working color peptides 51mer

Author: Evelyn Schmidt

scripts/color_peptides51mer.py

@@ -0,0 +1,343 @@
+import numpy as np
+import pandas as pd
+import re
+from bs4 import BeautifulSoup
+import argparse
+
+
+class AminoAcid:
+
+    def __init__(self, nucleotide, bold, color, underline, large, position, open_tag, close_tag):
+        self.nucleotide = nucleotide
+        self.bold = bold
+        self.color = color
+        self.underline = underline
+        self.large = large
+        self.position = position 
+        self.open_tag = open_tag
+        self.close_tag = close_tag
+
+# ---- PARSE ARGUMENTS -------------------------------------------------------
+# Parses command line arguments
+# Enables user help
+def parse_arguments():
+    # Parse command line arugments
+    parser = argparse.ArgumentParser(description='Color the 51mer peptide')
+
+    parser.add_argument('-p',
+                        help='The path to the Peptides 51 mer', required=True)
+    parser.add_argument('-classI',
+                        help='The path to the classI all_epitopes.aggregated.tsv used in pVACseq', required=True)
+    parser.add_argument('-classII',
+                        help='The path to the classII all_epitopes.aggregated.tsv used in pVACseq', required=True)
+
+    parser.add_argument('-o', help="Output location", required=True)
+
+    return(parser.parse_args())
+
+    # Function to rearrange string so that G518D looks like 518G/D
+def rearrange_string(s):
+    match = re.match(r'([A-Za-z]+)([\d-]+)([A-Za-z]*)', s)
+    if match:
+        letters_before = match.group(1)
+        numbers = match.group(2)
+        letters_after = match.group(3)
+            
+            #return f"{numbers}{letters_before}/{letters_after}"
+            # Just use the postion for the key to avoid FS problem
+        return f"{numbers}"
+    else:
+        return s
+        
+
+def annotate_every_nucleotide(sequence, classI_peptide, classII_peptide):
+
+    peptide_sequence = []
+
+    # Make the sequence a list of AminoAcid objects
+    for i in range(len(sequence)):
+        new_AA = AminoAcid(sequence[i], False, False, False, False, -1, False, False)
+
+        if sequence[i] == 'C':
+            new_AA.large = True
+        
+        peptide_sequence.append(new_AA)
+
+    # CLASS I
+    positions = []
+    # Get the positions in the peptide_sequence where the classI is located
+    for i in range(len(peptide_sequence)):
+        for j in range(len(classI_peptide)):
+            if peptide_sequence[i].nucleotide == classI_peptide[j]:
+                positions.append(i)
+                i+=1
+            else:
+                break
+
+        if len(positions) == len(classI_peptide):
+            break
+        else:
+            positions = []
+            
+    # set those positions to red
+    j = 0
+    for i in range(len(peptide_sequence)):
+        if j < len(positions) and i == positions[j]:
+            peptide_sequence[i].color = True
+            j+=1
+
+    # CLASS II
+    positions = []
+    for i in range(len(peptide_sequence)):
+        for j in range(len(classII_peptide)):
+            if peptide_sequence[i].nucleotide == classII_peptide[j]:
+                positions.append(i)
+                i+=1
+            else:
+                break
+
+        if len(positions) == len(classII_peptide):
+            break
+        else:
+            positions = []
+
+    j = 0
+    for i in range(len(peptide_sequence)):
+        if j < len(positions) and i == positions[j]:
+            peptide_sequence[i].bold = True
+            j+=1
+    
+
+    return(peptide_sequence)
+
+def set_underline(peptide_sequence, mutant_peptide_pos):
+
+    frameshift = False
+    classI_position = 0
+
+    if '-' in mutant_peptide_pos:
+        positions = mutant_peptide_pos.split("-")
+
+        start_position = int(positions[0])
+        end_position = int(positions[1])
+
+        frameshift = True
+
+    else:
+        mutant_peptide_pos = int(mutant_peptide_pos)
+
+    if frameshift:
+
+        continue_underline = False
+        
+        for i in range(len(peptide_sequence)):
+
+            if peptide_sequence[i].color:
+                classI_position += 1
+            else:
+                 classI_position = 0
+                 continue_underline = False
+
+            if classI_position == start_position:
+                peptide_sequence[i].underline = True
+                continue_underline = True
+            elif continue_underline:
