_starts=(2, 7), interval_ends=(6, 10), strand=Strand.MINUS.name),
),
],
)
def test_intersection_no_parent(self, schema, value, expected):
feat = schema.to_feature_interval()
val = feat.intersect(value)
expected = FeatureIntervalModel.Schema().load(expected).to_feature_interval()
assert str(expected) == str(val)
@pytest.mark.parametrize(
"schema,parent,expected_spliced",
[
(e3_spliced, parent, "ATTCTGGCTA"),
(e3_spliced, parent_genome2, "GTATCTTACC"),
(
# explicit strand on Parent has no effect
e3_spliced,
parent_genome2_minus,
"GTATCTTACC",
),
(e3_spliced_minus, parent, "TAGCCAGAAT"),
(e3_spliced_minus, parent_genome2, "GGTAAGATAC"),
(
# explicit strand on Parent has no effect
e3_spliced_minus,
parent_genome2_minus,
"GGTAAGATAC",
),
],
)
def test_spliced_sequence(self, schema, parent, expected_spliced):
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
assert str(feat.get_spliced_sequence()) == expected_spliced
@pytest.mark.parametrize(
"schema,parent,expected_genomic,expected_stranded_genomic",
[
# positive strand, so genomic == genomic_stranded
(e3_spliced, parent, "ATTCTTGGACCTA", "ATTCTTGGACCTA"),
(e3_spliced, parent_genome2, "GTATTCTTGGACC", "GTATTCTTGGACC"),
(
# explicit strand on Parent has no effect
e3_spliced,
parent_genome2_minus,
"GTATTCTTGGACC",
"GTATTCTTGGACC",
),
(e3_spliced_minus, parent, "ATTCTTGGACCTA", "TAGGTCCAAGAAT"),
(e3_spliced_minus, parent_genome2, "GTATTCTTGGACC", "GGTCCAAGAATAC"),
(
# explicit strand on Parent flips it
e3_spliced_minus,
parent_genome2_minus,
"GTATTCTTGGACC",
"GGTCCAAGAATAC",
),
],
)
def test_genomic_sequence(self, schema, parent, expected_genomic, expected_stranded_genomic):
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
assert str(feat.get_reference_sequence()) == expected_genomic
assert str(feat.get_genomic_sequence()) == expected_stranded_genomic
@pytest.mark.parametrize(
"schema,expected",
[(e3_spliced, "GTATCTTACC"), (e3_spliced_minus, "GGTAAGATAC")],
)
def test_reset_parent(self, schema, expected):
feat = schema.to_feature_interval(parent_or_seq_chunk_parent=parent)
feat._reset_parent(parent_genome2)
assert str(feat.get_spliced_sequence()) == str(expected)
def test_object_conversion(self):
for model in [se_unspliced, e3_spliced_minus, e3_spliced]:
obj = model.to_feature_interval()
new_model = FeatureIntervalModel.from_feature_interval(obj)
new_model.feature_interval_guid = None
assert model == new_model
obj = model.to_feature_interval(parent_or_seq_chunk_parent=parent)
new_model = FeatureIntervalModel.from_feature_interval(obj)
new_model.feature_interval_guid = None
assert model == new_model
def test_identifiers(self):
feat = se_unspliced.to_feature_interval()
feat.feature_name = "test"
feat.feature_id = "testid"
assert feat.identifiers == {"test", "testid"}
assert feat.identifiers_dict == {"feature_name": "test", "feature_id": "testid"}
def test_intersection_exception(self):
schema = FeatureIntervalModel.Schema().load(
dict(interval_starts=[10], interval_ends=[15], strand=Strand.MINUS.name)
)
feat = schema.to_feature_interval()
s = SingleInterval(0, 5, Strand.PLUS)
with pytest.raises(EmptyLocationException):
_ = feat.intersect(s)
def test_gff_export_exceptions(self):
feat = se_unspliced.to_feature_interval(parent_or_seq_chunk_parent=parent)
with pytest.raises(GFF3MissingSequenceNameError):
_ = "\n".join(str(x) for x in feat.to_gff())
feat.sequence_name = "myseq"
with pytest.raises(NoSuchAncestorException):
_ = "\n".join(str(x) for x in feat.to_gff(chromosome_relative_coordinates=False))
def test_gff_export(self):
feat = se_unspliced.to_feature_interval(parent_or_seq_chunk_parent=parent)
feat.sequence_name = "myseq"
assert (
"\n".join(str(x) for x in feat.to_gff())
== "myseq\tBioCantor\tfeature_interval\t1\t18\t.\t+\t.\tID=6940c467-070a-3524-2dcb-a478a6fa0388\n"
