PTM Stoichiometry

mzLib/Omics/SpectrumMatch/SpectrumMatchFromTsv.cs

@@ -107,27 +107,32 @@ public static Dictionary<int, List<string>> ParseModifications(string fullSeq)
             //int patternMatches = regex.Matches(fullSeq).Count;
             Dictionary<int, List<string>> modDict = new();
 
+
+            // If there is a missed cleavage, then there will be a label on K and a Label on X modification.
+            // It'll be like [label]|[label] which complicates the positional stuff a little bit. Therefore, 
+            // RemoveSpecialCharacters will remove the "|", to ease things later on. 
             RemoveSpecialCharacters(ref fullSeq);
             MatchCollection matches = regex.Matches(fullSeq);
-            int currentPosition = 0;
+            int captureLengthSum = 0; 
             foreach (Match match in matches)
             {
                 GroupCollection group = match.Groups;
                 string val = group[1].Value;
                 int startIndex = group[0].Index;
                 int captureLength = group[0].Length;
-                int position = group["(.+?)"].Index;
 
                 List<string> modList = new List<string>();
                 modList.Add(val);
+
+                // The position of the amino acids is tracked by the positionToAddToDict variable. It takes the 
+                // startIndex of the modification Match and removes the cumulative length of the modifications
+                // found (including the brackets). The difference will be the number of nonmodification characters, 
+                // or the number of amino acids prior to the startIndex in the sequence. 
+                int positionToAddToDict = startIndex - captureLengthSum;
+
                 // check to see if key already exist
-                // if there is a missed cleavage, then there will be a label on K and a Label on X modification.
-                // And, it'll be like [label]|[label] which complicates the positional stuff a little bit.
                 // if the already key exists, update the current position with the capture length + 1.
                 // otherwise, add the modification to the dict.
-
-                // int to add is startIndex - current position
-                int positionToAddToDict = startIndex - currentPosition;
                 if (modDict.ContainsKey(positionToAddToDict))
                 {
                     modDict[positionToAddToDict].Add(val);
@@ -136,7 +141,7 @@ public static Dictionary<int, List<string>> ParseModifications(string fullSeq)
                 {
                     modDict.Add(positionToAddToDict, modList);
                 }
-                currentPosition += startIndex + captureLength;
+                captureLengthSum += captureLength;
             }
             return modDict;
         }

mzLib/Test/FileReadingTests/TestPsmFromTsv.cs

@@ -10,6 +10,21 @@
 using Omics.SpectrumMatch;
 using Proteomics;
 using Readers;
+using UsefulProteomicsDatabases;
+using MathNet.Numerics.Distributions;
+using static UsefulProteomicsDatabases.ProteinDbRetriever;
+using System.Windows.Input;
+using Readers.QuantificationResults;
+using Easy.Common.Extensions;
+using MzIdentML;
+using Proteomics.ProteolyticDigestion;
+using Omics.Modifications;
+using Proteomics.AminoAcidPolymer;
+using FlashLFQ;
+using System.Printing.IndexedProperties;
+using pepXML.Generated;
+using System.Diagnostics.Eventing.Reader;
+using Omics;
 
 namespace Test.FileReadingTests
 {
@@ -76,7 +91,104 @@ public static void ReadOGlycoPsmsLocalizedGlycans()
             }
 
