diff --git a/alignement.py b/alignement.py
index 1c11a0737517221c5edf939c0991ada89efe594c..4e290fb52501fc7e04826d0fe063ad6b9d48506d 100644
--- a/alignement.py
+++ b/alignement.py
@@ -42,7 +42,7 @@ def numerical_to_alphabetical(arr):
return seq
def align(dataset, reference):
- seq_ref = reference['sequence']
+ seq_ref = reference['Sequence']
seq_common = dataset['Sequence']
seq_ref = seq_ref.tolist()
seq_common = seq_common.tolist()
@@ -57,10 +57,10 @@ def align(dataset, reference):
indices_common = dict((k, i) for i, k in enumerate(seq_common))
indices_common = [indices_common[x] for x in inter]
- rt_ref = reference['irt'][ind_dict_ref].reset_index()
+ rt_ref = reference['Retention time'][ind_dict_ref].reset_index()
rt_data = dataset['Retention time'][indices_common].reset_index()
- xout, yout, wout = loess_1d(np.array(rt_data['Retention time'].tolist()), np.array(rt_ref['irt'].tolist()),
+ xout, yout, wout = loess_1d(np.array(rt_data['Retention time'].tolist()), np.array(rt_ref['Retention time'].tolist()),
xnew=dataset['Retention time'],
degree=1, frac=0.5,
npoints=None, rotate=False, sigy=None)
@@ -68,38 +68,38 @@ def align(dataset, reference):
return dataset
-data_ori = RT_Dataset(None, 'database/data_train.csv', 'train', 25).data
-data_ori['sequence'] = data_ori['sequence'].map(numerical_to_alphabetical)
+data_ori = load_data('msms/msms30_01.txt').reset_index(drop=True)
+# data_ori['sequence'] = data_ori['sequence'].map(numerical_to_alphabetical)
data_train = load_data('msms/msms16_01.txt').reset_index(drop=True)
# data_train = pd.read_pickle('database/data_DIA_16_01.pkl').reset_index(drop=True)
data_align = align(data_train, data_ori)
-data_align.to_pickle('database/data_DIA_16_01_aligned.pkl')
+data_align.to_pickle('database/data_DIA_16_01_aligned30_01.pkl')
data_train = load_data('msms/msms17_01.txt').reset_index(drop=True)
# data_train = pd.read_pickle('database/data_DIA_17_01.pkl').reset_index(drop=True)
data_align = align(data_train, data_ori)
-data_align.to_pickle('database/data_DIA_17_01_aligned.pkl')
+data_align.to_pickle('database/data_DIA_17_01_aligned30_01.pkl')
data_train = load_data('msms/msms20_01.txt').reset_index(drop=True)
# data_train = pd.read_pickle('database/data_DIA_20_01.pkl').reset_index(drop=True)
data_align = align(data_train, data_ori)
-data_align.to_pickle('database/data_DIA_20_01_aligned.pkl')
+data_align.to_pickle('database/data_DIA_20_01_aligned30_01.pkl')
data_train = load_data('msms/msms23_01.txt').reset_index(drop=True)
# data_train = pd.read_pickle('database/data_DIA_23_01.pkl').reset_index(drop=True)
data_align = align(data_train, data_ori)
-data_align.to_pickle('database/data_DIA_23_01_aligned.pkl')
+data_align.to_pickle('database/data_DIA_23_01_aligned30_01.pkl')
data_train = load_data('msms/msms24_01.txt').reset_index(drop=True)
# data_train = pd.read_pickle('database/data_DIA_24_01.pkl').reset_index(drop=True)
data_align = align(data_train, data_ori)
-data_align.to_pickle('database/data_DIA_24_01_aligned.pkl')
+data_align.to_pickle('database/data_DIA_24_01_aligned30_01.pkl')
-data_train = load_data('msms/msms30_01.txt').reset_index(drop=True)
-# data_train = pd.read_pickle('database/data_DIA_30_01.pkl').reset_index(drop=True)
-data_align = align(data_train, data_ori)
-data_align.to_pickle('database/data_DIA_30_01_aligned.pkl')
+# data_train = load_data('msms/msms30_01.txt').reset_index(drop=True)
+# # data_train = pd.read_pickle('database/data_DIA_30_01.pkl').reset_index(drop=True)
