def test_reference_sequence_repository_register_genomic_ranges():
rsr = ReferenceSequenceRepository()
grs = [
GenomicRange('X', 1, 100, '+'),
GenomicRange('X', 500, 800, '+')
]
rsr.register_genomic_ranges(grs)
assert rsr.region_count == len(grs)
def test_genomic_range_contains(a_range, b_range, b_contains_a):
chromosome = 'X'
strand = '+'
a = GenomicRange(chromosome, *a_range, strand)
b = GenomicRange(chromosome, *b_range, strand)
assert a in a
assert (a in b if b_contains_a else a not in b)
assert b not in a
def test_reference_sequence_get_subsequence():
seq = 'AAACCC'
gr = GenomicRange('X', 100, 105, '+')
gr_sub = GenomicRange('X', 100, 102, '+')
# Initialise reference sequence
ref_seq = ReferenceSequence(seq, gr)
# Extract reference subsequence
sub_ref_seq = ref_seq.get_subsequence(gr_sub)
# Check reference subsequence
assert sub_ref_seq.genomic_range == gr_sub
assert sub_ref_seq.sequence == 'AAA'
def test_oligo_compute_mutations(targetons, mutator, pam_protection):
def get_segment(seq):
return TargetonOligoSegment(get_targeton(seq, pam_protection),
{mutator})
ct = CodonTable.load(get_data_file_path(CODON_TABLE_FP))
ref_seq = ''.join(targetons)
pam_ref_seq = PamProtectedReferenceSequence(
ref_seq, GenomicRange('X', 1, sum(len(seq) for seq in targetons), '+'),
ref_seq)
adaptor_5 = 'AAAAAA'
adaptor_3 = 'AAAAAA'
segments = list(map(get_segment, targetons))
ot = OligoTemplate(TRANSCRIPT_INFO, pam_ref_seq, set(), set(), adaptor_5,
adaptor_3, segments)
for _, target_segment in ot.target_segments:
mutation_collections = target_segment.compute_mutations(ct)
mutation_collection = mutation_collections[mutator]
if pam_protection:
if mutator == TargetonMutator.DEL1:
assert all(
set(m.sequence) == {DUMMY_PAM_PROTECTION_NT}
for m in mutation_collection.mutations)
elif mutator == TargetonMutator.SNV:
assert all(
set(m.sequence) - {m.new} == {DUMMY_PAM_PROTECTION_NT}
for m in mutation_collection.mutations)
def test_reference_sequence_init(seq, start, end, valid):
gr = GenomicRange('X', start, end, '+')
with pytest.raises(ValueError) if not valid else nullcontext():
# Initialise reference sequence
ReferenceSequence(seq, gr)
def test_utr_repository_get_transcript_infos():
utr = UTRRepository(UTR_RANGES)
d = utr.get_transcript_infos(REF_RANGES)
assert len(d) == 1
gr = GenomicRange('chrX', 800, 1200, '+')
ti = TranscriptInfo('G1', 'T1')
assert d[gr] == ti
def test_cds_context_repository_compute_cds_contexts(start, end, exp_ext_5, exp_ext_3):
cds_ranges = PyRanges(df=CDS_RANGES_DF)
chromosome = 'X'
strand = '+'
gr = GenomicRange(chromosome, start, end, strand)
target_ranges = PyRanges(df=pd.DataFrame.from_records([
gr.as_pyrange()
], columns=PYRANGES_FIELDS))
target_ranges.is_const = False
# Initialise repository
ccr = CDSContextRepository(cds_ranges)
ccr.register_target_ranges(target_ranges)
# Compute CDS contexts
ccr.compute_cds_contexts()
# Check CDS contexts
assert len(ccr._target_cds_contexts) == 1
exon_info, (ext_5, ext_3) = ccr._target_cds_contexts[gr]
# Check exon information
assert isinstance(exon_info, ExonInfo)
assert exon_info.gene_id == GID
assert exon_info.transcript_id == TID
# Check CDS extension
if exp_ext_5 is not None:
assert ext_5 == GenomicRange(chromosome, *exp_ext_5, strand)
else:
assert ext_5 is None
if exp_ext_3 is not None:
assert ext_3 == GenomicRange(chromosome, *exp_ext_3, strand)
else:
assert ext_3 is None
# Check information retrieval
assert ccr.get_cds_extensions(gr) == (ext_5, ext_3)
assert ccr.get_exon_info(gr) == exon_info
assert ccr.get_transcript_info(gr) == exon_info.transcript_info
def test_reference_sequence_repository_get_genomic_range_subsequence():
genomic_range = GenomicRange('X', 1023, 1032, '+')
a = 'AAAAA'
b = 'GGGGG'
sequence = a + b
rsr = ReferenceSequenceRepository()
rsr.register_genomic_range(genomic_range)
rsr._sequences[genomic_range.chromosome][(genomic_range.start, genomic_range.end)] = sequence
