def test_annotate_pheno_with_no_pheno(self):
expected_df = DataFrame()
expected_df['SAMPLE'] = ["A", "B", "C"]
exprs_data = parse_gct("test_data/ds/dummy/small.gct")
annotated_pheno = annotate_ds_pheno_data(exprs_data)
assert_frame_equal(annotated_pheno, expected_df)
def test_annotate_pheno_mismatching_counts(self):
pheno_df = DataFrame()
pheno_df['CellType'] = ["Microglia", "Macrophage"]
pheno_df['Donor'] = ["Donor A", "Donor A"]
exprs_data = parse_gct("test_data/ds/dummy/small.gct")
with pytest.raises(ROGERUsageError):
annotate_ds_pheno_data(exprs_data, pheno_df)
def test_check_matrix_fail_on_noh_integer_data(self):
exprs_data = parse_gct("test_data/ds/dummy/small.gct")
design_matrix = DataFrame(index=["A", "B", "C"])
design_matrix['Group1'] = [1, 1, 1]
design_matrix['Group2'] = [0, "A", 0]
design_matrix['Group3'] = [0, 0, 1]
with pytest.raises(ROGERUsageError):
check_design_matrix(exprs_data.columns, design_matrix)
def test_check_matrix_fail_on_row_count_mismatch(self):
exprs_data = parse_gct("test_data/ds/dummy/small.gct")
design_matrix = DataFrame()
design_matrix['Group1'] = [1, 1]
design_matrix['Group2'] = [0, 1]
design_matrix['Group3'] = [0, 0]
with pytest.raises(ROGERUsageError):
check_design_matrix(exprs_data.columns, design_matrix)
def test_annotate_pheno_witn_simple_pheno(self):
pheno_df = DataFrame()
pheno_df['CellType'] = ["Microglia", "Macrophage", "Macrophage"]
pheno_df['Donor'] = ["Donor A", "Donor A", "Donor A"]
expected_df = DataFrame()
expected_df['SAMPLE'] = ["A", "B", "C"]
expected_df['CellType'] = ["Microglia", "Macrophage", "Macrophage"]
expected_df['Donor'] = ["Donor A", "Donor A", "Donor A"]
exprs_data = parse_gct("test_data/ds/dummy/small.gct")
annotated_pheno = annotate_ds_pheno_data(exprs_data, pheno_df)
assert_frame_equal(annotated_pheno, expected_df)
def test_check_matrix(self):
exprs_data = parse_gct("test_data/ds/dummy/small.gct")
design_matrix = DataFrame()
design_matrix['Group1'] = [1, 1, 1]
design_matrix['Group2'] = [0, 1, 0]
design_matrix['Group3'] = [0, 0, 1]
check_design_matrix(exprs_data.columns, design_matrix)
design_matrix = DataFrame(index=["A", "B", "C"])
design_matrix['Group1'] = [1, 1, 1]
design_matrix['Group2'] = [0, 1, 0]
design_matrix['Group3'] = [0, 0, 1]
check_design_matrix(exprs_data.columns, design_matrix)
def test_annotate_entrezgene(self, sqlite_in_memory):
session = sqlite_in_memory.session()
roger.persistence.geneanno.add_species(
session, roger.persistence.geneanno.human_dataset,
roger.persistence.geneanno.human_tax_id)
gct_data = parse_gct(
file_path="test_data/ds/rnaseq-example-readCounts.gct")
(feature_data, annotation_version) = roger.logic.geneanno.annotate(
session, gct_data, roger.persistence.geneanno.human_tax_id,
"entrezgene")
assert "Human genes" in annotation_version
assert_frame_equal(
read_df("test_data/ds/rnaseq-example-rogerFeatureAnno.txt"),
feature_data)
def test_annotate_chip_data(self, sqlite_in_memory):
session = sqlite_in_memory.session()
roger.persistence.geneanno.add_species(
session, roger.persistence.geneanno.human_dataset,
roger.persistence.geneanno.human_tax_id)
roger.persistence.geneanno.add_species(session, mouse_dataset,
mouse_tax_id)
gct_data = parse_gct(file_path="test_data/ds/ma-example-signals.gct")
(feature_data, annotation_version) = roger.logic.geneanno.annotate(
session, gct_data, mouse_tax_id, "affy_mouse430_2")
assert "Mouse genes" in annotation_version
assert_frame_equal(
read_df("test_data/ds/ma-example-rogerFeatureAnno.txt"),
feature_data)
def create_ds(session,
ds_type: Type[DataSet],
exprs_file,
tax_id,
symbol_type,
pheno_file=None,
name=None,
normalization_method=None,
description=None,
xref=None):
name = get_or_guess_name(name, exprs_file)
# Input checking
species_list = list_species(session)
if species_list[species_list.TaxID == tax_id].empty:
raise ROGERUsageError('Unknown taxon id: %s' % tax_id)
if session.query(DataSet).filter(
DataSet.Name == name).one_or_none() is not None:
raise ROGERUsageError("Data set with name '%s' already exists" % name)
exprs_data = parse_gct(file_path=exprs_file)
(annotation_data, annotation_version) = annotate(session, exprs_data,
tax_id, symbol_type)
pheno_data = pd.DataFrame()
if pheno_file is not None:
pheno_data = read_df(pheno_file)
annotated_pheno_data = annotate_ds_pheno_data(exprs_data, pheno_data)
return DataSetProperties(ds_type, tax_id, exprs_file, pheno_file,
exprs_data, annotated_pheno_data, annotation_data,
annotation_version, name, normalization_method,
description, xref)
