Changed the method of summing peptiede intensity to protein intensity…

…, changed the method "razor" to same as MaxQuant, and added a new method "rank".

Docs/ChangeLog.md

@@ -1,3 +1,8 @@
+# Version: 1.107.7
+## Date: 2024-06-24
+### Changes:
+- Change: Changed the method of summing peptiede intensity to protein intensity, changed the method "razor" to same as MaxQuant, and added a new method "rank".
+
 # Version: 1.107.6
 ## Date: 2024-06-19
 ### Changes:

Docs/MetaX_Cookbook.md

@@ -227,9 +227,16 @@ The Data Overview provides basic information about your data, such as the number
 
 Click **Create Proteins Intensity Table** to sum peptides to proteins if the Protein column is in the original table.
 
-- **Occam's Razor and Anti-Razor:** Methods available for inferring shared peptides. 
-  1. Build the rank of proteins.
-  2. Choose the protein with a higher rank for the shared peptide.
+- **Occam's Razor**, **Anti-Razor** and **Rank:** Methods available for inferring shared peptides.
+  - Razor:
+    1. Build a minimal set of proteins to cover all peptides (Set Cover Problem).
+    2. For each peptide, choose the protein which has most peptides (if multiple proteins have the same number of peptides, share intensity to them).
+  - Anti-Razor:
+    - All proteins are shared the intensity of each peptide.
+  - Rank:
+    1. Build the rank of proteins.
+    2. Choose the protein with a higher rank for the shared peptide.
+
 
 - **Methods to Build Protein Rank:**
   - unique_counts: Use the counts of proteins inferred by unique peptides.

utils/AnalyzerUtils/SumProteinIntensity.py

@@ -1,48 +1,52 @@
 # This file is used to sum the protein intensity for each sample
-# Method: razor or anti-razor
+# Method: razor, anti-razor or rank
 # By sample: True or False
 # Output: a dataframe with protein as index and sample as columns
 ############################################## 
 # USAGE:
 # from utils.AnalyzerUtils.SumProteinIntensity import SumProteinIntensity
 # out = SumProteinIntensity(sw)
-# df1 = out.sum_protein_intensity(method='razor', by_sample=False, rank_method='count')
-# df2 = out.sum_protein_intensity(method='razor', by_sample=False, rank_method='shared')
-# df3 = out.sum_protein_intensity(method='razor', by_sample=False, rank_method='unique')
+# df0 = out.sum_protein_intensity(method='razor')
+# df1 = out.sum_protein_intensity(method='rank', by_sample=False, rank_method='all_counts')
+# df2 = out.sum_protein_intensity(method='rank', by_sample=False, rank_method='shared_intensity')
+# df3 = out.sum_protein_intensity(method='rank', by_sample=False, rank_method='unique_counts')
 # df4 = out.sum_protein_intensity(method='anti-razor')
 ##############################################
 
+from collections import defaultdict
 import pandas as pd
+from tqdm import tqdm
+
 
 class SumProteinIntensity:
     def __init__(self, taxa_func_analyzer):
         self.tfa = taxa_func_analyzer
         self.res_intensity_dict = {} #store all sample to output
         self.rank_dict = {} #store the rank of protein intensity for each sample temporarily
-        self.rank_method = None
+        self.rank_method = None # only used for rank method
         self.extract_col_name = [self.tfa.peptide_col_name, self.tfa.protein_col_name] + self.tfa.sample_list
         self.df = self.tfa.original_df.loc[:,self.extract_col_name]
         self._init_dicts()
 
 
     def sum_protein_intensity(self, method='razor', by_sample=False, rank_method='unique_counts'):
 