assert rsr.get_genomic_range_subsequence(genomic_range, 1023, 1027) == a
assert rsr.get_genomic_range_subsequence(genomic_range, 1028, 1032) == b
assert a + b == sequence
def test_compute_pam_protected_sequence(seq, pos, ref, alt, ppseq, valid):
chromosome = 'X'
start = 100
end = start + len(seq) - 1
variant = PamVariant(GenomicPosition(chromosome, pos), ref, alt)
gr = GenomicRange(chromosome, start, end, '+')
ref_seq = ReferenceSequence(seq, gr)
with pytest.raises(Exception) if not valid else nullcontext():
pam_ref_seq = compute_pam_protected_sequence(ref_seq, {variant})
if valid:
assert pam_ref_seq.sequence == ref_seq.sequence
assert pam_ref_seq.pam_protected_sequence == ppseq
def test_snv_table_get_snvs_triplet(triplet, alt_triplets, strand,
mut_types_plus, mut_types_minus):
gr = GenomicRange('X', 10, 12, strand)
tr = codon_table.translate if strand == '+' else codon_table.translate_rc
mut_types = mut_types_plus if strand == '+' else mut_types_minus
# Retrieve SNV metadata
snv_meta = snv_table.get_snvs(strand, triplet, gr, 0, 0, reset_index=True)
snv_meta = snv_meta.sort_values(['pos', 'alt']).reset_index(drop=True)
# Check amino acid change
ref_aas = snv_meta.ref_aa.unique().astype('string')
assert len(ref_aas) == 1
assert ref_aas[0] == tr(triplet)
assert np.array_equal(snv_meta.alt_aa, np.array(list(map(tr,
alt_triplets))))
assert np.array_equal(snv_meta.mut_type, mut_types)
def test_reference_sequence_repository_register_get_sequence():
chromosome = 'X'
start = 1
end = 100
seq = 'AACCGGTT'
gr = GenomicRange(chromosome, start, end, '+')
# Initialise repository
rsr = ReferenceSequenceRepository()
rsr.register_genomic_range(gr)
# Register sequence
rsr.register_sequence(chromosome, start, end, seq)
# Retrieve sequence
assert rsr.get_sequence(chromosome, start, end) == seq
assert rsr.get_genomic_range_sequence(gr) == seq
def test_get_inframe_mutations(seq, pre, suf, strand, exp_pos, exp_ref,
exp_mseq):
# Generate target
gr = GenomicRange('X', 10, 10 + len(seq) - 1, strand)
t = CDSTargeton(PamProtectedReferenceSequence(seq, gr, seq), pre, suf)
# Generate in-frame deletions
mc = t.get_inframe_mutations()
# Check metadata table
assert np.array_equal(mc.df.mut_position, np.array(exp_pos))
assert np.array_equal(
mc.df.ref.astype('string').to_numpy(), np.array(exp_ref))
assert np.array_equal(
mc.df.mseq.astype('string').to_numpy(), np.array(exp_mseq))
assert mc.df.new.isna().all()
assert mc.df.var_type.unique()[0] == del_var_type
def test_get_2del_mutations(offset, seq, exp_pos, exp_ref, exp_mseq, cds):
# Generate target
gr = GenomicRange('X', 10, 10 + len(seq) - 1, '+')
t = (CDSTargeton(PamProtectedReferenceSequence(seq, gr, seq), 'AA', 'A')
if cds else Targeton(PamProtectedReferenceSequence(seq, gr, seq)))
# Generate in-frame deletions
mc = getattr(t, del_offset_method[offset])()
# Check metadata table
assert np.array_equal(mc.df.mut_position, np.array(exp_pos))
assert np.array_equal(
mc.df.ref.astype('string').to_numpy(), np.array(exp_ref))
assert np.array_equal(
mc.df.mseq.astype('string').to_numpy(), np.array(exp_mseq))
assert mc.df.new.isna().all()
assert mc.df.var_type.unique()[0] == del_var_type
def test_get_snvre_aa_mutations(aa, strand, seq, exp_mseq):
# Generate target
gr = GenomicRange('X', 10, 10 + len(seq) - 1, strand)
t = CDSTargeton(PamProtectedReferenceSequence(seq, gr, seq), '', '')
# Generate codon substitution mutations
mc = getattr(t, CONST_CODON_METHODS[aa])(aux_tables=aux)
# Check results
assert mc.df.alt_aa.unique()[0] == aa
assert np.array_equal(mc.df.mseq.to_numpy(), np.array(exp_mseq))
assert mc.df.var_type.unique()[0] == sub_var_type
assert np.array_equal(
mc.df.ref.astype('string').to_numpy(),
np.array([
triplet for i, triplet in enumerate(seq2triplets(seq))
if 3 * i in mc.df.mut_position.unique()
]))
def test_reference_subsequence(ref_fp):
ref = ReferenceSequenceRepository()
fasta_file = get_fasta_file(ref_fp)
chromosome = 'X'
strand = '+'