"myseq\tBioCantor\tsubregion\t1\t18\t.\t+\t.\tID=feature-6940c467-070a-3524-2dcb-a478a6fa0388-1;"
"Parent=6940c467-070a-3524-2dcb-a478a6fa0388"
)
def test_gff_export_subset(self):
feat = e3_spliced.to_feature_interval(parent_or_seq_chunk_parent=parent_genome2_1_15)
feat.sequence_name = "myseq"
assert (
"\n".join(str(x) for x in feat.to_gff())
== "myseq\tBioCantor\tfeature_interval\t3\t15\t.\t+\t.\tID=c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80\n"
"myseq\tBioCantor\tsubregion\t3\t6\t.\t+\t.\tID=feature-c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80-1;"
"Parent=c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80\n"
"myseq\tBioCantor\tsubregion\t8\t10\t.\t+\t.\tID=feature-c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80-2;"
"Parent=c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80\n"
"myseq\tBioCantor\tsubregion\t13\t15\t.\t+\t.\tID=feature-c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80-3;"
"Parent=c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80"
)
assert (
"\n".join(str(x) for x in feat.to_gff(chromosome_relative_coordinates=False))
== "myseq\tBioCantor\tfeature_interval\t2\t14\t.\t+\t.\tID=c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80\n"
"myseq\tBioCantor\tsubregion\t2\t5\t.\t+\t.\tID=feature-c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80-1;"
"Parent=c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80\n"
"myseq\tBioCantor\tsubregion\t7\t9\t.\t+\t.\tID=feature-c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80-2;"
"Parent=c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80\n"
"myseq\tBioCantor\tsubregion\t12\t14\t.\t+\t.\tID=feature-c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80-3;"
"Parent=c45e8d7b-cbd6-43b2-bb08-429d9cb7fe80"
)
@pytest.mark.parametrize(
"feature,parent,expected_span",
[
(e3_spliced, None, SingleInterval(2, 15, Strand.PLUS)),
(e3_spliced_minus, None, SingleInterval(2, 15, Strand.MINUS)),
(se_unspliced, None, SingleInterval(0, 18, Strand.PLUS)),
(se_unspliced, None, SingleInterval(0, 18, Strand.PLUS)),
(e3_spliced, parent_genome2, SingleInterval(2, 15, Strand.PLUS, parent=parent_genome2)),
(e3_spliced_minus, parent_genome2, SingleInterval(2, 15, Strand.MINUS, parent=parent_genome2)),
(se_unspliced, parent_genome2, SingleInterval(0, 18, Strand.PLUS, parent=parent_genome2)),
(se_unspliced, parent_genome2, SingleInterval(0, 18, Strand.PLUS, parent=parent_genome2)),
(
e3_spliced,
parent_genome2_1_15,
SingleInterval(
2, 15, Strand.PLUS, parent=parent_genome2_1_15.first_ancestor_of_type(SequenceType.CHROMOSOME)
),
),
(
e3_spliced_minus,
parent_genome2_1_15,
SingleInterval(
2, 15, Strand.MINUS, parent=parent_genome2_1_15.first_ancestor_of_type(SequenceType.CHROMOSOME)
),
),
(
e3_spliced,
parent_genome2_2_8,
SingleInterval(
2, 15, Strand.PLUS, parent=parent_genome2_2_8.first_ancestor_of_type(SequenceType.CHROMOSOME)
),
),
(
e3_spliced_minus,
parent_genome2_2_8,
SingleInterval(
2, 15, Strand.MINUS, parent=parent_genome2_2_8.first_ancestor_of_type(SequenceType.CHROMOSOME)
),
),
],
)
def test_chromosome_span(self, feature, parent, expected_span):
feat = feature.to_feature_interval(parent)
assert feat.chromosome_span == expected_span
@pytest.mark.parametrize(
"feature,parent,expected_gaps",
[
(e3_spliced, None, CompoundInterval([6, 10], [7, 12], Strand.PLUS)),
(e3_spliced_minus, None, CompoundInterval([6, 10], [7, 12], Strand.MINUS)),
(se_unspliced, None, EmptyLocation()),
(se_unspliced, None, EmptyLocation()),
(e3_spliced, parent_genome2, CompoundInterval([6, 10], [7, 12], Strand.PLUS, parent=parent_genome2)),
(e3_spliced_minus, parent_genome2, CompoundInterval([6, 10], [7, 12], Strand.MINUS, parent=parent_genome2)),
(se_unspliced, parent_genome2, EmptyLocation()),
(se_unspliced, parent_genome2, EmptyLocation()),
(
e3_spliced,
parent_genome2_1_15,
CompoundInterval(
