def test_heidelberg_fasta_ac(subtype_heidelberg_pass):
st, df = subtype_contigs(fasta_path=fasta_heidelberg_pass,
genome_name=genome_name,
scheme=scheme_heidelberg)
stgz, dfgz = subtype_contigs(fasta_path=fasta_gz_heidelberg_pass,
genome_name=genome_name,
scheme=scheme_heidelberg)
assert isinstance(st, Subtype)
assert isinstance(df, DataFrame)
assert isinstance(stgz, Subtype)
assert isinstance(dfgz, DataFrame)
check_subtype_attrs(st, stgz, subtype_heidelberg_pass)
check_df_fasta_cols(df)
check_df_fasta_cols(dfgz)
def test_ac_vs_bad_contigs(subtype_enteritidis_fail):
st, df = subtype_contigs(fasta_path=fasta_enteritidis_fail,
genome_name=genome_name,
scheme=scheme_enteritidis)
stgz, dfgz = subtype_contigs(fasta_path=fasta_gz_enteritidis_fail,
scheme=scheme_enteritidis,
genome_name=genome_name)
assert isinstance(st, Subtype)
assert isinstance(df, DataFrame)
assert isinstance(stgz, Subtype)
assert isinstance(dfgz, DataFrame)
check_subtype_attrs(st, stgz, subtype_enteritidis_fail)
check_df_fasta_cols(df)
check_df_fasta_cols(dfgz)
def test_typhimurium_scheme(subtype_typhimurium_pass):
st, df = subtype_contigs(fasta_path=fasta_typhimurium_pass,
genome_name=genome_name,
scheme=scheme_typhimurium)
assert isinstance(st, Subtype)
assert isinstance(df, DataFrame)
check_subtype_attrs(st, subtype_typhimurium_pass)
def test_tuberculosis_scheme(subtype_tb_AP018036_pass):
st, df = subtype_contigs(fasta_path=fasta_tb_pass,
genome_name=genome_name,
scheme=scheme_tuberculosis)
assert isinstance(st, Subtype)
assert isinstance(df, DataFrame)
check_subtype_attrs(st, subtype_tb_AP018036_pass)
def test_too_many_kmers():
scheme = 'tests/data/too_many_kmers.fasta'
fasta = 'tests/data/fail-qc-missing-levels.fasta'
with pytest.raises(SystemExit):
assert subtype_contigs(fasta_path=fasta,
genome_name=genome_name,
scheme=scheme) == SystemExit
def test_enteritidis_scheme_vs_qc_failing_contigs_unconfident_ac(subtype_enteritidis_fail_unconfident):
st, df = subtype_contigs(fasta_path=fasta_enteritidis_unconfident,
genome_name=genome_name,
scheme=scheme_enteritidis)
stgz, dfgz = subtype_contigs(fasta_path=fasta_gz_enteritidis_unconfident,
genome_name=genome_name,
scheme=scheme_enteritidis)
assert isinstance(st, Subtype)
assert isinstance(df, DataFrame)
assert isinstance(stgz, Subtype)
assert isinstance(dfgz, DataFrame)
check_subtype_attrs(st, stgz, subtype_enteritidis_fail_unconfident)
assert 'Unconfident Results Error 4' in st.qc_message
assert 'Unconfident Results Error 4' in stgz.qc_message
check_df_fasta_cols(df)
check_df_fasta_cols(dfgz)
def test_heidelberg_scheme_and_lowcase_seq_inputs(
subtype_heidelberg_SRR1002850_pass):
st, df = subtype_contigs(fasta_path=fasta_heidelberg_pass,
genome_name=genome_name,
scheme=scheme_heidelberg)
assert isinstance(st, Subtype)
assert isinstance(df, DataFrame)
check_subtype_attrs(st, subtype_heidelberg_SRR1002850_pass)
def test_missing_hierarchy_levels_in_subtype():
scheme = 'heidelberg'
fasta = 'tests/data/fail-qc-missing-levels.fasta'
st, df = subtype_contigs(fasta_path=fasta,
genome_name=genome_name,
scheme=scheme)
assert isinstance(st, Subtype)
assert isinstance(df, pd.DataFrame)
assert st.scheme == scheme
assert st.qc_status == QC.FAIL
assert QC.UNCONFIDENT_RESULTS_ERROR_4 in st.qc_message
assert "kmers for nested hierarchical subtype(s)" in st.qc_message
assert "2.1" in st.qc_message
assert "2.1.1" in st.qc_message
def test_unconfident_subtype():
scheme = 'enteritidis'
fasta = 'tests/data/fail-qc-unconfident-subtype.fasta'
st, df = subtype_contigs(fasta_path=fasta,
genome_name=genome_name,
scheme=scheme)
assert isinstance(st, Subtype)
assert isinstance(df, pd.DataFrame)
assert st.scheme == scheme
assert st.qc_status == QC.FAIL
assert QC.UNCONFIDENT_RESULTS_ERROR_4 in st.qc_message
assert "kmers for downstream subtype(s)" in st.qc_message
assert "'2.1.5.4.2'" in st.qc_message
assert "'2.1.5.4.1'" in st.qc_message
def test_mixed_subtype_positive_negative_kmers_same_target():
scheme = 'heidelberg'
fasta = 'tests/data/fail-qc-mixed-subtype-pos-neg-kmers.fasta'
st, df = subtype_contigs(fasta_path=fasta,
genome_name=genome_name,
scheme=scheme)
assert isinstance(st, Subtype)
assert isinstance(df, pd.DataFrame)
assert st.scheme == scheme
assert st.qc_status == QC.FAIL
expected_qc_msg = (
'FAIL: Mixed subtype; the positive and negative kmers were found for the same '
'target sites 202001, 600783, 1049933, 1193219, 2778621, 2904061, '
'3278067, 3867228, 4499501, 4579224, 4738855, 202001, '
'600783, 1049933, 1193219, 2778621, 2904061, 3278067, '
'3867228, 4499501, 4579224, 4738855 for subtype "1.1".')
print(st.qc_message)
print(expected_qc_msg)
assert expected_qc_msg in st.qc_message