start = 41341615
end = 41341635
ref_range = GenomicRange(chromosome, start, end, strand)
ref.register_genomic_range(ref_range)
ref.fetch_sequences(fasta_file)
seq = ref.get_genomic_range_sequence(ref_range)
offset = 10
pre = ref.get_genomic_range_subsequence(ref_range, start, start + offset)
print(seq)
print(pre)
print(len(pre))
assert pre == seq[:offset]
def cds_seq_to_genomic_range(cds_seq, strand, plen, slen, offset=10):
start = offset + plen
end = offset + len(cds_seq) - slen - 1
return GenomicRange('X', start, end, strand)
# organisation for which payment is received. If you are interested in using the Software commercially, please contact # [email protected]. Contact details are: [email protected] quoting reference Valiant-software. ############################# import pandas as pd from pyranges import PyRanges import pytest from valiant.models.base import GenomicRange from valiant.models.exon import ExonInfo, CDSContextRepository PYRANGES_FIELDS = ['Chromosome', 'Strand', 'Start', 'End'] GID = 'GENE_ID_001' TID = 'TRANSCRIPT_ID_001' RANGE = ('X', '+', 100, 120) GR = GenomicRange('X', 100, 120, '+') RANGES = pd.DataFrame.from_records([ RANGE ], columns=PYRANGES_FIELDS) CDS_RANGES = [ ('X', '+', 100, 120, GID, TID, 0, 0), ('X', '+', 200, 207, GID, TID, 2, 1), ('X', '+', 300, 304, GID, TID, 1, 2) ] CDS_RANGES_DF = pd.DataFrame.from_records([ (chromosome, strand, start - 1, end, gene_id, transcript_id, frame, exon_index) for chromosome, strand, start, end, gene_id, transcript_id, frame, exon_index in CDS_RANGES ], columns=PYRANGES_FIELDS + ['gene_id', 'transcript_id', 'frame', 'exon_index'])
def test_genomic_range_get_from_5_prime(strand, n, exp_start, exp_end):
chromosome = 'X'
gr = GenomicRange(chromosome, 100, 200, strand)
assert gr.get_from_5_prime(n) == GenomicRange(
chromosome, exp_start, exp_end, strand)
def test_genomic_range_region():
assert GenomicRange('X', 1, 10, '+').region == 'X:1-10'
def test_genomic_range_length():
assert len(GenomicRange('X', 5, 10, '+')) == 6
def test_genomic_range_init(chromosome, start, end, strand, valid):
with pytest.raises(ValueError) if not valid else nullcontext():
GenomicRange(chromosome, start, end, strand)
def test_genomic_range_as_unstranded_pyrange(chromosome, start, end, exp_start, strand):
assert GenomicRange(chromosome, start, end, strand).as_unstranded_pyrange() == (chromosome, exp_start, end)
def test_genomic_range_contains_position(pos,exp):
gr = GenomicRange('X', 100, 200, '+')
assert gr.contains_position(pos) == exp
chromosome, exp_start, exp_end, strand)
@pytest.mark.parametrize('strand,n,exp_start,exp_end', [
('-', 3, 100, 102),
('+', 3, 198, 200)
])
def test_genomic_range_get_from_3_prime(strand, n, exp_start, exp_end):
chromosome = 'X'
gr = GenomicRange(chromosome, 100, 200, strand)
assert gr.get_from_3_prime(n) == GenomicRange(
chromosome, exp_start, exp_end, strand)
@pytest.mark.parametrize('parent,child,exp_range,valid', [
(GenomicRange('X', 100, 200, '+'), GenomicRange('X', 150, 160, '+'), (50, 61), True),
(GenomicRange('X', 100, 200, '+'), GenomicRange('X', 20, 160, '+'), None, False),
(GenomicRange('X', 100, 200, '+'), GenomicRange('X', 150, 160, '-'), None, False)
])
def test_genomic_range_get_relative_subrange(parent, child, exp_range, valid):
with pytest.raises(ValueError) if not valid else nullcontext():
assert exp_range == parent.get_relative_subrange(child)
@pytest.mark.parametrize('pos,exp', [
(GenomicPosition('X', 100), True),
(GenomicPosition('X', 200), True),
(GenomicPosition('X', 150), True),
(GenomicPosition('X', 300), False),
(GenomicPosition('Y', 100), False)
def get_pam_protected_sequence(seq, pam_protection, chromosome='X', strand='+', pos=1):
gr = GenomicRange(chromosome, pos, len(seq), strand)
ref_seq = ReferenceSequence(seq, gr)
return PamProtectedReferenceSequence.from_reference_sequence(
ref_seq, get_dummy_pam_protected(seq) if pam_protection else seq)
