def test_match_mz_multiple_matches_mode_reduce():
tolerance = 2
mz1 = np.array([50, 75, 100, 125, 150], dtype=float)
mz2 = np.array([49, 51, 78, 99, 100, 101, 126, 150, 151], dtype=float)
sp2 = np.array([100] * mz2.size, dtype=float)
# expected values for match/no match indices
# in closest mode, argmin is used to select the closest value. If more
# than one value has the same difference, the first one in the array is
# going to be selected.
mz1_match_index = np.array([0, 2, 3, 4], dtype=int)
mz2_match_index = np.array([0, 1, 3, 4, 5, 6, 7, 8], dtype=int)
mz2_no_match_index = np.array([2], dtype=int)
expected_mz2_match = [50.0, 100.0, 126.0, 150.5]
expected_sp2_match = [200, 300, 100, 200]
mode = "reduce"
test_mz1_index, mz2_match, sp2_match, mz2_no_match, sp2_no_match = \
lcms._match_mz(mz1, mz2, sp2, tolerance, mode, np.mean, np.sum)
# test match index
assert np.array_equal(mz1_match_index, test_mz1_index)
# test match mz and sp values
assert np.allclose(mz2_match, expected_mz2_match)
assert np.allclose(sp2_match, expected_sp2_match)
# test no match mz and sp values
assert np.array_equal(mz2[mz2_no_match_index], mz2_no_match)
assert np.array_equal(sp2[mz2_no_match_index], sp2_no_match)
def test_match_mz_invalid_mode():
tolerance = 2
mz1 = np.array([50, 75, 100, 125, 150])
mz2 = np.array([49, 51, 78, 99, 100, 101, 126, 150, 151])
sp2 = np.array([100] * mz2.size)
# expected values for match/no match indices
# in closest mode, argmin is used to select the closest value. If more
# than one value has the same difference, the first one in the array is
# going to be selected.
mz1_match_index = np.array([0, 2, 3, 4], dtype=int)
mz2_match_index = np.array([0, 4, 6, 7], dtype=int)
mz2_no_match_index = np.array([1, 2, 3, 5, 8], dtype=int)
mode = "invalid-mode"
with pytest.raises(ValueError):
test_mz1_index, mz2_match, sp2_match, mz2_no_match, sp2_no_match = \
lcms._match_mz(mz1, mz2, sp2, tolerance, mode, np.mean, np.mean)
def test_match_mz_all_match():
tolerance = 2
mz1 = np.array([50, 75, 100, 125, 150])
mz2 = np.array([51, 77, 99, 126, 150])
sp2 = np.array([100] * mz2.size)
# expected values for match/no match indices
mz1_match_index = np.array([0, 1, 2, 3, 4], dtype=int)
mz2_match_index = np.array([0, 1, 2, 3, 4], dtype=int)
mz2_no_match_index = np.array([], dtype=int)
mode = "closest"
test_mz1_index, mz2_match, sp2_match, mz2_no_match, sp2_no_match = \
lcms._match_mz(mz1, mz2, sp2, tolerance, mode, np.mean, np.mean)
# test match index
assert np.array_equal(mz1_match_index, test_mz1_index)
# test match mz and sp values
assert np.array_equal(mz2[mz2_match_index], mz2_match)
assert np.array_equal(sp2[mz2_match_index], sp2_match)
# test no match mz and sp values
assert np.array_equal(mz2[mz2_no_match_index], mz2_no_match)
assert np.array_equal(sp2[mz2_no_match_index], sp2_no_match)
def test_match_mz_multiple_matches_mode_closest():
tolerance = 2
mz1 = np.array([50, 75, 100, 125, 150])
mz2 = np.array([49, 51, 78, 99, 100, 101, 126, 150, 151])
sp2 = np.array([100] * mz2.size)
# expected values for match/no match indices
# in closest mode, argmin is used to select the closest value. If more
# than one value has the same difference, the first one in the array is
# going to be selected.
mz1_match_index = np.array([0, 2, 3, 4], dtype=int)
mz2_match_index = np.array([0, 4, 6, 7], dtype=int)
mz2_no_match_index = np.array([1, 2, 3, 5, 8], dtype=int)
mode = "closest"
test_mz1_index, mz2_match, sp2_match, mz2_no_match, sp2_no_match = \
lcms._match_mz(mz1, mz2, sp2, tolerance, mode, np.mean, np.mean)
# test match index
assert np.array_equal(mz1_match_index, test_mz1_index)
# test match mz and sp values
assert np.array_equal(mz2[mz2_match_index], mz2_match)
assert np.array_equal(sp2[mz2_match_index], sp2_match)
# test no match mz and sp values
assert np.array_equal(mz2[mz2_no_match_index], mz2_no_match)
assert np.array_equal(sp2[mz2_no_match_index], sp2_no_match)