def setUp(self):
self.f1 = Peakel(1256.52, 0.0, 0.0, 1256.52)
self.f1.area_by_sample_name = {'a': 102564, 'b': 130156, 'c': 150000, 'd': 10000}
self.f1.charge = 1
self.f1.polarity = -1
self.f2 = Peakel(1258.52, 0.0, 0.0, 1258.52)
self.f2.area_by_sample_name = {'a': 102564 / 10.0, 'b': 130156 / 10.0, 'c': 150000 / 10.0, 'd': 10000 / 10.0}
self.f3 = Peakel(1261.52, 0.0, 0.0, 1261.52)
self.f3.area_by_sample_name = {'a': 102564 / 2.0, 'b': 130156 / 2.3, 'c': 150000 / 2.25, 'd': 10000 / 1.89}
self.f4 = Peakel(1262.52, 0.0, 0.0, 1262.52)
self.f4.area_by_sample_name = {'a': 102564 * 3.0, 'b': 130156 / 1.8, 'c': 150000 / 4.1, 'd': 10000 * 0.5}
self.f5 = Peakel(1274.52, 0.0, 0.0, 1274.52)
self.f5.area_by_sample_name = {'a': 102564 * 12.23, 'b': 130156 * 30.0, 'c': 150000 * 33.0, 'd': 10000 * 44.0}
self.f6 = Peakel(1275.52, 0.0, 0.0, 1275.52)
self.f6.area_by_sample_name = {'a': 102564 * 44.0, 'b': 130156 * 60.0, 'c': 150000 * 66.0, 'd': 10000 * 88.0}
self.f7 = Peakel(1281.52, 0.0, 0.0, 1281.52)
self.f7.area_by_sample_name = {'a': 102564 * 6.0, 'b': 130156 * 4.56, 'c': 150000 / 78.0, 'd': 10000 / 1236.0}
self.features = [self.f1, self.f2, self.f3, self.f4, self.f5, self.f6, self.f7]
self.f1.isotopes.add(self.f2)
self.f2.set_main_attribution(Attribution('isotope c13', self.f1.id, 1))
self.f1.adducts.add(self.f3)
self.f3.set_main_attribution(Attribution('[M+Na+]', self.f1.id, 1))
t = ("acession", "name", "formula", "inchi", "mono_mass", "average_mass", "description", "status", "origin",
"kegg_id", "isotopic_pattern_pos", "isotopic_pattern_neg")
self.f1.annotations.append(Annotation(metabolite=Metabolite(*t), for_adduct='H2'))
def _to_peakel_obj(self, d):
""" convert csv data to peakel objects """
p = Peakel(float(d[PeakListReader.KEYS[0]]),
float(d[PeakListReader.KEYS[1]]),
float(d[PeakListReader.KEYS[2]]),
float(d[PeakListReader.KEYS[3]]),
float(d[PeakListReader.KEYS[4]]),
float(d[PeakListReader.KEYS[5]]))
# set the right polarity
polarity = None
if "Mode" in list(d.keys()):
polarity = 1 if d["Mode"] == 'Positif' else -1
else:
p.polarity = self.exp_design.polarity
# remove keys
for k in (PeakListReader.KEYS + self.directories + [
"", "BIO", "mzmed", "rt.minutes", "Var", "Blc.Ext", "BLC",
"Mode", "Correlation_Dilution_Log", "NOT_M.QC", "NOT_M.Blc",
"NOT_QC.Blc", "NOT_CV..", "NOT_CV", "NOT_Correl", "Correl",
"NOT_BIO.Blc", "NOT_nom", "rt.min", "Negatifs"
]):
try:
del d[k]
except KeyError:
pass
# assign area of each sample
p.area_by_sample_name.update({a: float(b) for a, b in list(d.items())})
p.area = np.median(list(p.area_by_sample_name.values()))
if not p.area:
p.area = np.mean(list(p.area_by_sample_name.values()))
logging.debug(
"an elution peak has the median of "
"its area equals to 0 ! Using mean instead: {}.".format(
p.area))
return p
def test_dbscan_clustering_for_alignment(self):
f1 = Peakel(1256.52, 0.0, 0.0, 100.0)
f2 = Peakel(1258.52, 0.0, 0.0, 500.52)
f3 = Peakel(1257.52, 0.0, 0.0, 101.52)
f4 = Peakel(1600.52, 0.0, 0.0, 99.52)
f7 = Peakel(1600.86, 0.0, 0.0, 107.12)
f5 = Peakel(1600.52, 0.0, 0.0, 3205.52)
f6 = Peakel(1456.52, 0.0, 0.0, 600.52)
peakels = [f1, f2, f3, f4, f5, f6, f7]
peakels_by_sample = {'a': {f1, f2, f4}, 'b': {f3, f5, f6}}
sample_by_peakel = {f1: 'a', f2: 'a', f4: 'a', f3: 'b', f5: 'b', f6: 'b', f7: 'b'}
values = [[x.moz, x.rt] for x in peakels]
clusters = clusterize_dbscan(values, peakels, eps=5, min_samples=1)
for c in clusters:
for p in c:
print(p.moz, p.rt, sample_by_peakel[p])
print('\n')