+                peptide_sequence[i].underline = True
+            elif classI_position == end_position:
+                peptide_sequence[i].underline = True
+                continue_underline = False
+            i+=1
+    else:
+        for i in range(len(peptide_sequence)):
+
+            if peptide_sequence[i].color:
+                classI_position += 1
+            else:
+                 classI_position = 0
+
+            if classI_position == int(mutant_peptide_pos):
+                peptide_sequence[i].underline = True
+            i+=1
+        
+def set_span_tags(peptide_sequence):
+
+    currently_bold = False
+    currently_red = False
+    currently_underlined = False
+    currently_large = False
+    inside_span = False
+
+    for nucleotide in peptide_sequence:
+
+        if currently_bold != nucleotide.bold or currently_red != nucleotide.color or currently_underlined != nucleotide.underline or currently_large != nucleotide.large:
+            
+            nucleotide.open_tag = True
+
+            if inside_span:
+                nucleotide.close_tag = True # only if its isnide a span tag
+            else:
+                nucleotide.close_tag = False
+
+
+            currently_bold = nucleotide.bold 
+            currently_red = nucleotide.color
+            currently_underlined = nucleotide.underline
+            currently_large = nucleotide.large
+
+            inside_span = True
+        
+    return(peptide_sequence)
+
+def create_stylized_sequence(peptide_sequence):
+
+    new_string = ''
+
+    for nucleotide in peptide_sequence:
+
+        if nucleotide.open_tag or nucleotide.close_tag:
+            if nucleotide.close_tag:
+                new_string += '</span>'
+
+            if nucleotide.open_tag:
+
+                if nucleotide.large: # we are assuming that a cystine is never in the classI and classIi
+                    new_string += '<span style="font-size:105%">'
+                    new_string += nucleotide.nucleotide
+
+                if nucleotide.bold and nucleotide.color and nucleotide.underline:
+                    new_string += '<span style="font-weight:bold;color:#ff0000;text-decoration:underline;">'
+                    new_string += nucleotide.nucleotide
+                elif nucleotide.bold and not nucleotide.color and not nucleotide.underline:
+                    new_string += '<span style="font-weight:bold;">'
+                    new_string += nucleotide.nucleotide
+                elif not nucleotide.bold and nucleotide.color and not nucleotide.underline:
+                    new_string += '<span style="color:#ff0000;">'
+                    new_string += nucleotide.nucleotide
+                elif not nucleotide.bold and not nucleotide.color and nucleotide.underline:
+                    new_string += '<span style="text-decoration:underline;">'
+                    new_string += nucleotide.nucleotide
+                elif nucleotide.bold and nucleotide.color and not nucleotide.underline:
+                    new_string += '<span style="font-weight:bold;color:#ff0000;">'
+                    new_string += nucleotide.nucleotide
+                elif not nucleotide.bold and nucleotide.color and nucleotide.underline:
+                    new_string += '<span style="color:#ff0000;text-decoration:underline;">'
+                    new_string += nucleotide.nucleotide
+                elif nucleotide.bold and not nucleotide.color and nucleotide.underline:
+                    new_string += '<span style="font-weight:bold;text-decoration:underline;">'
+                    new_string += nucleotide.nucleotide
+        else:
+            new_string += nucleotide.nucleotide
+
+    return(new_string)   
+
+def main():
+    args = parse_arguments()
+
+    # read in classI and class II
+    #peptides_51mer = pd.read_excel("/Volumes/mgriffit/Active/griffithlab/gc2596/e.schmidt/neoag_vaccine_scripts/scripts/data_files/10146-0021_Peptides_51-mer.xlsx")
+    #classI = pd.read_csv("/Volumes/mgriffit/Active/griffithlab/gc2596/e.schmidt/neoag_vaccine_scripts/scripts/data_files/classI.TWJF-10146-0021-Tumor_Lysate.all_epitopes.aggregated.tsv", sep="\t")
+    #classII = pd.read_csv("/Volumes/mgriffit/Active/griffithlab/gc2596/e.schmidt/neoag_vaccine_scripts/scripts/data_files/classII.TWJF-10146-0021-Tumor_Lysate.all_epitopes.aggregated.tsv", sep="\t")
+
+    peptides_51mer = pd.read_excel(args.p)
+    classI = pd.read_csv(args.classI, sep="\t")
+    classII = pd.read_csv(args.classII, sep="\t")
