Mod_Pair part1 ModBox and Reader

MetaMorpheus/EngineLayer/EngineLayer.csproj

@@ -67,6 +67,9 @@
     <None Update="Digestion\rnases.tsv">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
     </None>
+    <None Update="Glycan_Mods\General_Mods\ModBox Database.txt">
+      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
+    </None>
     <None Update="Glycan_Mods\NGlycan\NGlycan.gdb">
       <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
     </None>

MetaMorpheus/EngineLayer/Glycan_Mods/General_Mods/ModBox Database.txt

@@ -0,0 +1,12 @@
+HexNAc(1)	N1H0F0S0	203.0794
+HexNAc(1)Hex(1)	N1H1F0S0	365.1322
+HexNAc(2)	N2H0F0S0	406.1588
+HexNAc(1)NeuAc(1)	N1H0F0S1	494.1748
+HexNAc(1)Hex(1)Fuc(1)	N1H1F1S0	511.1901
+HexNAc(1)Hex(2)	N1H2F0S0	527.185
+HexNAc(2)Hex(1)	N2H1F0S0	568.2116
+HexNAc(1)Hex(1)NeuAc(1)	N1H1F0S1	656.2276
+HexNAc(1)Hex(2)Fuc(1)	N1H2F1S0	673.2429
+HexNAc(2)Hex(1)Fuc(1)	N2H1F1S0	714.2695
+Oxidation on M	O1	15.9994
+Oxidation on Q	O1	15.9994

MetaMorpheus/EngineLayer/GlycoSearch/AdjNode.cs

@@ -1,7 +1,7 @@
 using System;
 using System.Collections.Generic;
 using System.Text;
-
+using EngineLayer.ModSearch;
 namespace EngineLayer.GlycoSearch
 {
     //the class is for localization graph matrix. Each node in the matrix is represented by AdjNode.

MetaMorpheus/EngineLayer/GlycoSearch/Glycan.cs

@@ -6,6 +6,7 @@
 using Proteomics;
 using MassSpectrometry;
 using Omics.Modifications;
+using System.Text.RegularExpressions;
 
 namespace EngineLayer
 {
@@ -60,7 +61,7 @@ public string Composition                 // Glycan composition string. Ex. H2N2
         public HashSet<int> DiagnosticIons // B ions (the sugar fragment dropped from the glycopeptide), used for the N-glycan. There are more ions to set...
         {
             get
-            {   
+            {
                 HashSet<int> diagnosticIons = new HashSet<int>();
                 if (Kind[0] >= 1) //if we have Hexose(the number more than one), then we have the corresponding diagonsitic ions as below.
                 {
@@ -508,6 +509,39 @@ public static string GetKindString(byte[] Kind)
             return kindString;
         }
 
+        /// <summary>
+        /// Convert the glycan name to the glycan kind array.
+        /// </summary>
+        /// <param name="glycanName"></param>
+        /// <returns></returns>
+        public static byte[] StringToKind(string glycanName)
+        {
+            // Initialize the kind array with zeros
+            byte[] kind = new byte[NameCharDic.Count];
+
+            // Regular expression to match the sugar type and count
+            Regex regex = new Regex(@"([A-Z][a-z]*)(\d+)");
+
+            // Match the input string
+            var matches = regex.Matches(glycanName);
+
+            // Update the kind array based on the matches
+            foreach (Match match in matches)
+            {
+                string sugarType = match.Groups[1].Value;
+                int count = int.Parse(match.Groups[2].Value);
+
+                // Find the corresponding index for the sugar type
+                var sugarInfo = NameCharDic.FirstOrDefault(x => x.Value.Item1.ToString() == sugarType);
+                if (sugarInfo.Value != null)
+                {
+                    kind[sugarInfo.Value.Item2] = (byte)count;
+                }
+            }
+
+            return kind;
+        }
+
         #endregion
 
         //TO THINK: Is it reasonable to transfer Glycan to Modification the first time Glycan is read in? Which could save time.
@@ -557,7 +591,7 @@ public static Modification OGlycanToModification(Glycan glycan) //try to transfe
             //TO THINK: what the neutralLoss for O-Glyco?
             Dictionary<DissociationType, List<double>> neutralLosses = new Dictionary<DissociationType, List<double>>();
 
-            if (glycan.Ions!=null)
+            if (glycan.Ions != null)
             {
                 List<double> lossMasses = glycan.Ions.Select(p => (double)p.LossIonMass / 1E5).OrderBy(p => p).ToList();
                 neutralLosses.Add(DissociationType.HCD, lossMasses);