+# data_align = align(data_train, data_ori)
+# data_align.to_pickle('database/data_DIA_30_01_aligned30_01.pkl')
#
# plt.scatter(data_train['Retention time'], data_align['Retention time'], s=1)
# plt.savefig('test_align_2.png')
diff --git a/data_viz.py b/data_viz.py
index 9a609cc9b838e896c09ef4e47aa15c6d4dd69720..de2373da73dd6c09664588d71603657617218da6 100644
--- a/data_viz.py
+++ b/data_viz.py
@@ -139,19 +139,19 @@ def histo_abs_error(dataframe, display=False, save=False, path=None):
plt.savefig(path)
-def random_color_deterministic(df):
+def random_color_deterministic(df, column):
def rd10(str):
color = list(mcolors.CSS4_COLORS)
random.seed(str)
return color[random.randint(0,147)]
- df['color']=df['seq'].map(rd10)
+ df['color']=df[column].map(rd10)
-def scatter_rt(dataframe, display=False, save=False, path=None, color = False):
+def scatter_rt(dataframe, display=False, save=False, path=None, color = False, col = 'seq'):
fig, ax = plt.subplots()
if color :
- random_color_deterministic(dataframe)
+ random_color_deterministic(dataframe, col)
ax.scatter(dataframe['true rt'], dataframe['rt pred'], s=.1, color = dataframe['color'])
else :
ax.scatter(dataframe['true rt'], dataframe['rt pred'], s=.1)
@@ -243,6 +243,7 @@ def compare_error(df1, df2, display=False, save=False, path=None):
if save:
plt.savefig(path)
+
def add_length(dataframe):
def fonc(a):
a = a.replace('[', '')
@@ -253,23 +254,30 @@ def add_length(dataframe):
dataframe['length']=dataframe['seq'].map(fonc)
-df = pd.read_csv('output/out_common_ISA_ISA_eval.csv')
-add_length(df)
-df['abs_error'] = np.abs(df['rt pred']-df['true rt'])
-histo_abs_error(df, display=False, save=True, path='fig/custom model res/histo_ISA_ISA_eval.png')
-scatter_rt(df, display=False, save=True, path='fig/custom model res/RT_pred_ISA_ISA_eval.png', color=True)
-histo_length_by_error(df, bins=10, display=False, save=True, path='fig/custom model res/histo_length_ISA_ISA_eval.png')
-
-df = pd.read_csv('output/out_common_prosit_prosit_eval.csv')
-add_length(df)
-df['abs_error'] = np.abs(df['rt pred']-df['true rt'])
-histo_abs_error(df, display=False, save=True, path='fig/custom model res/histo_prosit_prosit_eval.png')
-scatter_rt(df, display=False, save=True, path='fig/custom model res/RT_pred_prosit_prosit_eval.png', color=True)
-histo_length_by_error(df, bins=10, display=False, save=True, path='fig/custom model res/histo_length_prosit_prosit_eval.png')
-
-df = pd.read_csv('output/out_common_transfereval.csv')
+# df = pd.read_csv('output/out_common_ISA_ISA_eval.csv')
+# add_length(df)
+# df['abs_error'] = np.abs(df['rt pred']-df['true rt'])
+# histo_abs_error(df, display=False, save=True, path='fig/custom model res/histo_ISA_ISA_eval.png')
+# scatter_rt(df, display=False, save=True, path='fig/custom model res/RT_pred_ISA_ISA_eval.png', color=True)
+# histo_length_by_error(df, bins=10, display=False, save=True, path='fig/custom model res/histo_length_ISA_ISA_eval.png')
+#
+# df = pd.read_csv('output/out_common_prosit_prosit_eval.csv')
+# add_length(df)
+# df['abs_error'] = np.abs(df['rt pred']-df['true rt'])
+# histo_abs_error(df, display=False, save=True, path='fig/custom model res/histo_prosit_prosit_eval.png')
+# scatter_rt(df, display=False, save=True, path='fig/custom model res/RT_pred_prosit_prosit_eval.png', color=True)
+# histo_length_by_error(df, bins=10, display=False, save=True, path='fig/custom model res/histo_length_prosit_prosit_eval.png')
+#
+# df = pd.read_csv('output/out_common_transfereval.csv')
