Merge pull request #39 from RECETOX/fontsize

Updated plotting library

analysis/Python_scripts/Fig3_correlations.py

@@ -5,6 +5,7 @@
 from plotting import *
 
 matchms_scores = load_matchms_scores()
+_, _, label_fontsize, tick_fontsize, text_width, _ = init()
 
 df = normalize_df(matchms_scores, matches_norm_col=None)
 del df['peak_comments']
@@ -36,12 +37,20 @@
 corr = df[properties].corr().round(2)
 
 plt.figure(figsize=(24, 20))
-cax = sns.heatmap(corr, annot=True, cmap='coolwarm', center=0, vmin=-1, vmax=1,annot_kws={"size": 20})
+cax = sns.heatmap(
+    corr,
+    annot=True,
+    cmap='coolwarm',
+    center=0,
+    vmin=-1,
+    vmax=1,
+    annot_kws={"size": label_fontsize}
+)
 # plt.title('Pearson Correlations')
-plt.tick_params(axis='both', which='major', labelsize=20)
+plt.tick_params(axis='both', which='major', labelsize=tick_fontsize)
 # Get the colorbar from the HeatMap and set the fontsize for its tick labels
 cbar = cax.collections[0].colorbar
-cbar.ax.tick_params(labelsize=20)
+cbar.ax.tick_params(labelsize=tick_fontsize)
 
 plt.savefig("paper_plots/Fig3_correlations.png", bbox_inches='tight')
 # plot name in the manuscript:

analysis/Python_scripts/Fig4_superclass_histograms.py

@@ -1,15 +1,11 @@
 import pandas as pd
-import os
-import numpy as np
-import math
 from matplotlib import pyplot as plt
-from rdkit import Chem
-import plotly.graph_objs as go
-from plotly.subplots import make_subplots
 
 from utils import *
 from plotting import *
 
+_, _, label_fontsize, tick_fontsize, text_width, _ = init()
+
 matchms_scores = load_matchms_scores()
 
 matchms_scores_superclass = preprocess_data(normalize_df(matchms_scores.copy()), ["superclass"])
@@ -21,11 +17,13 @@
 plt.clf()
 plt.set_cmap('viridis')
 plt.hist2d(matches_normalized * 100, matchms_scores['scores'] * 1000, bins=(5, 5), range=[[0, 100], [0, 1000]])
-plt.colorbar()
+cbar = plt.colorbar()
+cbar.ax.tick_params(labelsize=tick_fontsize)
+
 plt.clim(0, 70)
-plt.xlabel('ions matching reference (%)', fontsize=20)
-plt.ylabel('scores', fontsize=20)
-plt.tick_params(labelsize=13)
+plt.xlabel('ions matching reference (%)', fontsize=label_fontsize)
+plt.ylabel('scores', fontsize=label_fontsize)
+plt.tick_params(labelsize=tick_fontsize)
 plt.gcf().set_size_inches(8, 6)
 plt.savefig("paper_plots/Fig4a_superclasses_histogram.png", bbox_inches='tight')
 
@@ -43,12 +41,13 @@
 
 plt.clf()
 plt.set_cmap('viridis')
-plt.hist2d(matchms_scores_top5['matches'], matchms_scores_top5['scores'] * 1000, bins=([0,1,2,3,4,5], 5))
-plt.colorbar()
+plt.hist2d(matchms_scores_top5['matches'], matchms_scores_top5['scores'] * 1000, bins=([0,1,2,3,4,5], 5), range=[[0, 5], [0, 1000]])
+cbar = plt.colorbar()
+cbar.ax.tick_params(labelsize=tick_fontsize)
 plt.clim(0, 70)
 
-plt.xlabel('ion matches', fontsize=20)
-plt.ylabel('scores', fontsize=20)
-plt.tick_params(labelsize=13)
+plt.xlabel('ion matches', fontsize=label_fontsize)
+plt.ylabel('scores', fontsize=label_fontsize)
+plt.tick_params(labelsize=tick_fontsize)
 plt.gcf().set_size_inches(8, 6)
 plt.savefig("paper_plots/Fig4b_superclasses_histogram.png", bbox_inches='tight')

analysis/Python_scripts/Fig6_benzene_subclasses_boxplot.py

@@ -19,5 +19,5 @@
 for group in grouped_class.groups:
     grp = grouped_class.get_group(group).groupby(grouping).filter(lambda x: len(x) > 6)
     if len(grp) > 0 and group == "Benzene and substituted derivatives":
-        fig = create_plot(grp, grouping, showlegend=False, hide_labels=True)
+        fig = create_plot(grp, grouping, showlegend=False, hide_labels=False)
         fig.savefig(f"paper_plots/Fig6_benzene_subclasses.png", bbox_inches='tight')

analysis/Python_scripts/Fig9_comparisons.py

@@ -0,0 +1,46 @@
+import pandas as pd
+
+from utils import *
+from plotting import make_simple_boxplot
+
+wang_db = pd.read_csv("../data/reference/13321_2020_470_MOESM1_ESM.csv")
+wang_db_tms = pd.read_csv("../data/reference/wang2022_TMS_annotated.tsv", sep="\t", decimal=",")
+df = normalize_df(load_matchms_scores())
+
+wang_db["study"] = "Wang et al.$^{27}$"
+wang_db_tms["study"] = "Wang et al.$^{28}$"
+df["study"] = "This study"
+cols = ["study", "Dot"]
+
+combined = pd.concat([df.rename(columns={"scores": "Dot"})[cols], wang_db[cols], wang_db_tms[cols]])
+
+make_simple_boxplot(combined, "study", "Dot", text_width=60).savefig("paper_plots/Fig9a_studies.png", bbox_inches='tight')
+
+
+df_wang2020_db_with_classes = load_sdf_into_df("../data/reference/wang2020/3411.sdf")
+wang_db_merged = df_wang2020_db_with_classes.set_index("inchikey").join(wang_db.set_index("Inchikey "))[["superclass","class","subclass","Dot"]].dropna()
+order = [
+    "Lipids and lipid-like molecules",
+    "Organoheterocyclic compounds",
+    "Organic oxygen compounds",
+    "Organic acids and derivatives",
+    "Organic nitrogen compounds",
+    "Hydrocarbons"
+]
+
+make_simple_boxplot(wang_db_merged, x="superclass", y="Dot", order=order).savefig("paper_plots/Fig9b_wang2020.png", bbox_inches='tight')
+
+df_wang2022_db_tms_with_classes = load_sdf_into_df("../data/reference/wang2022/3412.sdf")
+wang_db_tms_merged = df_wang2022_db_tms_with_classes.set_index("inchikey").join(wang_db_tms.set_index("inchikey")["Dot"])[["superclass","class","subclass","Dot"]].dropna()
+order = [
+    "Benzenoids",
+    "Lipids and lipid-like molecules",
+    "Organoheterocyclic compounds",
+    "Phenylpropanoids and polyketides",
+    "Organic oxygen compounds",
+    "Organic acids and derivatives",
+    "Organic nitrogen compounds",
+    "Organosulfur compounds",
+    "Organometallic compounds"
+]
+make_simple_boxplot(wang_db_tms_merged, x="superclass", y="Dot", order = order).savefig("paper_plots/Fig9c_wang2022.png", bbox_inches='tight')