[6, 10],
[7, 12],
Strand.PLUS,
parent=parent_genome2_1_15.first_ancestor_of_type(SequenceType.CHROMOSOME),
),
),
(
e3_spliced_minus,
parent_genome2_1_15,
CompoundInterval(
[6, 10],
[7, 12],
Strand.MINUS,
parent=parent_genome2_1_15.first_ancestor_of_type(SequenceType.CHROMOSOME),
),
),
(se_unspliced, parent_genome2_1_15, EmptyLocation()),
(se_unspliced, parent_genome2_1_15, EmptyLocation()),
(
e3_spliced,
parent_genome2_2_8,
CompoundInterval(
[6, 10],
[7, 12],
Strand.PLUS,
parent=parent_genome2_1_15.first_ancestor_of_type(SequenceType.CHROMOSOME),
),
),
(
e3_spliced_minus,
parent_genome2_2_8,
CompoundInterval(
[6, 10],
[7, 12],
Strand.MINUS,
parent=parent_genome2_1_15.first_ancestor_of_type(SequenceType.CHROMOSOME),
),
),
],
)
def test_chromosome_gaps_location(self, feature, parent, expected_gaps):
feat = feature.to_feature_interval(parent)
assert feat.chromosome_gaps_location == expected_gaps
id="genome2_minus",
sequence=Sequence(genome2, Alphabet.NT_STRICT),
strand=Strand.MINUS,
sequence_type=SequenceType.CHROMOSOME,
)
# slice the genome down to contain some of the transcripts
parent_genome2_1_15 = Parent(
id="genome2_1_15",
sequence=Sequence(
genome2[1:15],
Alphabet.NT_EXTENDED_GAPPED,
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(
location=SingleInterval(
1, 15, Strand.PLUS, parent=Parent(id="genome2", sequence_type=SequenceType.CHROMOSOME)
)
),
),
)
parent_genome2_2_8 = Parent(
id="genome2_2_8",
sequence=Sequence(
genome2[2:8],
Alphabet.NT_EXTENDED_GAPPED,
type=SequenceType.SEQUENCE_CHUNK,
parent=Parent(
location=SingleInterval(
2, 8, Strand.PLUS, parent=Parent(id="genome2", sequence_type=SequenceType.CHROMOSOME)
)
def test_location(self):
assert Parent(location=SingleInterval(0, 1, Strand.PLUS)).location == SingleInterval(0, 1, Strand.PLUS)
class TestParent:
@pytest.mark.parametrize(
"id,sequence_type,strand,location,sequence,expected",
[
(None, None, None, None, None, Parent()),
(
"id",
"seqtype",
Strand.MINUS,
SingleInterval(
0,
1,
Strand.MINUS,
Parent(
id="id",
sequence_type="seqtype",
strand=Strand.MINUS,
sequence=Sequence(
"AAA",
Alphabet.NT_STRICT,
id="id",
type="seqtype",
parent=Parent(
id="id2",
sequence=Sequence("CCC", Alphabet.NT_STRICT, id="id2"),
),
),
),
),
Sequence(
"AAA",
Alphabet.NT_STRICT,
id="id",
type="seqtype",
parent=Parent(id="id2", sequence=Sequence("CCC", Alphabet.NT_STRICT, id="id2")),
),
Parent(
id="id",
sequence_type="seqtype",
strand=Strand.MINUS,
location=SingleInterval(
0,
1,
Strand.MINUS,
Parent(
id="id",
sequence_type="seqtype",
strand=Strand.MINUS,
sequence=Sequence(
"AAA",
Alphabet.NT_STRICT,
id="id",
type="seqtype",
parent=Parent(
id="id2",
sequence=Sequence("CCC", Alphabet.NT_STRICT, id="id2"),
),
),
),
),
sequence=Sequence(
"AAA",
Alphabet.NT_STRICT,
id="id",
type="seqtype",
parent=Parent(
id="id2",
sequence=Sequence("CCC", Alphabet.NT_STRICT, id="id2"),
),
),
),
),
],
)
def test_init(self, id, sequence_type, strand, location, sequence, expected):
assert expected == Parent(
id=id,
sequence_type=sequence_type,
strand=strand,
location=location,
sequence=sequence,
)
@pytest.mark.parametrize(
"id,sequence_type,strand,location,sequence,parent,expected_exception",
[
("id1", None, None, SingleInterval(0, 5, Strand.PLUS, parent="id2"), None, None, ParentException),
("id1", None, None, None, Sequence("AAA", Alphabet.NT_STRICT, id="id2"), None, ParentException),
(
None,
None,
None,
SingleInterval(0, 5, Strand.PLUS, parent="id1"),
Sequence("AAC", Alphabet.NT_STRICT, id="id2"),
None,
ParentException,
),
(
None,
"seqtype",
None,
SingleInterval(0, 5, Strand.PLUS, parent=Parent(sequence_type="unknown")),
None,
None,
ParentException,
),
(None, "seqtype", None, None, Sequence("AAT", Alphabet.NT_STRICT, type="unknown"), None, ParentException),
(
None,
None,
None,
SingleInterval(0, 5, Strand.PLUS, parent=Parent(sequence_type="unknown")),
