Merge branch 'precursorFragmentGroup' of https://github.com/zhuoxinshi/MetaMorpheus into precursorFragmentGroup

Author: zhuoxinshi

.github/workflows/DockerUpload.yml

@@ -0,0 +1,70 @@
+name: Docker Uploads
+
+on:
+  push:
+    branches: [ master ]         # dev tag on merges to master
+  release:
+    types: [released, edited]    # latest + version tag on new releases
+  
+  # Docker (Linux) push on branch merges (not PRs)
+jobs:
+  docker-dev:
+    if: github.event_name == 'push' && github.ref == 'refs/heads/master'
+    runs-on: ubuntu-latest
+    timeout-minutes: 20
+    env:
+      DOCKER_USER: ${{ secrets.DOCKER_USERNAME }}
+      DOCKER_TOKEN: ${{ secrets.DOCKER_TOKEN }}
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v4
+    - name: Set up .NET
+      uses: actions/setup-dotnet@v4
+      with:
+        dotnet-version: 8.0.x
+    - name: Restore dependencies
+      run: dotnet restore ./MetaMorpheus/MetaMorpheus.sln
+    - name: Build (CMD)
+      run: dotnet publish /p:Configuration=Release ./MetaMorpheus/CMD/CMD.csproj  -f net8.0
+    - name: Login to Docker Hub
+      uses: docker/login-action@v3
+      with:
+        username: ${{ env.DOCKER_USER }}
+        password: ${{ env.DOCKER_TOKEN }}
+    - name: Build and push :dev
+      uses: docker/build-push-action@v6
+      with:
+        context: "${{ github.workspace }}/MetaMorpheus"
+        push: true
+        tags: smithchemwisc/metamorpheus:dev
+
+# Push to "latest" and specific release tags on new release
+  push-release:
+    if: github.event_name == 'release'
+    runs-on: ubuntu-latest
+    timeout-minutes: 20
+    env:
+      DOCKER_USER: ${{ secrets.DOCKER_USERNAME }}
+      DOCKER_TOKEN: ${{ secrets.DOCKER_TOKEN }}
+    steps:
+    - name: Checkout
+      uses: actions/checkout@v4
+    - name: Set up .NET
+      uses: actions/setup-dotnet@v4
+      with:
+        dotnet-version: 8.0.x
+    - name: Restore dependencies
+      run: dotnet restore ./MetaMorpheus/MetaMorpheus.sln
+    - name: Build (CMD)
+      run: dotnet publish /p:Configuration=Release ./MetaMorpheus/CMD/CMD.csproj  -f net8.0
+    - name: Login to Docker Hub
+      uses: docker/login-action@v3
+      with:
+        username: ${{ env.DOCKER_USER }}
+        password: ${{ env.DOCKER_TOKEN }}
+    - name: Build and push :latest
+      uses: docker/build-push-action@v6
+      with:
+        context: "${{ github.workspace }}/MetaMorpheus"
+        push: true
+        tags: smithchemwisc/metamorpheus:${{ github.event.release.tag_name }}, smithchemwisc/metamorpheus:latest

MetaMorpheus/Dockerfile

@@ -1,8 +1,8 @@
 ## Base image is the Alpine Linux distro with .NET Core runtime
-FROM mcr.microsoft.com/dotnet/runtime:6.0-alpine
+FROM mcr.microsoft.com/dotnet/runtime:8.0-alpine
 
 ## Copies contents of the build folder into the Docker image
-ADD CMD/bin/Release/net6.0/ /metamorpheus/
+ADD CMD/bin/Release/net8.0/ /MetaMorpheus/
 
 ## Installs bash (seemingly necessary for NextFlow)
 RUN apk add --no-cache bash
@@ -11,4 +11,4 @@ RUN apk add --no-cache bash
 RUN apk add --no-cache procps
 
 ## Set the entrypoint of the Docker image to CMD.dll
-ENTRYPOINT ["dotnet", "/metamorpheus/CMD.dll"]
+ENTRYPOINT ["dotnet", "/MetaMorpheus/CMD.dll"]

MetaMorpheus/Dockerfile_WindowsInstaller

@@ -1,8 +1,12 @@
-## Base image is the Alpine Linux distro with .NET Core runtime
-FROM mcr.microsoft.com/dotnet/runtime:6.0-nanoserver-1809
+## This is an experimental script. 
+## The docker container built from this script runs a stripped down version of windows
+## The docker container built by this script has never been deployed to docker hub
+
+## Base image is a windows nanoserver with .NET Core runtime
+FROM mcr.microsoft.com/dotnet/runtime:8.0-nanoserver-1809
 
 ## Copies contents of the installer folder into the Docker image
 ADD /InstalledFiles/ /metamorpheus/
 
 ## Set the entrypoint of the Docker image to CMD.exe
-ENTRYPOINT ["/metamorpheus/CMD.exe"]
+ENTRYPOINT ["/metamorpheus/CMD.exe"]