MetaMorpheus/EngineLayer/GlycoSearch/GlycanBox.cs

@@ -1,11 +1,8 @@
 using System.Collections.Generic;
-using System.IO;
 using System.Linq;
-using Chemistry;
 using System;
-using Proteomics;
-using MassSpectrometry;
 using Omics.Modifications;
+using EngineLayer.ModSearch;
 
 namespace EngineLayer
 {
@@ -128,9 +125,9 @@ public static IEnumerable<GlycanBox> BuildChildOGlycanBoxes(int maxNum, int[] gl
         public GlycanBox(int[] ids, bool Istarget = true):base(ids)
         {
             byte[] kind = new byte[] { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
-            foreach (var id in ModIds) //ModIds is the same as ids.
+            foreach (var id in ids) //ModIds is the same as ids.
             {
-                for (int i = 0; i < kind.Length; i++)   
+                for (int i = 0; i < kind.Length; i++)
                 {
                     kind[i] += GlobalOGlycans[id].Kind[i]; //kind is the sum of all glycan Kind in the Box.
                 }
@@ -139,7 +136,7 @@ public GlycanBox(int[] ids, bool Istarget = true):base(ids)
 
             if (Istarget)
             {
-                Mass = (double)Glycan.GetMass(Kind) / 1E5;
+                Mass = (double)(Glycan.GetMass(Kind)) / 1E5;
             }
             else
             {

MetaMorpheus/EngineLayer/GlycoSearch/GlycoPeptides.cs

@@ -1,12 +1,12 @@
 using MassSpectrometry;
-using Proteomics;
 using Omics.Fragmentation;
 using Proteomics.ProteolyticDigestion;
 using System;
 using System.Collections.Generic;
 using System.Linq;
 using Chemistry;
 using Omics.Modifications;
+using EngineLayer.ModSearch;
 
 namespace EngineLayer.GlycoSearch
 {
@@ -438,15 +438,15 @@ public static List<double> GetLocalFragment(List<Product> products, int[] modPos
 
             foreach (var c in local_c_fragments)
             {
-                var newMass = c.NeutralMass + localOGlycanBox.Mass;
+                var newMass = c.NeutralMass + localOGlycanBox.Mass.Value;  // use the value to get the double value of the mass.
                 newFragments.Add(newMass);
             }
 
             var local_z_fragments = products.Where(p => p.ProductType == ProductType.zDot && p.AminoAcidPosition >= modPoses[modInd] && p.AminoAcidPosition < modPoses[modInd + 1]).ToList();
 
             foreach (var z in local_z_fragments)
             {
-                var newMass = z.NeutralMass + (OGlycanBox.Mass - localOGlycanBox.Mass);
+                var newMass = z.NeutralMass + (OGlycanBox.Mass.Value - localOGlycanBox.Mass.Value);
                 newFragments.Add(newMass);
             }
 
@@ -456,7 +456,7 @@ public static List<double> GetLocalFragment(List<Product> products, int[] modPos
         //Find FragmentMass for the fragments that doesn't contain localization Information. For example, "A|TAABBS|B", c1 and c7, z1 and z7, z8 ion don't contain localization information.
         public static List<double> GetUnlocalFragment(List<Product> products, int[] modPoses, ModBox OGlycanBox)
         {
-            var mass = OGlycanBox.Mass;
+            var mass = OGlycanBox.Mass.Value;
 
             List<double> newFragments = new List<double>();
             var c_fragments = products.Where(p => p.ProductType == ProductType.c && p.AminoAcidPosition < modPoses.First() - 1).Select(p => p.NeutralMass);

MetaMorpheus/EngineLayer/GlycoSearch/GlycoSearchEngine.cs

@@ -17,7 +17,7 @@ public class GlycoSearchEngine : ModernSearchEngine
     {
         public static readonly double ToleranceForMassDifferentiation = 1e-9;
         private readonly int OxoniumIon204Index = 9;               // Check Glycan.AllOxoniumIons
-        protected readonly List<GlycoSpectralMatch>[] GlobalGsms;  // Why don't we call it GlobalGsms?
+        protected readonly List<GlycoSpectralMatch>[] GlobalGsms;  
 
         private GlycoSearchType GlycoSearchType;
         private readonly int TopN;              // DDA top Peak number.
@@ -204,7 +204,7 @@ protected override MetaMorpheusEngineResults RunSpecific()
 