Sequence("AAG", Alphabet.NT_STRICT, type="seqtype"),
None,
ParentException,
),
(None, None, Strand.MINUS, SingleInterval(0, 5, Strand.PLUS), None, None, InvalidStrandException),
(
None,
None,
None,
SingleInterval(0, 10, Strand.PLUS),
Sequence("A", Alphabet.NT_STRICT),
None,
InvalidPositionException,
),
(
None,
None,
None,
None,
Sequence("AA", Alphabet.NT_STRICT),
Parent(sequence=Sequence("A", Alphabet.NT_STRICT)),
LocationException,
),
(None, None, Strand.PLUS, SingleInterval(5, 10, Strand.MINUS), None, None, InvalidStrandException),
(
None,
None,
None,
None,
Sequence("AA", Alphabet.NT_STRICT, parent="id1"),
Parent(id="id2"),
MismatchedParentException,
),
],
)
def test_init_error(self, id, sequence_type, strand, location, sequence, parent, expected_exception):
with pytest.raises(expected_exception):
Parent(
id=id,
sequence_type=sequence_type,
strand=strand,
location=location,
sequence=sequence,
parent=parent,
)
@pytest.mark.parametrize(
"obj,expected",
[
(
Sequence("AAA", Alphabet.NT_STRICT),
Parent(sequence=Sequence("AAA", Alphabet.NT_STRICT)),
),
("parent", Parent(id="parent")),
(
SingleInterval(5, 10, Strand.PLUS),
Parent(location=SingleInterval(5, 10, Strand.PLUS)),
),
(
CompoundInterval([5], [10], Strand.PLUS),
Parent(location=CompoundInterval([5], [10], Strand.PLUS)),
),
(EmptyLocation(), Parent(location=EmptyLocation())),
(Strand.MINUS, Parent(strand=Strand.MINUS)),
(
Parent(
id="parent",
sequence_type="chr",
strand=Strand.MINUS,
parent=Parent(id="grandparent"),
),
Parent(
id="parent",
sequence_type="chr",
strand=Strand.MINUS,
parent=Parent(id="grandparent"),
),
),
],
)
def test_make_parent(self, obj, expected):
assert make_parent(obj) == expected
@pytest.mark.parametrize(
"parent1,parent2,expected",
[
(Parent(), Parent(), True),
(Parent(), Parent(id=None, sequence_type=None), True),
(Parent(id="id1"), Parent(id="id2"), False),
(
Parent(sequence_type=None),
Parent(sequence_type="unknown"),
False,
),
(Parent(strand=Strand.UNSTRANDED), Parent(strand=Strand.MINUS), False),
(
Parent(location=SingleInterval(0, 5, Strand.PLUS, parent="id1")),
Parent(location=SingleInterval(0, 5, Strand.PLUS, parent="id2")),
False,
),
(
Parent(sequence=Sequence("A", Alphabet.NT_STRICT)),
Parent(sequence=Sequence("A", Alphabet.NT_STRICT, parent=Parent(id="parent"))),
False,
),
(
Parent(parent="parent1"),
Parent(parent="parent2"),
False,
),
],
)
def test_eq(self, parent1, parent2, expected):
assert (parent1 == parent2) is expected
@pytest.mark.parametrize(
"parent1,parent2,expected",
[
(Parent(), Parent(), True),
(Parent(id="id1"), Parent(id="id2"), False),
(
Parent(sequence_type=None),
Parent(sequence_type="unknown"),
False,
),
(Parent(strand=Strand.UNSTRANDED), Parent(strand=Strand.MINUS), True),
(
Parent(location=SingleInterval(0, 5, Strand.PLUS, parent="id1")),
Parent(location=SingleInterval(0, 5, Strand.PLUS, parent="id2")),
False,
),
(
Parent(sequence=Sequence("A", Alphabet.NT_STRICT)),
Parent(sequence=Sequence("A", Alphabet.NT_STRICT, parent="parent")),
False,
),
(
Parent(parent="parent1"),
Parent(parent="parent2"),
False,
),
],
)
def test_equals_except_location(self, parent1, parent2, expected):
assert parent1.equals_except_location(parent2) is expected
@pytest.mark.parametrize(
"id,location,sequence,expected",
[
("id", None, None, "id"),
(
None,
SingleInterval(0, 1, Strand.PLUS, parent="id"),
None,
"id",
),
(
None,
None,
Sequence("A", Alphabet.NT_STRICT, id="id", parent="id2"),
"id",
),
(
"id",
SingleInterval(0, 1, Strand.PLUS, parent="id"),
Sequence("A", Alphabet.NT_STRICT, id="id", parent="id2"),
"id",
),
],
)
def test_id(self, id, location, sequence, expected):
assert Parent(id=id, location=location, sequence=sequence).id == expected
@pytest.mark.parametrize(
"sequence_type,location,sequence,expected",
[
("seqtype", None, None, "seqtype"),
(