MetaMorpheus/EngineLayer/Calibration/DataPointAcquisitionEngine.cs

@@ -1,9 +1,7 @@
 using Chemistry;
 using MassSpectrometry;
 using MzLibUtil;
-using Proteomics.AminoAcidPolymer;
 using System;
-using System.Collections.Concurrent;
 using System.Collections.Generic;
 using System.Linq;
 using System.Threading.Tasks;
@@ -76,13 +74,6 @@ protected override MetaMorpheusEngineResults RunSpecific()
 
                     var representativeSinglePeptide = identification.BestMatchingBioPolymersWithSetMods.First().SpecificBioPolymer;
 
-                    // Get the peptide, don't forget to add the modifications!!!!
-                    var SequenceWithChemicalFormulas = representativeSinglePeptide.SequenceWithChemicalFormulas;
-                    if (SequenceWithChemicalFormulas == null || representativeSinglePeptide.AllModsOneIsNterminus.Any(b => b.Value.NeutralLosses != null))
-                        continue;
-
-                    Peptide coolPeptide = new Peptide(SequenceWithChemicalFormulas);
-
                     List<LabeledDataPoint> ms2tuple = SearchMS2Spectrum(GoodScans[matchIndex], identification, ProductMassTolerance);
 
                     lock (_ms2Lock)
@@ -91,7 +82,7 @@ protected override MetaMorpheusEngineResults RunSpecific()
                     }
 
                     // Calculate theoretical isotopic distribution of the full peptide
-                    var dist = IsotopicDistribution.GetDistribution(coolPeptide.GetChemicalFormula(), FineResolutionForIsotopeDistCalculation, 0.001);
+                    var dist = IsotopicDistribution.GetDistribution(representativeSinglePeptide.ThisChemicalFormula, FineResolutionForIsotopeDistCalculation, 0.001);
 
                     double[] theoreticalMasses = dist.Masses.ToArray();
                     double[] theoreticalIntensities = dist.Intensities.ToArray();
@@ -137,6 +128,7 @@ protected override MetaMorpheusEngineResults RunSpecific()
             theIndex = direction == 1 ? ms2spectrumIndex + 1 : identification.PrecursorScanNumber ?? ms2spectrumIndex;
 
             bool addedAscan = true;
+            bool positiveMode = identification.ScanPrecursorCharge > 0;
 
             int highestKnownChargeForThisPeptide = peptideCharge;
             while (theIndex >= 1 && theIndex <= MyMsDataFile.NumSpectra && addedAscan) //as long as we're finding the peptide in ms1 scans
@@ -157,16 +149,26 @@ protected override MetaMorpheusEngineResults RunSpecific()
                     break;
 
                 bool startingToAddCharges = false;
-                int chargeToLookAt = 1;
+                int chargeToLookAt = positiveMode ? 1 : -1; 
+                int chargeLimit = positiveMode ? highestKnownChargeForThisPeptide + 1 : highestKnownChargeForThisPeptide - 1;
+
                 do
                 {
-                    if (theoreticalMasses[0].ToMz(chargeToLookAt) > scanWindowRange.Maximum)
+                    double mz = theoreticalMasses[0].ToMz(chargeToLookAt);
+
+                    // If m/z is above the scan window, try next charge
+                    if (mz > scanWindowRange.Maximum)
                     {
-                        chargeToLookAt++;
+                        chargeToLookAt += positiveMode ? 1 : -1;
                         continue;
                     }
-                    if (theoreticalMasses[0].ToMz(chargeToLookAt) < scanWindowRange.Minimum)
+
+                    // If m/z is below the scan window, break early, more charge will only make the mz smaller. 
+                    if (mz < scanWindowRange.Minimum)
+                    {
                         break;
+                    }
+
                     var trainingPointsToAverage = new List<LabeledDataPoint>();
                     foreach (double a in theoreticalMasses)
                     {
@@ -187,15 +189,17 @@ protected override MetaMorpheusEngineResults RunSpecific()
                         var closestPeakIndex = fullMS1spectrum.GetClosestPeakIndex(theMZ);
                         var closestPeakMZ = fullMS1spectrum.XArray[closestPeakIndex];
 