+# add_length(df)
+# df['abs_error'] = np.abs(df['rt pred']-df['true rt'])
+# histo_abs_error(df, display=False, save=True, path='fig/custom model res/histo_prosit_ISA_eval.png')
+# scatter_rt(df, display=False, save=True, path='fig/custom model res/RT_pred_prosit_ISA_eval.png', color=True)
+# histo_length_by_error(df, bins=10, display=False, save=True, path='fig/custom model res/histo_length_prosit_ISA_eval.png')
+
+df = pd.read_csv('output/out_common_ISA_ISA_eval_2.csv')
add_length(df)
df['abs_error'] = np.abs(df['rt pred']-df['true rt'])
-histo_abs_error(df, display=False, save=True, path='fig/custom model res/histo_prosit_ISA_eval.png')
-scatter_rt(df, display=False, save=True, path='fig/custom model res/RT_pred_prosit_ISA_eval.png', color=True)
-histo_length_by_error(df, bins=10, display=False, save=True, path='fig/custom model res/histo_length_prosit_ISA_eval.png')
\ No newline at end of file
+histo_abs_error(df, display=False, save=True, path='fig/custom model res/histo_ISA_ISA_eval_2.png')
+scatter_rt(df, display=False, save=True, path='fig/custom model res/RT_pred_ISA_ISA_eval_2_seq.png', color=True, col = 'seq')
+histo_length_by_error(df, bins=10, display=False, save=True, path='fig/custom model res/histo_length_ISA_ISA_eval_2.png')
\ No newline at end of file
diff --git a/database/data_DIA_ISA_55_test_30_01.pkl b/database/data_DIA_ISA_55_test_30_01.pkl
new file mode 100644
index 0000000000000000000000000000000000000000..b6091e333179587f777da5d16c2880e5bdda77d7
Binary files /dev/null and b/database/data_DIA_ISA_55_test_30_01.pkl differ
diff --git a/database/data_DIA_ISA_55_train_30_01.pkl b/database/data_DIA_ISA_55_train_30_01.pkl
new file mode 100644
index 0000000000000000000000000000000000000000..0ee7e1b319fa2376ae34864af21725b36c7ec267
Binary files /dev/null and b/database/data_DIA_ISA_55_train_30_01.pkl differ
diff --git a/msms_processing.py b/msms_processing.py
index b202abb2aba113030d18362ddf0b73e11ed45881..05cbac233ff85d0d21efe0fd7eda470963c44344 100644
--- a/msms_processing.py
+++ b/msms_processing.py
@@ -82,15 +82,15 @@ def mscatter(x,y, ax=None, m=None, **kw):
#data gradient 3 :
# 17/01 23/01 24/01
if __name__ == '__main__':
- data_1 = pd.read_pickle('database/data_DIA_16_01_aligned.pkl')
+ data_1 = pd.read_pickle('database/data_DIA_16_01_aligned30_01.pkl')
data_1['file']= 1
- data_2 = pd.read_pickle('database/data_DIA_17_01_aligned.pkl')
+ data_2 = pd.read_pickle('database/data_DIA_17_01_aligned30_01.pkl')
data_2['file'] = 2
- data_3 = pd.read_pickle('database/data_DIA_20_01_aligned.pkl')
+ data_3 = pd.read_pickle('database/data_DIA_20_01_aligned30_01.pkl')
data_3['file'] = 3
- data_4 = pd.read_pickle('database/data_DIA_23_01_aligned.pkl')
+ data_4 = pd.read_pickle('database/data_DIA_23_01_aligned30_01.pkl')
data_4['file'] = 4
- data_5 = pd.read_pickle('database/data_DIA_24_01_aligned.pkl')
+ data_5 = pd.read_pickle('database/data_DIA_24_01_aligned30_01.pkl')
data_5['file'] = 5
data_6 = pd.read_pickle('database/data_DIA_30_01_aligned.pkl')
data_6['file'] = 6
@@ -115,8 +115,8 @@ if __name__ == '__main__':
dataset_train = pd.concat(train_set).reset_index(drop=True)
dataset_test = pd.concat(test_set).reset_index(drop=True)
- dataset_train.to_pickle('database/data_DIA_ISA_55_train.pkl')
- dataset_test.to_pickle('database/data_DIA_ISA_55_test.pkl')
+ dataset_train.to_pickle('database/data_DIA_ISA_55_train_30_01.pkl')
+ dataset_test.to_pickle('database/data_DIA_ISA_55_test_30_01.pkl')