None,
SingleInterval(
0,
5,
Strand.PLUS,
parent=Parent(sequence_type="seqtype"),
),
None,
"seqtype",
),
(
None,
None,
Sequence("A", Alphabet.NT_STRICT, type="seqtype"),
"seqtype",
),
(
None,
None,
Sequence(
"A",
Alphabet.NT_STRICT,
type="seqtype",
parent=Parent(sequence_type="seqtype_2"),
),
"seqtype",
),
],
)
def test_sequence_type(self, sequence_type, location, sequence, expected):
assert Parent(sequence_type=sequence_type, location=location, sequence=sequence).sequence_type == expected
@pytest.mark.parametrize(
"strand,location,sequence,expected",
[
(Strand.PLUS, None, None, Strand.PLUS),
(None, SingleInterval(0, 5, Strand.MINUS), None, Strand.MINUS),
(
Strand.PLUS,
None,
Sequence("A", Alphabet.NT_STRICT, parent=Strand.MINUS),
Strand.PLUS,
),
],
)
def test_strand(self, strand, location, sequence, expected):
assert Parent(strand=strand, location=location, sequence=sequence).strand == expected
def test_location(self):
assert Parent(location=SingleInterval(0, 1, Strand.PLUS)).location == SingleInterval(0, 1, Strand.PLUS)
def test_sequence(self):
assert Parent(sequence=Sequence("A", Alphabet.NT_STRICT)).sequence == Sequence("A", Alphabet.NT_STRICT)
@pytest.mark.parametrize(
"parent,expected",
[
(Parent(parent="id"), Parent(id="id")),
(
Parent(
sequence=Sequence(
"AA",
Alphabet.NT_STRICT,
parent=Parent(sequence_type="chr"),
)
),
Parent(sequence_type="chr"),
),
],
)
def test_parent(self, parent, expected):
assert parent.parent == expected
@pytest.mark.parametrize(
"parent,expected",
[
(Parent(), Parent()),
(Parent(strand=Strand.PLUS), Parent()),
(
Parent(strand=Strand.PLUS, location=SingleInterval(5, 10, Strand.PLUS)),
Parent(),
),
(
Parent(
id="parent",
sequence_type="unknown",
strand=Strand.PLUS,
location=SingleInterval(0, 1, Strand.PLUS),
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
Parent(
id="parent",
sequence_type="unknown",
sequence=Sequence("AAA", Alphabet.NT_STRICT),
parent="grandparent",
),
),
],
)
def test_strip_location_info(self, parent, expected):
assert parent.strip_location_info() == expected
@pytest.mark.parametrize(
"parent,sequence_type,include_self,expected",
[
(
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
"seqtype",
True,
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
),
(
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
"seqtype",
False,
Parent(id="parent", sequence_type="seqtype"),
),
(
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
parent=Parent(id="grandparent", sequence_type="seqtype_2"),
),
),
"seqtype_2",
True,
Parent(
id="parent",
sequence_type="seqtype_2",
parent=Parent(id="grandparent", sequence_type="seqtype_2"),
),
),
],
)
def test_first_ancestor_of_type(self, parent, sequence_type, include_self, expected):
assert parent.first_ancestor_of_type(sequence_type, include_self=include_self) == expected
@pytest.mark.parametrize(
"parent,sequence_type,include_self",
[
(Parent(id="self"), "seqtype_2", True),
(
Parent(id="self", parent="parent"),
"seqtype_2",
True,
),
(
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
parent=Parent(id="grandparent", sequence_type="seqtype_2"),
),
),
"chr",
True,
),
],
)
def test_first_ancestor_of_type_error(self, parent, sequence_type, include_self):
with pytest.raises(NoSuchAncestorException):
parent.first_ancestor_of_type(sequence_type, include_self=include_self)
@pytest.mark.parametrize(
"parent,sequence_type,include_self,expected",
[
(
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
"seqtype",
True,
True,
),
(
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(id="parent", sequence_type="seqtype"),
),
"seqtype",
False,
True,
),
(
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
parent=Parent(id="grandparent", sequence_type="seqtype_2"),
),
),
"seqtype_2",
True,
True,
),
(
Parent(id="self"),
"seqtype_2",
True,
False,
),
(
Parent(id="self", parent="parent"),