-        if method not in ['razor', 'anti-razor']:
-            raise ValueError('Method must in ["razor", "anti-razor"]')
+        if method not in ['razor', 'anti-razor', 'rank']:
+            raise ValueError('Method must in ["razor", "anti-razor", "rank"]')
         if rank_method not in ['shared_intensity', 'all_counts', 'unique_counts', 'unique_intensity']:
             raise ValueError('Rank method must in ["shared_intensity", "all_counts", "unique_counts", "unique_intensity"]')
 
         self.rank_method = rank_method
 
-        if method == 'razor':
+        if method == 'rank':
             print(f"\n-------------Start to sum protein intensity using method: [{method}]  by_sample: [{by_sample}] rank_method: [{rank_method}]-------------")   
             # make a dict to count the intensity of each protein, intensity sahred by peptides will be divided by the number of peptides
             if by_sample:
                 for sample in self.tfa.sample_list:
                     # update the dict for each sample
                     print(f'Creating protein rank dict for [{sample}] by shared intensity', end='\r')
                     self._update_protein_rank_dict(sample_name = sample, rank_method = rank_method)
-                    self._sum_protein_razor(sample, by_sample)
+                    self._sum_protein_rank(sample, by_sample)
 
             else: # without sample
                 # only need to create the dict once
@@ -51,8 +55,12 @@ def sum_protein_intensity(self, method='razor', by_sample=False, rank_method='un
                 self._update_protein_rank_dict(sample_name = None, rank_method = rank_method)
 
                 for sample in self.tfa.sample_list:
-                    self._sum_protein_razor(sample, by_sample)
-
+                    self._sum_protein_rank(sample, by_sample)
+        elif method == 'razor':
+            print('start to sum protein intensity using method: [razor]')
+            # use Set Cover Problem to get the protein list, then sum the intensity
+            pep_to_protein = self._create_pep_to_protein_razor()
+            self._sum_protein_razor(pep_to_protein)
 
         elif method == 'anti-razor':
             print(f"\n-------------Start to sum protein intensity using method: [{method}]  by_sample: [True] rank_method: [Shared]-------------")    
@@ -77,6 +85,82 @@ def sum_protein_intensity(self, method='razor', by_sample=False, rank_method='un
         return res_df
 
 
+    def _create_pep_to_protein_razor(self) -> dict:
+        """
+        Create a dictionary mapping peptides to proteins based on a minimum protein set.
+
+        Returns:
+            dict: A dictionary mapping peptides to proteins.
+            key: peptide
+            value: a list of proteins
+        """
+        # crate a function to find the minimum protein set
+        def find_minimum_protein_set(peptides, protein_to_peptides):
+            print('Start to create protein dict using "Set Cover Problem"')
+            peptides_to_cover = set(peptides)
+            selected_proteins = set()
+
+            with tqdm(total=len(peptides_to_cover), desc="Covering peptides") as pbar:
+                while peptides_to_cover:
+                    best_protein = None
+                    peptides_covered_by_best = set()
+                    for protein, covered_peptides in protein_to_peptides.items():
+                        covered = peptides_to_cover & covered_peptides
+                        if len(covered) > len(peptides_covered_by_best):
+                            best_protein = protein
+                            peptides_covered_by_best = covered
+
+                    if not best_protein:
+                        break
+
+                    selected_proteins.add(best_protein)
+                    peptides_to_cover -= peptides_covered_by_best
+                    pbar.update(len(peptides_covered_by_best))
+
+            return selected_proteins
+
+        df = self.df.loc[:, [self.tfa.peptide_col_name, self.tfa.protein_col_name]]
+        # Create a dictionary mapping proteins to peptides
+        protein_to_peptides = defaultdict(set)
+        peptides = set(df[self.tfa.peptide_col_name])
+
+        for _, row in tqdm(df.iterrows(), total=df.shape[0], desc="Creating protein to peptides mapping"):
+            sequence = row[self.tfa.peptide_col_name]
+            proteins = row[self.tfa.protein_col_name].split(';')
+            for protein in proteins:
+                protein_to_peptides[protein].add(sequence)
+
+        mini_protein_set = find_minimum_protein_set(peptides, protein_to_peptides)
+
+        # remove the proteins not in the mini_protein_set from the protein_to_peptides
+        filtered_protein_to_peptides = {protein: protein_to_peptides[protein] for protein in mini_protein_set}
+        # Assign each peptide to the protein that contains it with the highest peptide count
+        print('Assigning peptides to proteins')
+        peptide_to_protein = defaultdict(list)
+        for peptide in tqdm(peptides, desc="Assigning peptides to proteins"):
+            possible_proteins = [protein for protein, peps in filtered_protein_to_peptides.items() if peptide in peps]
+            if possible_proteins:
+                # 找到包含该肽最多的蛋白质
+                max_protein_count = max(len(filtered_protein_to_peptides[protein]) for protein in possible_proteins)
+                best_proteins = [protein for protein in possible_proteins if len(filtered_protein_to_peptides[protein]) == max_protein_count]
+                peptide_to_protein[peptide].extend(best_proteins)
+
+        return peptide_to_protein
+
+    def _sum_protein_razor(self, peptide_to_protein: dict):
+
+        for sample in tqdm(self.tfa.sample_list):
+            print(f'Assigning protein intensity for [{sample}]')
+            df = self.df.loc[:,[ self.tfa.peptide_col_name, sample]]
+            # create a dict to store the intensity of each peptide
+            df.set_index(self.tfa.peptide_col_name, inplace=True)
+            peptide_intensity_dict = df.to_dict()[sample]
+            for peptide, proteins in peptide_to_protein.items():
+                intensity = peptide_intensity_dict[peptide]
+                self._update_output_dict(proteins, sample, intensity)
+
+
+
     def _init_dicts(self):
         for sample in self.tfa.sample_list:
             self.res_intensity_dict[sample] = {}
@@ -147,7 +231,7 @@ def _update_output_dict(self, protein_list: list, sample_name:str, intensity:flo
                     self.res_intensity_dict[sample_name][protein] = intensity
 