                         if (GlobalGsms[scanIndex] == null)
                         {
-                            GlobalGsms[scanIndex] = new List<GlycoSpectralMatch>(); //the first one finished task, create teh new gsms list.
+                            GlobalGsms[scanIndex] = new List<GlycoSpectralMatch>(); //the first one finished task, create the new gsms list.
                         }
                         else
                         {
@@ -392,23 +392,23 @@ private void FindSingle(Ms2ScanWithSpecificMass theScan, int scanIndex, int scor
         /// <summary>
         /// Match the mass of the peptide candidate with the precursor mass. Try to generate the Gsms for the Scan. Gsms will be stored in the possibleMatches.
         /// </summary>
-        /// <param name="theScan"></param>
-        /// <param name="scanIndex"></param>
-        /// <param name="scoreCutOff"></param>
+        /// <param name="theScan"> The experimental data</param>
+        /// <param name="scanIndex"> The index for the scan</param>
+        /// <param name="scoreCutOff"> the cutoff for the indexing score</param>
         /// <param name="theScanBestPeptide"> peptide candidate </param>
         /// <param name="ind"></param>
-        /// <param name="possibleGlycanMassLow"> The precursor mass </param>
-        /// <param name="oxoniumIonIntensities"></param>
+        /// <param name="possibleGlycanMassLow"> The mass shift between the exact mass and the experimental mass </param>
+        /// <param name="oxoniumIonIntensities"> The oxonium for the filtering</param>
         /// <param name="possibleMatches"> The space to store the gsms </param>
         private void FindOGlycan(Ms2ScanWithSpecificMass theScan, int scanIndex, int scoreCutOff, PeptideWithSetModifications theScanBestPeptide, int ind, double possibleGlycanMassLow, double[] oxoniumIonIntensities, ref List<GlycoSpectralMatch> possibleMatches)
         {
             // The glycanBoxes will be filtered by the oxonium ions. If the oxonium ions don't make sense, we will remove the glycanBox.
 
-
-            int iDLow = GlycoPeptides.BinarySearchGetIndex(GlycanBox.OGlycanBoxes.Select(p => p.Mass).ToArray(), possibleGlycanMassLow); // try to find the index that closet match to the "possibleGlycanMassLow" within the glycanBox
+            // Using "possibleGlycanMassLow" to find index for the best matached glycanBox
+            int iDLow = GlycoPeptides.BinarySearchGetIndex(GlycanBox.OGlycanBoxes.Select(p => p.Mass.Value).ToArray(), possibleGlycanMassLow); 
 
             int[] modPos = GlycoSpectralMatch.GetPossibleModSites(theScanBestPeptide, new string[] { "S", "T" }).OrderBy(p => p).ToArray(); //list all of the possible glycoslation site/postition
-
+            
             var localizationScan = theScan;
             List<Product> products = new List<Product>(); // product list for the theoretical fragment ions
 
@@ -438,7 +438,7 @@ private void FindOGlycan(Ms2ScanWithSpecificMass theScan, int scanIndex, int sco
 
             List<LocalizationGraph> localizationGraphs = new List<LocalizationGraph>(); // if we also have ETD, then we will search the localization
 
-            while (iDLow < GlycanBox.OGlycanBoxes.Count() && (PrecusorSearchMode.Within(theScan.PrecursorMass, theScanBestPeptide.MonoisotopicMass + GlycanBox.OGlycanBoxes[iDLow].Mass))) // verify the glycan mass is invaild (within the range and match with mass shift)
+            while (iDLow < GlycanBox.OGlycanBoxes.Count() && (PrecusorSearchMode.Within(theScan.PrecursorMass, theScanBestPeptide.MonoisotopicMass + GlycanBox.OGlycanBoxes[iDLow].Mass.Value))) // verify the glycan mass is invaild (within the range and match with mass shift)
             {
                 if (OxoniumIonFilter && !GlycoPeptides.DiagonsticFilter(oxoniumIonIntensities, GlycanBox.OGlycanBoxes[iDLow])) // if the filter is turned on, we need to check does the oxoiums make sense.
                 {
@@ -447,6 +447,7 @@ private void FindOGlycan(Ms2ScanWithSpecificMass theScan, int scanIndex, int sco
                 }
                 if (modPos.Length >= GlycanBox.OGlycanBoxes[iDLow].NumberOfMods) // the glycosite number should be larger than the possible glycan number.
                 {
+
                     LocalizationGraph localizationGraph = new LocalizationGraph(modPos, GlycanBox.OGlycanBoxes[iDLow], GlycanBox.OGlycanBoxes[iDLow].ChildGlycanBoxes, iDLow);
                     LocalizationGraph.LocalizeOGlycan(localizationGraph, localizationScan, CommonParameters.ProductMassTolerance, products); //create the localization graph with the glycan mass and the possible glycosite.
 