"seqtype_2",
True,
False,
),
(
Parent(
id="self",
sequence_type="seqtype",
parent=Parent(
id="parent",
sequence_type="seqtype_2",
parent=Parent(id="grandparent", sequence_type="seqtype_2"),
),
),
"chr",
True,
False,
),
],
)
def test_has_ancestor_of_type(self, parent, sequence_type, include_self, expected):
assert parent.has_ancestor_of_type(sequence_type, include_self=include_self) is expected
@pytest.mark.parametrize(
"parent,expected",
[
(
Parent(
id="parent",
location=SingleInterval(3, 5, Strand.PLUS),
parent=Parent(id="grandparent", location=SingleInterval(10, 20, Strand.PLUS)),
),
SingleInterval(13, 15, Strand.PLUS, parent="grandparent"),
),
(
Parent(
id="parent",
location=SingleInterval(0, 5, Strand.PLUS),
sequence_type="unknown",
strand=Strand.PLUS,
parent=Parent(
id="grandparent",
location=SingleInterval(100, 200, Strand.MINUS),
),
),
SingleInterval(195, 200, Strand.MINUS, parent="grandparent"),
),
(
Parent(
id="parent",
location=SingleInterval(6, 9, Strand.MINUS),
parent=Parent(
id="grandparent",
location=SingleInterval(0, 10, Strand.PLUS),
sequence_type="chr",
strand=Strand.PLUS,
parent="great grandparent",
),
),
SingleInterval(
6,
9,
Strand.MINUS,
parent=Parent(
id="grandparent",
sequence_type="chr",
parent="great grandparent",
),
),
),
(
Parent(
id="parent",
sequence_type="chr",
strand=Strand.MINUS,
location=SingleInterval(6, 8, Strand.MINUS),
parent=Parent(
id="grandparent",
sequence_type="unknown",
strand=Strand.MINUS,
location=SingleInterval(5, 15, Strand.MINUS),
parent="great grandparent",
),
),
SingleInterval(
7,
9,
Strand.PLUS,
parent=Parent(
id="grandparent",
sequence_type="unknown",
parent="great grandparent",
),
),
),
(
Parent(
id="parent",
location=SingleInterval(3, 5, Strand.UNSTRANDED),
parent=Parent(id="grandparent", location=SingleInterval(10, 20, Strand.PLUS)),
),
SingleInterval(13, 15, Strand.UNSTRANDED, parent="grandparent"),
),
(
Parent(
id="parent",
location=SingleInterval(3, 5, Strand.UNSTRANDED),
parent=Parent(id="grandparent", location=SingleInterval(10, 20, Strand.MINUS)),
),
SingleInterval(15, 17, Strand.UNSTRANDED, parent="grandparent"),
),
],
)
def test_lift_child_location_contiguous_to_parent_single_interval(self, parent, expected):
assert parent.lift_child_location_to_parent() == expected
@pytest.mark.parametrize(
"parent,expected",
[
(
Parent(
id="parent",
location=CompoundInterval([3, 7], [5, 10], Strand.PLUS),
parent=Parent(id="grandparent", location=SingleInterval(10, 20, Strand.PLUS)),
),
CompoundInterval([13, 17], [15, 20], Strand.PLUS, parent="grandparent"),
),
(
Parent(
id="parent",
location=CompoundInterval([0, 10], [5, 15], Strand.PLUS),
sequence_type="unknown",
strand=Strand.PLUS,
parent=Parent(
id="grandparent",
location=SingleInterval(100, 200, Strand.MINUS),
),
),
CompoundInterval(
[185, 195],
[190, 200],
Strand.MINUS,
parent="grandparent",
),
),
(
Parent(
id="parent",
location=CompoundInterval([6], [9], Strand.MINUS),
parent=Parent(
id="grandparent",
location=SingleInterval(0, 10, Strand.PLUS),
sequence_type="chr",
strand=Strand.PLUS,
parent="great grandparent",
),
),
SingleInterval(
6,
9,
Strand.MINUS,
parent=Parent(
id="grandparent",
sequence_type="chr",
parent="great grandparent",
),
),
),
(
Parent(
id="parent",
sequence_type="chr",
strand=Strand.MINUS,
location=CompoundInterval([6], [8], Strand.MINUS),
parent=Parent(
id="grandparent",
sequence_type="unknown",
strand=Strand.MINUS,
location=SingleInterval(5, 15, Strand.MINUS),
parent="great grandparent",
),
),
SingleInterval(
7,
9,
Strand.PLUS,
parent=Parent(
id="grandparent",
sequence_type="unknown",
parent="great grandparent",
),
),
),
(
Parent(
id="parent",
location=CompoundInterval([3, 7], [5, 10], Strand.UNSTRANDED),
parent=Parent(id="grandparent", location=SingleInterval(10, 20, Strand.PLUS)),
),
CompoundInterval(
[13, 17],
