small refactoring

RECETOX · Jul 10, 2024 · e564efe · e564efe
1 parent 5872db6
commit e564efe
Show file tree

Hide file tree

Showing 7 changed files with 18 additions and 30 deletions.
diff --git a/analysis/Python_scripts/Fig3_correlations.py b/analysis/Python_scripts/Fig3_correlations.py
@@ -4,12 +4,9 @@
 from utils import *
 from plotting import *
 
-matchms_scores = pd.read_csv("../data/output_matching/matchms/matchms_tol_0.0035_1%I_all_peaks_with_0s_only_matching.tsv", sep="\t")
-matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_scores': 'scores'}, inplace=True)
-matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_matches': 'matches'}, inplace=True)
+matchms_scores = load_matchms_scores()
 
-matchms_scores = append_classes(matchms_scores, 'query')
-df = normalize_df(append_spectrum_metadata(matchms_scores), matches_norm_col=None)
+df = normalize_df(matchms_scores, matches_norm_col=None)
 del df['peak_comments']
 
 matches_col = 'matches'

diff --git a/analysis/Python_scripts/Fig4_superclass_histograms.py b/analysis/Python_scripts/Fig4_superclass_histograms.py
@@ -10,11 +10,7 @@
 from utils import *
 from plotting import *
 
-matchms_scores = pd.read_csv("../data/output_matching/matchms/matchms_tol_0.0035_1%I_all_peaks_with_0s_only_matching.tsv", sep="\t")
-matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_scores': 'scores'}, inplace=True)
-matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_matches': 'matches'}, inplace=True)
-matchms_scores = append_classes(matchms_scores, 'query')
-matchms_scores = append_spectrum_metadata(matchms_scores)
+matchms_scores = load_matchms_scores()
 
 matchms_scores_superclass = preprocess_data(normalize_df(matchms_scores.copy()), ["superclass"])
 larger_superclasses = matchms_scores_superclass.groupby("superclass").filter(lambda x: len(x) > 2)

diff --git a/analysis/Python_scripts/Fig5_classes_boxplots.py b/analysis/Python_scripts/Fig5_classes_boxplots.py
@@ -11,11 +11,7 @@
 from plotting import *
 
 
-matchms_scores = pd.read_csv("../data/output_matching/matchms/matchms_tol_0.0035_1%I_all_peaks_with_0s_only_matching.tsv", sep="\t")
-matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_scores': 'scores'}, inplace=True)
-matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_matches': 'matches'}, inplace=True)
-matchms_scores = append_classes(matchms_scores, 'query')
-matchms_scores = append_spectrum_metadata(matchms_scores)
+matchms_scores = load_matchms_scores()
 merged = normalize_df(matchms_scores.copy())
 
 scores_preprocessed_hierarchy = preprocess_data(merged, ["superclass", "class", "subclass"])

diff --git a/analysis/Python_scripts/Fig6_benzene_subclasses_boxplot.py b/analysis/Python_scripts/Fig6_benzene_subclasses_boxplot.py
@@ -9,11 +9,7 @@
 from plotting import *
 
 
-matchms_scores = pd.read_csv("../data/output_matching/matchms/matchms_tol_0.0035_1%I_all_peaks_with_0s_only_matching.tsv", sep="\t")
-matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_scores': 'scores'}, inplace=True)
-matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_matches': 'matches'}, inplace=True)
-matchms_scores = append_classes(matchms_scores, 'query')
-matchms_scores = append_spectrum_metadata(matchms_scores)
+matchms_scores = load_matchms_scores()
 merged = normalize_df(matchms_scores.copy())
 
 scores_preprocessed_hierarchy = preprocess_data(merged, ["superclass", "class", "subclass"])

diff --git a/analysis/Python_scripts/Fig7_nitrogen_comparison.py b/analysis/Python_scripts/Fig7_nitrogen_comparison.py
@@ -1,11 +1,8 @@
 from utils import *
 from plotting import boxplot_comparison
 
-all_peaks_same = pd.read_csv("../data/output_matching/matchms/matchms_tol_0.0035_1%I_all_peaks_with_0s_only_matching.tsv", sep='\t')
-all_peaks_same.rename(columns={'CosineHungarian_0.0035_0.0_1.0_scores': 'scores'}, inplace=True)
-all_peaks_same.rename(columns={'CosineHungarian_0.0035_0.0_1.0_matches': 'matches'}, inplace=True)
-all_peaks_same = append_spectrum_metadata(all_peaks_same)
-merged_all_peaks_same = normalize_df(append_classes(all_peaks_same, 'query'))
+matchms_scores = load_matchms_scores()
+merged_all_peaks_same = normalize_df(matchms_scores)
 mdf_comp = preprocess_data(merged_all_peaks_same, ["composition"])
 
 baseline_cols= ['C,H', 'C,H,O', 'C,H,O,S', 'C,Cl,H,O', 'Br,C,H,O', 'C,Cl,H', 'C,Cl,H,O,S', 'C,Cl,F,H,O', 'C,H,O,P', 'C,H,O,P,S']

diff --git a/analysis/Python_scripts/Fig8_p_and_s.py b/analysis/Python_scripts/Fig8_p_and_s.py
@@ -1,11 +1,8 @@
 from utils import *
 from plotting import create_plot
 
-all_peaks_same = pd.read_csv("../data/output_matching/matchms/matchms_tol_0.0035_1%I_all_peaks_with_0s_only_matching.tsv", sep='\t')
-all_peaks_same.rename(columns={'CosineHungarian_0.0035_0.0_1.0_scores': 'scores'}, inplace=True)
-all_peaks_same.rename(columns={'CosineHungarian_0.0035_0.0_1.0_matches': 'matches'}, inplace=True)
-all_peaks_same = append_spectrum_metadata(all_peaks_same)
-merged_all_peaks_same = normalize_df(append_classes(all_peaks_same, 'query'))
+matchms_scores = load_matchms_scores()
+merged_all_peaks_same = normalize_df(matchms_scores)
 mdf_comp = preprocess_data(merged_all_peaks_same, ["composition"])
 
 mdf_comp_ps = mdf_comp[mdf_comp['composition'].str.contains('S|P')]

diff --git a/analysis/Python_scripts/utils.py b/analysis/Python_scripts/utils.py
@@ -332,3 +332,12 @@ def normalize_df(df: pd.DataFrame, use_nist: bool = True, matches_norm_col: str
     if matches_norm_col:
         df[matches_col] = (df[matches_col] / df[matches_norm_col]) * 100
     return df
+
+
+def load_matchms_scores():
+    matchms_scores = pd.read_csv("../data/output_matching/matchms/matchms_tol_0.0035_1%I_all_peaks_with_0s_only_matching.tsv", sep="\t")
+    matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_scores': 'scores'}, inplace=True)
+    matchms_scores.rename(columns={'CosineHungarian_0.0035_0.0_1.0_matches': 'matches'}, inplace=True)
+    matchms_scores = append_classes(matchms_scores, 'query')
+    matchms_scores = append_spectrum_metadata(matchms_scores)
+    return matchms_scores