             Assert.AreEqual(1, localGlycans.Count);
+        }
+        [Test]
+        public static void Peter()
+        {
+            // load database file and quantified peak file. 
+            string dbFilePath1 = @"C:\Users\student\Downloads\MetaMorpheusVignette\MetaMorpheusVignette\uniprot-mouse-reviewed-1-24-2018.xml.gz";
+            string dbFilePath2 = @"C:\Users\student\Downloads\MetaMorpheusVignette\MetaMorpheusVignette\uniprot-cRAP-1-24-2018.xml.gz";
+            string quantifiedPeakFilePath = @"C:\Users\student\Downloads\MetaMorpheusVignette\MetaMorpheusVignette\2024-07-25-11-27-37\Task1-SearchTask\AllQuantifiedPeaks.tsv";
+            string quantifiedProteinGroupsFilePath = @"C:\Users\student\Downloads\MetaMorpheusVignette\MetaMorpheusVignette\2024-07-25-11-27-37\Task1-SearchTask\AllQuantifiedPeaks.tsv";
+
+
+            List<Protein> proteinDb = ProteinDbLoader.LoadProteinXML(dbFilePath1, true, DecoyType.None, null, false, null, out var a);
+            proteinDb.AddRange(ProteinDbLoader.LoadProteinXML(dbFilePath2, true, DecoyType.None, null, false, null, out var b));
+            var proteinDbDict = proteinDb.ToEasyDictionary(n => n.Accession);
+
+            var peaksFile = new QuantifiedPeakFile(quantifiedPeakFilePath);
+            peaksFile.LoadResults();
+
+            var ProteinGroupsFile = new SpectrumMatchFromTsvFile(quantifiedProteinGroupsFilePath);
+            ProteinGroupsFile.LoadResults();
+
+            var occupancyDict = new Dictionary<String, Dictionary<int, Dictionary<string, double>>>();
+            var proteinSeqRangesSeen = new Dictionary<String, List<Tuple<int, int, double>>>();
+            var proteinsNotInDb = new List<string>();
+
+            // go through the quantified peaks 
+            foreach (var peak in peaksFile)
+            {
+                // if ambiguous peptides for a peak, go through each peptide
+                var peptidesFull = peak.FullSequence.Split('|');
+                List<string> peptidesBase = new List<string>();
+                foreach (string peptide in peptidesFull) 
+                    peptidesBase.Add(IBioPolymerWithSetMods.GetBaseSequenceFromFullSequence(peptide)); 
+
+                for (int i = 0; i < peptidesFull.Length; i++)
+                {
+                    var peptideMods = Omics.SpectrumMatch.SpectrumMatchFromTsv.ParseModifications(peptidesFull[i]);
+
+                    // go through each protein of the protein group
+                    foreach (string protein in peak.ProteinGroup.SplitAndTrim(new char[] { '|', ';' }))
+                    {
+                        int offset = 0;
+                        // try setting offset to the peptide start residue index, and catch when the protein is not in the database
+                        try
+                        {
+                            offset = proteinDbDict[protein].BaseSequence.IndexOf(peptidesBase[i]);
+
+                            // if the peptide is not a substring of the protein sequence, move on to the next protein
+                            if (offset < 0)
+                                break;
+                        }
+                        catch
+                        {
+                            if (!proteinsNotInDb.Contains(protein))
+                                proteinsNotInDb.Add(protein);
+                        }
 
+                        // add the protein to the occupancy dictionary if it has not been seen, yet
+                        if (!occupancyDict.ContainsKey(protein))
+                        {
+                            occupancyDict.Add(protein, new Dictionary<int, Dictionary<string, double>>());
+                            proteinSeqRangesSeen.Add(protein, new List<Tuple<int, int, double>>());
+                        }
+
+                        proteinSeqRangesSeen[protein].Add(new Tuple<int, int, double>(offset, offset + peptidesBase[i].Length, peak.PeakIntensity));
+
+                        // get the localized modifications from the peptide full sequence and add any amino acid/modification combination not
+                        // seen yet to the occupancy dictionary
+                        foreach (KeyValuePair<int, List<string>> aaWithModList in peptideMods)
+                        {
+                            int aa = aaWithModList.Key + offset;
+                            if (!occupancyDict[protein].ContainsKey(aaWithModList.Key))
+                                occupancyDict[protein].Add(aaWithModList.Key, new Dictionary<string, double> { { "Total", 0 } });
+
+                            foreach (string mod in aaWithModList.Value)
+                            {
+                                if (!occupancyDict[protein][aaWithModList.Key].ContainsKey(mod))
+                                    occupancyDict[protein][aaWithModList.Key][mod] = 0;
+
+                                occupancyDict[protein][aaWithModList.Key][mod] += peak.PeakIntensity;
+                            }
+                        }
+                    }
+                }
+            }
+
+            // Add the total intensity for each aa of each protein with a modification
+            foreach (string protein in occupancyDict.Keys)
+            {
+                foreach (int aa in occupancyDict[protein].Keys)
+                {
+                    foreach ((int start, int end, double intensity) in proteinSeqRangesSeen[protein])
+                    {
+                        if ((start <= aa) && (aa <= end))
+                            occupancyDict[protein][aa]["Total"] += intensity;
+                    }
+                }
+            }
         }
 
         [Test]