 
-    def _sum_protein_razor(self, sample_name:str, by_sample=False):
+    def _sum_protein_rank(self, sample_name:str, by_sample=False):
         # print in one line
         print(f'Asigning protein intensity for [{sample_name}]', end='\r')
         df = self.df.loc[:,[ self.tfa.protein_col_name, sample_name]]
@@ -180,7 +264,4 @@ def _sum_protein_anti_razor(self, sample_name:str):
         for row in df.itertuples():
             proteins = row[1].split(';')
             intensity = row[2]
-            self._update_output_dict(proteins, sample_name, intensity)
-
-
-
+            self._update_output_dict(proteins, sample_name, intensity)

utils/GUI.py

@@ -831,21 +831,21 @@ def change_theme(self, theme, silent=False):
 
     def change_event_checkBox_create_protein_table(self):
         if self.checkBox_create_protein_table.isChecked():
-            self.checkBox_infrence_protein_by_sample.setEnabled(True)
-            self.comboBox_protein_ranking_method.setEnabled(True)
+            # self.checkBox_infrence_protein_by_sample.setEnabled(True)
+            # self.comboBox_protein_ranking_method.setEnabled(True)
             self.comboBox_method_of_protein_inference.setEnabled(True)
         else:
             self.comboBox_method_of_protein_inference.setEnabled(False)
             self.checkBox_infrence_protein_by_sample.setEnabled(False)
             self.comboBox_protein_ranking_method.setEnabled(False)
 
     def update_method_of_protein_inference(self):
-        if self.comboBox_method_of_protein_inference.currentText() == "anti-razor":
+        if self.comboBox_method_of_protein_inference.currentText() in ["razor", "anti-razor"]:
             # set checked 
             self.checkBox_infrence_protein_by_sample.setChecked(True)
             self.checkBox_infrence_protein_by_sample.setEnabled(False)
             self.comboBox_protein_ranking_method.setEnabled(False)
-        else:
+        else: # method is ["rank"]
             self.checkBox_infrence_protein_by_sample.setEnabled(True)
             self.comboBox_protein_ranking_method.setEnabled(True)
             self.checkBox_infrence_protein_by_sample.setChecked(False)