@@ -646,7 +647,6 @@ private List<GlycoSpectralMatch> FindOGlycopeptideHashLocal(Ms2ScanWithSpecificM
                     {
                         continue;
                     }
-
                     //Find O-Glycan
                     FindOGlycan(theScan, scanIndex, scoreCutOff, theScanBestPeptide, ind, possibleGlycanMassLow, oxoniumIonIntensities, ref possibleMatches);
                 }

MetaMorpheus/EngineLayer/GlycoSearch/GlycoSpectralMatch.cs

@@ -311,7 +311,7 @@ public string GlycoToString()
                     sb.Append(string.Join(",", glycans.Select(p => p.Struc.ToString()).ToArray())); //ex. (N(H)),(N(H(A))(N(H))),(N(H)(N(H(A))(F))
                 }
                 sb.Append("\t");
-
+                
                 sb.Append(CorrectLocalizationLevel(SiteSpeciLocalProb, LocalizationGraphs.First(), Routes.First(), LocalizedGlycan, LocalizationLevel)); sb.Append("\t");
 
                 string local_peptide = "";
@@ -357,7 +357,7 @@ public string GlycoToString()
                 sb.Append("\t");
                 sb.Append("\t");
 
-                 sb.Append("\t");
+                sb.Append("\t");
 
                 sb.Append(SiteSpeciLocalInfo(SiteSpeciLocalProb));
             }
@@ -420,7 +420,7 @@ public static List<Tuple<int, int, bool>> GetLocalizedGlycan(List<Route> OGlycan
 
             foreach (var seenMod in modSiteSeenCount)
             {
-                if (seenMod.Value == OGlycanBoxLocalization.Count) // Try to fine the glycan-site pair that always localized in all the cases.
+                if (seenMod.Value == OGlycanBoxLocalization.Count) // Try to find the glycan-site pair that always localized in all the cases.
                 {
                     localizedGlycan.Add(new Tuple<int, int, bool>(int.Parse(seenMod.Key.Split('-')[0]), int.Parse(seenMod.Key.Split('-')[1]), true));
                 }

MetaMorpheus/EngineLayer/GlycoSearch/LocalizationGraph.cs

@@ -2,9 +2,8 @@
 using System.Collections.Generic;
 using System.Linq;
 using Omics.Fragmentation;
-using Proteomics.ProteolyticDigestion;
-using Proteomics;
 using MzLibUtil;
+using EngineLayer.ModSearch;
 
 namespace EngineLayer.GlycoSearch
 {
@@ -46,7 +45,6 @@ public LocalizationGraph(int[] modPos, ModBox modBox, ModBox[] childModBoxes, in
         public static void LocalizeOGlycan(LocalizationGraph localizationGraph, Ms2ScanWithSpecificMass theScan, Tolerance productTolerance, List<Product> products)
         {
             var boxSatisfyBox = BoxSatisfyBox(localizationGraph.ChildModBoxes);
-
             for (int i = 0; i < localizationGraph.ModPos.Length; i++)
             {
                 //maxLength: the most mods we can have up to current mod pos; minlengtt: the least mods we can have up to current mod pos.
@@ -57,7 +55,7 @@ public static void LocalizeOGlycan(LocalizationGraph localizationGraph, Ms2ScanW
                 {
                     if (localizationGraph.ChildModBoxes[j].NumberOfMods <= maxLength && localizationGraph.ChildModBoxes[j].NumberOfMods >= minlength)
                     {
-                        AdjNode adjNode = new AdjNode(i, j, localizationGraph.ModPos[i], localizationGraph.ChildModBoxes[j]); //chekc the num of glycan in this node is make sense.
+                        AdjNode adjNode = new AdjNode(i, j, localizationGraph.ModPos[i], localizationGraph.ChildModBoxes[j]); //check the num of glycan in this node is make sense.
 
                         double cost = 0;
                         if (i != localizationGraph.ModPos.Length - 1) // check the node is not the last one.

MetaMorpheus/EngineLayer/GlycoSearch/Route.cs

@@ -9,7 +9,7 @@ public class Route
 
         public int ModBoxId { get; set; }
 
-        //Tuple<int, int, double> mod pos, glycan id, local peak exist
+        //Tuple<int, int, double> mod pos, Mod id, local peak exist
         //For the local peak exist, the idea is that, in the localization graph matrix, if the node is detected as a mod, then the node score and the previous node has a current score >0.
         public List<Tuple<int, int, bool>> Mods { get; private set; } = new List<Tuple<int, int, bool>>();