Refactored analysis

Author: aretaon

autoprot/analysis/PCA.py

@@ -15,7 +15,7 @@
 import seaborn as sns
 
 from sklearn.decomposition import PCA
-from .. import r_helper
+from autoprot import r_helper
 
 from gprofiler import GProfiler
 
@@ -25,6 +25,8 @@
 # check where this is actually used and make it local
 cmap = sns.diverging_palette(150, 275, s=80, l=55, n=9)
 
+__all__ = ["AutoPCA"]
+
 
 class AutoPCA:
     # noinspection PyUnresolvedReferences

autoprot/analysis/clustering.py

@@ -30,14 +30,16 @@
     davies_bouldin_score,
 )
 
-from .. import r_helper
+from autoprot import r_helper
 
 gp = GProfiler(user_agent="autoprot", return_dataframe=True)
 RFUNCTIONS, R = r_helper.return_r_path()
 
 # check where this is actually used and make it local
 cmap = sns.diverging_palette(150, 275, s=80, l=55, n=9)
 
+__all__ = ["HCA", "KMeans"]
+
 
 class _Cluster:
     r"""

autoprot/analysis/functional.py

@@ -16,8 +16,8 @@
 import matplotlib.pylab as plt
 import seaborn as sns
 
-from .. import visualization as vis
-from .. import r_helper
+from autoprot import visualization as vis
+from autoprot import r_helper
 
 from gprofiler import GProfiler
 
@@ -27,6 +27,8 @@
 # check where this is actually used and make it local
 cmap = sns.diverging_palette(150, 275, s=80, l=55, n=9)
 
+__all__ = ["go_analysis", "KSEA"]
+
 
 def go_analysis(
     gene_list: list[str],

autoprot/analysis/qc.py

@@ -17,7 +17,7 @@
 from operator import itemgetter
 import missingno as msn
 
-from .. import r_helper
+from autoprot import r_helper
 
 from gprofiler import GProfiler
 
@@ -27,6 +27,15 @@
 # check where this is actually used and make it local
 cmap = sns.diverging_palette(150, 275, s=80, l=55, n=9)
 
+__all__ = [
+    "miss_analysis",
+    "missed_cleavages",
+    "enrichment_specificity",
+    "SILAC_labeling_efficiency",
+    "dimethyl_labeling_efficieny",
+    "tmt6plex_labeling_efficiency",
+]
+
 
 def miss_analysis(
     df,
@@ -234,17 +243,21 @@ def missed_cleavages(df_evidence, enzyme="Trypsin/P", save=True, ax=None, title=
     today = date.today().isoformat()
 
     if "Experiment" not in df_evidence.columns.tolist():
-        print("Warning: Column [Experiment] either not unique or missing,\n\
-              column [Raw file] used")
+        print(
+            "Warning: Column [Experiment] either not unique or missing,\n\
+              column [Raw file] used"
+        )
         experiments = list(set((df_evidence["Raw file"])))
     else:
         experiments = list(set((df_evidence["Experiment"])))
 
     rawfiles = list(set((df_evidence["Raw file"])))
     if len(experiments) != len(rawfiles):
         experiments = rawfiles
-        print("Warning: Column [Experiment] either not unique or missing,\n\
-              column [Raw file] used")
+        print(
+            "Warning: Column [Experiment] either not unique or missing,\n\
+              column [Raw file] used"
+        )
 
     # calculate miss cleavage for each raw file in df_evidence
     df_missed_cleavage_summary = pd.DataFrame()
@@ -601,8 +614,10 @@ def dimethyl_labeling_efficieny(df_evidence, label, save=True) -> pd.DataFrame:
         experiments = list((df_evidence["Experiment"].unique()))
     else:
         experiments = list((df_evidence["Raw file"].unique()))
-        print("Warning: Column [Experiment] either not unique or missing,\n\
-              column [Raw file] used")
+        print(
+            "Warning: Column [Experiment] either not unique or missing,\n\
+              column [Raw file] used"
+        )
 
     df_labeling_eff = pd.DataFrame()
 

autoprot/analysis/stat_test.py

@@ -17,15 +17,17 @@
 from scipy.stats import ttest_1samp, ttest_ind
 from statsmodels.stats import multitest as mt
 
-from .. import preprocessing as pp
-from .. import r_helper
+from autoprot import preprocessing as pp
+from autoprot import r_helper
 
 gp = GProfiler(user_agent="autoprot", return_dataframe=True)
 RFUNCTIONS, R = r_helper.return_r_path()
 
 # check where this is actually used and make it local
 cmap = sns.diverging_palette(150, 275, s=80, l=55, n=9)
 
+__all__ = ["ttest", "adjust_p", "cohen_d", "limma", "rank_prod"]
+
 
 def ttest(
     df,

autoprot/analysis/stats.py

@@ -12,7 +12,7 @@
 import matplotlib.pylab as plt
 import seaborn as sns
 
-from .. import r_helper
+from autoprot import r_helper
 
 from gprofiler import GProfiler
 
@@ -22,6 +22,8 @@
 # check where this is actually used and make it local
 cmap = sns.diverging_palette(150, 275, s=80, l=55, n=9)
 
+__all__ = ["edm", "loess", "make_psm"]
+
 
 def edm(matrix_a, matrix_b):
     """