[15, 20],
Strand.UNSTRANDED,
parent="grandparent",
),
),
(
Parent(
id="parent",
location=CompoundInterval([3], [5], Strand.UNSTRANDED),
parent=Parent(id="grandparent", location=SingleInterval(10, 20, Strand.MINUS)),
),
SingleInterval(15, 17, Strand.UNSTRANDED, parent="grandparent"),
),
],
)
def test_lift_child_location_discontiguous_to_parent_single_interval(self, parent, expected):
assert parent.lift_child_location_to_parent() == expected
@pytest.mark.parametrize(
"parent,expected_error",
[
# No location
(
Parent(parent=SingleInterval(5, 10, Strand.PLUS)),
NullParentException,
),
# Parent has no location
(
Parent(
location=SingleInterval(5, 10, Strand.PLUS),
parent="grandparent",
),
NullParentException,
),
# Location on parent can't be unstranded
(
Parent(
location=SingleInterval(5, 10, Strand.PLUS),
parent=Parent(
id="grandparent",
location=SingleInterval(0, 100, Strand.UNSTRANDED),
),
),
InvalidStrandException,
),
# Location must fit inside location on parent
(
Parent(
location=SingleInterval(5, 10, Strand.PLUS),
parent=Parent(id="grandparent", location=SingleInterval(30, 31, Strand.PLUS)),
),
ValueError,
),
],
)
def test_lift_child_location_to_parent_single_interval_error(self, parent, expected_error):
with pytest.raises(expected_error):
parent.lift_child_location_to_parent()
@pytest.mark.parametrize(
"parent,expected",
[
# Location takes up entire parent location
(
Parent(
id="parent",
location=SingleInterval(0, 10, Strand.PLUS),
parent=Parent(
id="grandparent",
location=CompoundInterval([5, 20], [10, 25], Strand.PLUS),
),
),
CompoundInterval([5, 20], [10, 25], Strand.PLUS, parent="grandparent"),
),
# Location (unstranded) takes up part of parent location (minus)
(
Parent(
id="parent",
location=SingleInterval(10, 20, Strand.UNSTRANDED),
parent=Parent(
id="grandparent",
location=CompoundInterval([10, 20, 30], [18, 28, 38], Strand.MINUS),
),
),
CompoundInterval(
[14, 20],
[18, 26],
Strand.UNSTRANDED,
parent="grandparent",
),
),
# Location (minus) takes up one block of parent location (plus); location is at end of sequence
(
Parent(
id="parent",
location=SingleInterval(5, 10, Strand.MINUS),
parent=Parent(
id="grandparent",
location=CompoundInterval([30, 40], [35, 45], Strand.PLUS),
),
),
SingleInterval(40, 45, Strand.MINUS, parent="grandparent"),
),
# Location (minus) takes up part of one block of parent location (minus)
(
Parent(
id="parent",
location=SingleInterval(0, 4, Strand.MINUS),
parent=Parent(
id="grandparent",
location=CompoundInterval([30, 40], [35, 45], Strand.MINUS),
),
),
SingleInterval(41, 45, Strand.PLUS, parent="grandparent"),
),
],
)
def test_lift_child_location_contiguous_to_parent_compound_interval(self, parent, expected):
assert parent.lift_child_location_to_parent() == expected
@pytest.mark.parametrize(
"parent,expected",
[
# Location takes up entire parent location
(
Parent(
id="parent",
location=CompoundInterval([0, 5], [5, 10], Strand.PLUS),
parent=Parent(
id="grandparent",
location=CompoundInterval([5, 20], [10, 25], Strand.PLUS),
),
),
CompoundInterval([5, 20], [10, 25], Strand.PLUS, parent="grandparent"),
),
# Location (unstranded) takes up part of parent location (minus)
(
Parent(
id="parent",
location=CompoundInterval([10, 22], [20, 23], Strand.UNSTRANDED),
parent=Parent(
id="grandparent",
location=CompoundInterval([10, 20, 30], [18, 28, 38], Strand.MINUS),
),
),
CompoundInterval(
[11, 14, 20],
[12, 18, 26],
Strand.UNSTRANDED,
parent="grandparent",
),
),
# Location (minus) takes up one block of parent location (plus); location is at end of sequence
(
Parent(
id="parent",
location=CompoundInterval([5], [10], Strand.MINUS),
parent=Parent(
id="grandparent",
location=CompoundInterval([30, 40], [35, 45], Strand.PLUS),
),
),
SingleInterval(40, 45, Strand.MINUS, parent="grandparent"),
),
