def test_reading(files_source):
mwtabfile_generator = mwtab.read_files(files_source)
mwtabfiles_list = list(mwtabfile_generator)
mwtabfiles_study_ids_set = set(mwf.study_id for mwf in mwtabfiles_list)
mwtabfiles_analysis_ids_set = set(mwf.analysis_id
for mwf in mwtabfiles_list)
assert mwtabfiles_study_ids_set == {"ST000001", "ST000002"}
assert mwtabfiles_analysis_ids_set == {"AN000001", "AN000002"}
def test_from_local_file():
mwtabfile_generator = mwtab.read_files(
"tests/example_data/mwtab_files/ST000001_AN000001.txt",
"tests/example_data/mwtab_files/ST000002_AN000002.txt")
mwtabfile1 = next(mwtabfile_generator)
mwtabfile2 = next(mwtabfile_generator)
assert mwtabfile1.study_id == "ST000001" and mwtabfile2.study_id == "ST000002"
assert mwtabfile1.analysis_id == "AN000001" and mwtabfile2.analysis_id == "AN000002"
def test_converter_module(from_path, to_path, from_format, to_format):
converter = Converter(from_path=from_path,
to_path=to_path,
from_format=from_format,
to_format=to_format)
converter.convert()
mwtabfile_generator = mwtab.read_files(to_path)
mwtabfiles_list = list(mwtabfile_generator)
mwtabfiles_study_ids_set = set(mwf.study_id for mwf in mwtabfiles_list)
mwtabfiles_analysis_ids_set = set(mwf.analysis_id
for mwf in mwtabfiles_list)
assert mwtabfiles_study_ids_set.issubset({"ST000001", "ST000002"})
assert mwtabfiles_analysis_ids_set.issubset({"AN000001", "AN000002"})
def mwtab_to_df(path, id_mapping='pubchem_id'):
'''
Parse mwtab file to df
:param path: path of mwtab file
:param id_mapping: which db will be used to annotate metabolite names.
Those are valid inputs {None, 'PubChem ID', 'KEGG ID', 'HMDB'}
'''
f = next(mwtab.read_files(path))
id_factor_mapping = {
i['local_sample_id']: i['factors'].split('-')[1].strip()
if not i['factors'].startswith('Source:Method Blanks') else 'healthy'
for i in f['SUBJECT_SAMPLE_FACTORS']['SUBJECT_SAMPLE_FACTORS']
}
metabolites_names = {
i['metabolite_name']: i[id_mapping]
for i in f['METABOLITES']['METABOLITES_START']['DATA']
if id_mapping in i and i[id_mapping]
}
metabolite_measurements = dict()
for i in f['MS_METABOLITE_DATA']['MS_METABOLITE_DATA_START']['DATA']:
m = i['metabolite_name']
del i['metabolite_name']
if id_mapping:
if m in metabolites_names:
metabolite_measurements[metabolites_names[m]] = i
else:
metabolite_measurements[m] = i
df = pd.DataFrame(metabolite_measurements, dtype=float)
df = df[[i in id_factor_mapping for i in df.index]]
labels = [id_factor_mapping[i] for i in df.index]
labels = [i if i != 'Adenocarcnoma' else 'Adenocarcinoma' for i in labels]
df.insert(0, 'labels', labels)
df = df.reset_index()
del df['index']
return df.replace('', np.nan).dropna(axis=1, how='any')
def test_from_analysis_id():
mwtabfile_generator = mwtab.read_files("5")
mwtabfile = next(mwtabfile_generator)
assert mwtabfile.study_id == "ST000004"
assert mwtabfile.analysis_id == "AN000005"