def test_estimate_fdr_digitize_works():
fdr = FDR(job_id=0, decoy_sample_size=1, target_adducts=['+H'], db=None)
fdr.fdr_levels = [0.4, 0.8]
fdr.td_df = pd.DataFrame([['C1', '+H', '+Cu'],
['C2', '+H', '+Ag'],
['C3', '+H', '+Cl'],
['C4', '+H', '+Co']],
columns=['sf', 'ta', 'da'])
msm_df = pd.DataFrame([['C1', '+H', 1.0],
['C2', '+H', 0.75],
['C3', '+H', 0.5],
['C4', '+H', 0.25],
['C1', '+Cu', 0.75],
['C2', '+Ag', 0.3],
['C3', '+Cl', 0.25],
['C4', '+Co', 0.1]],
columns=['sf', 'adduct', 'msm']).set_index(['sf', 'adduct']).sort_index()
exp_sf_df = pd.DataFrame([['C1', '+H', 0.4],
['C2', '+H', 0.4],
['C3', '+H', 0.4],
['C4', '+H', 0.8]],
columns=['sf', 'adduct', 'fdr']).set_index(['sf', 'adduct'])
assert_frame_equal(fdr.estimate_fdr(msm_df), exp_sf_df)
def test_estimate_fdr_digitize_works():
fdr = FDR(job_id=0, decoy_sample_size=1, target_adducts=['+H'], db=None)
fdr.fdr_levels = [0.4, 0.8]
fdr.td_df = pd.DataFrame([['C1', '+H', '+Cu'],
['C2', '+H', '+Ag'],
['C3', '+H', '+Cl'],
['C4', '+H', '+Co']],
columns=['sf', 'ta', 'da'])
msm_df = pd.DataFrame([['C1', '+H', 1.0],
['C2', '+H', 0.75],
['C3', '+H', 0.5],
['C4', '+H', 0.25],
['C1', '+Cu', 0.75],
['C2', '+Ag', 0.3],
['C3', '+Cl', 0.25],
['C4', '+Co', 0.1]],
columns=['sf', 'adduct', 'msm']).set_index(['sf', 'adduct']).sort_index()
exp_sf_df = pd.DataFrame([['C1', '+H', 0.4],
['C2', '+H', 0.4],
['C3', '+H', 0.4],
['C4', '+H', 0.8]],
columns=['sf', 'adduct', 'fdr']).set_index(['sf', 'adduct'])
assert_frame_equal(fdr.estimate_fdr(msm_df), exp_sf_df)
def test_estimate_fdr_returns_correct_df():
fdr = FDR(job_id=0, decoy_sample_size=2, target_adducts=['+H'], db=None)
fdr.fdr_levels = [0.2, 0.8]
fdr.td_df = pd.DataFrame([['H2O', '+H', '+Cu'], ['H2O', '+H', '+Co'],
['C2H2', '+H', '+Ag'], ['C2H2', '+H', '+Ar']],
columns=['sf', 'ta', 'da'])
msm_df = pd.DataFrame(
[['H2O', '+H', 0.85], ['C2H2', '+H', 0.5], ['H2O', '+Cu', 0.5],
['H2O', '+Co', 0.5], ['C2H2', '+Ag', 0.75], ['C2H2', '+Ar', 0.0]],
columns=['sf', 'adduct', 'msm']).set_index(['sf',
'adduct']).sort_index()
exp_sf_df = pd.DataFrame([['H2O', '+H', 0.2], ['C2H2', '+H', 0.8]],
columns=['sf', 'adduct',
'fdr']).set_index(['sf', 'adduct'])
assert_frame_equal(fdr.estimate_fdr(msm_df), exp_sf_df)
def test_estimate_fdr_digitize_works():
fdr = FDR(0, 0, 1, ['+H'], None)
fdr.fdr_levels = [0.4, 0.8]
fdr.td_df = pd.DataFrame([[1, '+H', '+Cu'], [2, '+H', '+Ag'],
[3, '+H', '+Cl'], [4, '+H', '+Co']],
columns=['sf_id', 'ta', 'da'])
msm_df = pd.DataFrame(
[[1, '+H', 1.0], [2, '+H', 0.75], [3, '+H', 0.5], [4, '+H', 0.25],
[1, '+Cu', 0.75], [2, '+Ag', 0.3], [3, '+Cl', 0.25], [4, '+Co', 0.1]],
columns=['sf_id', 'adduct', 'msm']).set_index(['sf_id',
'adduct']).sort_index()
exp_sf_df = pd.DataFrame(
[[1, '+H', 0.4], [2, '+H', 0.4], [3, '+H', 0.4], [4, '+H', 0.8]],
columns=['sf_id', 'adduct', 'fdr']).set_index(['sf_id', 'adduct'])
assert_frame_equal(fdr.estimate_fdr(msm_df), exp_sf_df)
def test_estimate_fdr_returns_correct_df():
fdr = FDR(0, 0, 2, ['+H'], None)
fdr.fdr_levels = [0.2, 0.8]
fdr.td_df = pd.DataFrame([[1, '+H', '+Cu'], [1, '+H', '+Co'],
[2, '+H', '+Ag'], [2, '+H', '+Ar']],
columns=['sf_id', 'ta', 'da'])
msm_df = pd.DataFrame([[1, '+H', 0.85], [2, '+H', 0.5], [1, '+Cu', 0.5],
[1, '+Co', 0.5], [2, '+Ag', 0.75], [2, '+Ar', 0.0]],
columns=['sf_id', 'adduct',
'msm']).set_index(['sf_id',
'adduct']).sort_index()
exp_sf_df = pd.DataFrame([[1, '+H', 0.2], [2, '+H', 0.8]],
columns=['sf_id', 'adduct',
'fdr']).set_index(['sf_id', 'adduct'])
assert_frame_equal(fdr.estimate_fdr(msm_df), exp_sf_df)
def test_estimate_fdr_returns_correct_df():
fdr = FDR(job_id=0, decoy_sample_size=2, target_adducts=['+H'], db=None)
fdr.fdr_levels = [0.2, 0.8]
fdr.td_df = pd.DataFrame([['H2O', '+H', '+Cu'],
['H2O', '+H', '+Co'],
['C2H2', '+H', '+Ag'],
['C2H2', '+H', '+Ar']],
columns=['sf', 'ta', 'da'])
msm_df = pd.DataFrame([['H2O', '+H', 0.85],
['C2H2', '+H', 0.5],
['H2O', '+Cu', 0.5],
['H2O', '+Co', 0.5],
['C2H2', '+Ag', 0.75],
['C2H2', '+Ar', 0.0]],
columns=['sf', 'adduct', 'msm']).set_index(['sf', 'adduct']).sort_index()
exp_sf_df = pd.DataFrame([['H2O', '+H', 0.2], ['C2H2', '+H', 0.8]],
columns=['sf', 'adduct', 'fdr']).set_index(['sf', 'adduct'])
assert_frame_equal(fdr.estimate_fdr(msm_df), exp_sf_df)