+
+    # Create a universal ID by editing the peptide 51mer ID
+    peptides_51mer.rename(columns={'ID': 'full ID'}, inplace=True)
+    peptides_51mer['ID'] = peptides_51mer['full ID']
+
+    peptides_51mer['ID'] = peptides_51mer['ID'].apply(lambda x: '.'.join(x.split('.')[1:]))  # Removing before first period, periods will be removed 
+    peptides_51mer['ID'] = peptides_51mer['ID'].apply(lambda x: '.'.join(x.split('.')[1:]))  # Removing before second period
+    peptides_51mer['ID'] = peptides_51mer['ID'].apply(lambda x: '.'.join(x.split('.')[:3]) + '.' + '.'.join(x.split('.')[4:]))
+
+    for index, row in peptides_51mer.iterrows():
+        for i, char in enumerate(row['ID'][::-1]):
+            if char.isdigit():
+                peptides_51mer.at[index, 'ID'] = row['ID'][:-i]
+                break
+        else:
+            result = row['ID']
+
+    # create a key that is gene, transcript, AA change for CLASSI
+    classII['modified AA Change'] = classII['AA Change'] 
+
+    # Apply the function to the 'Value' column
+    classII['modified AA Change'] = classII['modified AA Change'].apply(rearrange_string)
+
+    classII['ID'] = classII['Gene'] + '.' + classII['Best Transcript'] + '.' + classII['modified AA Change'] 
+
+    # create a key that is gene, transcript, AA change for CLASSI
+    classI['modified AA Change'] = classI['AA Change'] 
+
+    # Apply the function to the 'Value' column
+    classI['modified AA Change'] = classI['modified AA Change'].apply(rearrange_string)
+
+    classI['ID'] = classI['Gene'] + '.' + classI['Best Transcript'] + '.' + classI['modified AA Change'] 
+
+    # Merge the sequences from classI and classII with peptide 51mer
+    merged_peptide_51mer = pd.merge(peptides_51mer, classII[['ID', 'Best Peptide']], on='ID', how='left')
+
+    merged_peptide_51mer.rename(columns = {"Best Peptide":"Best Peptide Class II"}, inplace=True)
+
+    merged_peptide_51mer = pd.merge(merged_peptide_51mer, classI[['ID', 'Best Peptide', 'Pos']], on='ID', how='left')
+
+    merged_peptide_51mer.rename(columns = {"Best Peptide":"Best Peptide Class I"}, inplace=True)
+
+    # convert peptide 51mer to HTML
+    peptides_51mer_html = peptides_51mer.to_html(index=False) # convert to html
+
+    # Creating a BeautifulSoup object and specifying the parser
+    peptides_51mer_soup = BeautifulSoup(peptides_51mer_html, 'html.parser')
+
+
+    for index, row in peptides_51mer.iterrows():
+
+        search_string = row['full ID']
+
+        #classII_sequence 
+        classII_peptide = merged_peptide_51mer.loc[merged_peptide_51mer['full ID'] == search_string, 'Best Peptide Class II'].values[0]
+        #classI_sequence 
+        classI_peptide = merged_peptide_51mer.loc[merged_peptide_51mer['full ID'] == search_string, 'Best Peptide Class I'].values[0]
+        # mutant pepetide position --- not working yet becasue of STUPID frameshift
+        mutant_peptide_pos = str(merged_peptide_51mer.loc[merged_peptide_51mer['full ID'] == search_string, 'Pos'].values[0])
+
+        # Find the tag containing the search string
+        tag_with_search_string = peptides_51mer_soup.find('td', string=search_string)
+
+        if tag_with_search_string and isinstance(classII_peptide, str):
+
+            # Find the parent <tr> tag of the tag containing the search string
+            parent_tr = tag_with_search_string.find_parent('tr')    
+            # Find the next two <td> tags
+            next_td_tags = parent_tr.findChildren('td', limit=3)
+            
+            sequence = next_td_tags[2].get_text()
+
+            # make sequence the list of objects
+            peptide_sequence = annotate_every_nucleotide(sequence, classI_peptide, classII_peptide)
+
+            # actaully lets break class I and classII into two steps and handle the mutated nucleotide in class I function
+            # it should be basically like at that position in the class I set 
+
+            set_underline(peptide_sequence, mutant_peptide_pos)
+
+            set_span_tags(peptide_sequence) # pass by reference
+            
+            new_string = create_stylized_sequence(peptide_sequence)
+
+            print(new_string)
+
+            next_td_tags[2].string = new_string
+
+            modified_html = peptides_51mer_soup.prettify(formatter=None)
+
+        else:
+            print("Search string not found.")
+
+    with open(args.o, "w", encoding = 'utf-8') as file:
+        file.write(modified_html)
+
+
+
+if __name__ == "__main__":
+    main()