# Location (minus) takes up part of one block of parent location (minus)
(
Parent(
id="parent",
location=CompoundInterval([0, 3], [1, 4], Strand.MINUS),
parent=Parent(
id="grandparent",
location=CompoundInterval([30, 40], [35, 45], Strand.MINUS),
),
),
CompoundInterval([41, 44], [42, 45], Strand.PLUS, parent="grandparent"),
),
],
)
def test_lift_child_location_discontiguous_to_parent_compound_interval(self, parent, expected):
assert parent.lift_child_location_to_parent() == expected
@pytest.mark.parametrize(
"parent,expected_error",
[
# Location must fit inside location on parent
(
Parent(
location=SingleInterval(5, 50, Strand.PLUS),
parent=Parent(
id="grandparent",
location=CompoundInterval([10, 20], [15, 25], Strand.PLUS),
),
),
InvalidPositionException,
),
],
)
def test_lift_child_location_to_parent_compound_interval_error(self, parent, expected_error):
with pytest.raises(expected_error):
parent.lift_child_location_to_parent()
@pytest.mark.parametrize(
"parent,location,expected",
[
(
Parent(),
SingleInterval(5, 10, Strand.PLUS),
Parent(location=SingleInterval(5, 10, Strand.PLUS), strand=Strand.PLUS),
),
(
Parent(
id="parent",
sequence_type="unknown",
strand=Strand.MINUS,
location=SingleInterval(0, 2, Strand.MINUS),
sequence=Sequence("AAA", Alphabet.NT_STRICT),
),
SingleInterval(2, 3, Strand.PLUS),
Parent(
id="parent",
sequence_type="unknown",
strand=Strand.PLUS,
location=SingleInterval(2, 3, Strand.PLUS),
sequence=Sequence("AAA", Alphabet.NT_STRICT),
),
),
(
Parent(
id="parent",
sequence_type="unknown",
strand=Strand.MINUS,
location=SingleInterval(0, 2, Strand.MINUS),
sequence=Sequence("AAA", Alphabet.NT_STRICT),
),
None,
Parent(
id="parent",
sequence_type="unknown",
sequence=Sequence("AAA", Alphabet.NT_STRICT),
),
),
],
)
def test_reset_location(self, parent, location, expected):
assert parent.reset_location(location) == expected
@pytest.mark.parametrize(
"parent,location,expected_exception",
[
(
Parent(sequence=Sequence("AAA", Alphabet.NT_STRICT)),
SingleInterval(0, 5, Strand.PLUS),
InvalidPositionException,
),
(
Parent(id="id1", sequence=Sequence("AAA", Alphabet.NT_STRICT)),
SingleInterval(
0,
1,
Strand.PLUS,
parent=Parent(id="id2", sequence=Sequence("AAA", Alphabet.NT_STRICT)),
),
ParentException,
),
],
)
def test_reset_location_error(self, parent, location, expected_exception):
with pytest.raises(expected_exception):
parent.reset_location(location)
@pytest.mark.parametrize(
"parent,sequence,include_self,expected",
[
(Parent(), Sequence("AA", Alphabet.NT_STRICT), True, False),
(Parent(), Sequence("AA", Alphabet.NT_STRICT), False, False),
(
Parent(sequence=Sequence("AA", Alphabet.NT_STRICT)),
Sequence("AA", Alphabet.NT_STRICT),
True,
True,
),
(
Parent(sequence=Sequence("AA", Alphabet.NT_STRICT)),
Sequence("AA", Alphabet.NT_STRICT),
False,
False,
),
(
Parent(
sequence=Sequence("AA", Alphabet.NT_STRICT),
parent=Sequence("AA", Alphabet.NT_STRICT),
),
Sequence("AA", Alphabet.NT_STRICT),
False,
True,
),
(
Parent(
sequence=Sequence("AA", Alphabet.NT_STRICT),
parent=Sequence("AAT", Alphabet.NT_STRICT),
),
Sequence("AAT", Alphabet.NT_STRICT),
False,
True,
),
(
Parent(
sequence=Sequence("AA", Alphabet.NT_STRICT),
parent=Sequence("AAT", Alphabet.NT_STRICT),
),
Sequence("AAT", Alphabet.NT_STRICT, id="id"),
True,
False,
),
(
Parent(
parent=Parent(parent=Parent(parent=Parent(sequence=Sequence("AAA", Alphabet.NT_STRICT, id="seq"))))
),
Sequence("AAA", Alphabet.NT_STRICT, id="seq"),
True,
True,
),
],
)
def test_has_ancestor_sequence(self, parent, sequence, include_self, expected):
assert parent.has_ancestor_sequence(sequence, include_self) == expected
def test_require_parent_has_location(self):
with pytest.raises(NullParentException):
ObjectValidation.require_parent_has_location(Parent(id="parent"))
ObjectValidation.require_parent_has_location(
Parent(location=SingleInterval(5, 6, Strand.PLUS)))