utils/MetaX_GUI/MainWindow.ui

@@ -46,7 +46,7 @@
            <enum>Qt::LeftToRight</enum>
           </property>
           <property name="currentIndex">
-           <number>6</number>
+           <number>2</number>
           </property>
           <widget class="QWidget" name="tab">
            <attribute name="title">
@@ -798,6 +798,11 @@
                   <string>anti-razor</string>
                  </property>
                 </item>
+                <item>
+                 <property name="text">
+                  <string>rank</string>
+                 </property>
+                </item>
                </widget>
               </item>
               <item row="1" column="2">
@@ -1315,7 +1320,7 @@
             <item row="1" column="0">
              <widget class="QTabWidget" name="tabWidget_4">
               <property name="currentIndex">
-               <number>1</number>
+               <number>0</number>
               </property>
               <widget class="QWidget" name="tab_12">
                <attribute name="title">
@@ -8026,7 +8031,7 @@
      <x>0</x>
      <y>0</y>
      <width>1059</width>
-     <height>21</height>
+     <height>23</height>
     </rect>
    </property>
    <widget class="QMenu" name="menuTools">

utils/MetaX_GUI/Ui_MainWindow.py

@@ -1,6 +1,6 @@
 # -*- coding: utf-8 -*-
 
-# Form implementation generated from reading ui file 'c:\Users\Qing\OneDrive - University of Ottawa\code\TaxaFunc\MetaX\utils\MetaX_GUI\MainWindow.ui'
+# Form implementation generated from reading ui file 'c:\Users\max\OneDrive - University of Ottawa\code\TaxaFunc\MetaX\utils\MetaX_GUI\MainWindow.ui'
 #
 # Created by: PyQt5 UI code generator 5.15.9
 #
@@ -438,6 +438,7 @@ def setupUi(self, metaX_main):
         self.comboBox_method_of_protein_inference.setObjectName("comboBox_method_of_protein_inference")
         self.comboBox_method_of_protein_inference.addItem("")
         self.comboBox_method_of_protein_inference.addItem("")
+        self.comboBox_method_of_protein_inference.addItem("")
         self.gridLayout_37.addWidget(self.comboBox_method_of_protein_inference, 0, 2, 1, 1)
         self.horizontalLayout_2 = QtWidgets.QHBoxLayout()
         self.horizontalLayout_2.setObjectName("horizontalLayout_2")
@@ -4208,7 +4209,7 @@ def setupUi(self, metaX_main):
         self.statusbar.setObjectName("statusbar")
         metaX_main.setStatusBar(self.statusbar)
         self.menuBar = QtWidgets.QMenuBar(metaX_main)
-        self.menuBar.setGeometry(QtCore.QRect(0, 0, 1059, 21))
+        self.menuBar.setGeometry(QtCore.QRect(0, 0, 1059, 23))
         self.menuBar.setObjectName("menuBar")
         self.menuTools = QtWidgets.QMenu(self.menuBar)
         self.menuTools.setObjectName("menuTools")
@@ -4268,9 +4269,9 @@ def setupUi(self, metaX_main):
 
         self.retranslateUi(metaX_main)
         self.stackedWidget.setCurrentIndex(0)
-        self.tabWidget_TaxaFuncAnalyzer.setCurrentIndex(6)
+        self.tabWidget_TaxaFuncAnalyzer.setCurrentIndex(2)
         self.toolBox_2.setCurrentIndex(0)
-        self.tabWidget_4.setCurrentIndex(1)
+        self.tabWidget_4.setCurrentIndex(0)
         self.tabWidget_3.setCurrentIndex(0)
         self.tabWidget.setCurrentIndex(1)
         self.tabWidget_2.setCurrentIndex(1)
@@ -4369,6 +4370,7 @@ def retranslateUi(self, metaX_main):
         self.checkBox_infrence_protein_by_sample.setText(_translate("metaX_main", "Inference by each Sample"))
         self.comboBox_method_of_protein_inference.setItemText(0, _translate("metaX_main", "razor"))
         self.comboBox_method_of_protein_inference.setItemText(1, _translate("metaX_main", "anti-razor"))
+        self.comboBox_method_of_protein_inference.setItemText(2, _translate("metaX_main", "rank"))
         self.label_136.setText(_translate("metaX_main", "Protein Ranking Method"))
         self.comboBox_protein_ranking_method.setItemText(0, _translate("metaX_main", "unique_counts"))
         self.comboBox_protein_ranking_method.setItemText(1, _translate("metaX_main", "all_counts"))

utils/version.py

@@ -1,2 +1,2 @@
-__version__ = '1.107.6'
+__version__ = '1.107.7'
 API_version = '1'