-                        highestKnownChargeForThisPeptide = Math.Max(highestKnownChargeForThisPeptide, chargeToLookAt);
+                        highestKnownChargeForThisPeptide = positiveMode
+                            ? Math.Max(highestKnownChargeForThisPeptide, chargeToLookAt)
+                            : Math.Min(highestKnownChargeForThisPeptide, chargeToLookAt);
                         trainingPointsToAverage.Add(new LabeledDataPoint(closestPeakMZ, -1, double.NaN, double.NaN, Math.Log(fullMS1spectrum.YArray[closestPeakIndex]), theMZ, null));
                     }
                     // If started adding and suddenly stopped, go to next one, no need to look at higher charges
                     if (trainingPointsToAverage.Count == 0 && startingToAddCharges)
                     {
                         break;
                     }
-                    if ((trainingPointsToAverage.Count == 0 || (trainingPointsToAverage.Count == 1 && theoreticalIntensities[0] < 0.65)) && (peptideCharge <= chargeToLookAt))
+                    if ((trainingPointsToAverage.Count == 0 || (trainingPointsToAverage.Count == 1 && theoreticalIntensities[0] < 0.65)) && (positiveMode ? peptideCharge <= chargeToLookAt : peptideCharge >= chargeToLookAt))
                     {
                         break;
                     }
@@ -213,8 +217,8 @@ protected override MetaMorpheusEngineResults RunSpecific()
                                                              trainingPointsToAverage.Select(b => b.TheoreticalMz).Average(),
                                                              identification));
                     }
-                    chargeToLookAt++;
-                } while (chargeToLookAt <= highestKnownChargeForThisPeptide + 1);
+                    chargeToLookAt += positiveMode ? 1 : -1;
+                } while (positiveMode ? chargeToLookAt <= chargeLimit : chargeToLookAt >= chargeLimit);
                 theIndex += direction;
             }
             return (result, numMs1MassChargeCombinationsConsidered, numMs1MassChargeCombinationsThatAreIgnoredBecauseOfTooManyPeaks);
@@ -240,16 +244,17 @@ private static List<LabeledDataPoint> SearchMS2Spectrum(Ms2ScanWithSpecificMass
                 if(envelopesThatMatch.Count == 0)
                     continue;
                 //only allow one envelope per charge state
-                bool[] chargeStateFound = new bool[envelopesThatMatch.Max(x => x.Charge) + 1];
+                bool[] chargeStateFound = new bool[envelopesThatMatch.Max(x => Math.Abs(x.Charge)) + 1];
 
                 foreach (var envelopeThatMatched in envelopesThatMatch)
                 {
+                    int charge = Math.Abs(envelopeThatMatched.Charge);
                     //if we haven't seen this charge state already
-                    if (!chargeStateFound[envelopeThatMatched.Charge])
+                    if (!chargeStateFound[charge])
                     {
-                        chargeStateFound[envelopeThatMatched.Charge] = true;
+                        chargeStateFound[charge] = true;
 
-                        double exptPeakMz = envelopeThatMatched.MonoisotopicMass.ToMz(envelopeThatMatched.Charge);
+                        double exptPeakMz = envelopeThatMatched.MonoisotopicMass.ToMz(charge);
                         double exptPeakIntensity = envelopeThatMatched.TotalIntensity;
                         double injTime = ms2DataScan.TheScan.InjectionTime ?? double.NaN;
 
@@ -259,8 +264,8 @@ private static List<LabeledDataPoint> SearchMS2Spectrum(Ms2ScanWithSpecificMass
                                 ms2DataScan.OneBasedScanNumber,
                                 Math.Log(ms2DataScan.TotalIonCurrent),
                                 Math.Log(injTime),
-                                Math.Log(exptPeakIntensity),
-                                matchedIon.NeutralTheoreticalProduct.NeutralMass.ToMz(envelopeThatMatched.Charge),
+                        Math.Log(exptPeakIntensity),
+                                matchedIon.NeutralTheoreticalProduct.NeutralMass.ToMz(charge),
                                 identification));
                     }
                 }

MetaMorpheus/EngineLayer/ClassicSearch/ClassicSearchEngine.cs

@@ -11,6 +11,7 @@
 using System.Collections.Concurrent;
 using EngineLayer.Util;
 using Omics;
+using Readers.SpectralLibrary;
 
 namespace EngineLayer.ClassicSearch
 {

MetaMorpheus/EngineLayer/ClassicSearch/MiniClassicSearchEngine.cs

@@ -7,6 +7,7 @@
 using System;
 using System.Collections.Generic;
 using System.Linq;
+using Readers.SpectralLibrary;
 
 namespace EngineLayer.ClassicSearch
 {

MetaMorpheus/EngineLayer/GlycoSearch/Glycan.cs

@@ -313,9 +313,13 @@ public static List<Glycan> Struct2Glycan(string theGlycanStruct, int id, bool is
             else 
             {
                 GlycanType glycanType = GlycanType.N_glycan;
-                Glycan N_glycan = new Glycan(theGlycanStruct, mass, kind, glycanIons.OrderBy(p => p.IonMass).ToList(), false, "N", glycanType);
-                N_glycan.GlyId = id;
-                glycans.Add(N_glycan);
+                Glycan N_glycan_Nxs = new Glycan(theGlycanStruct, mass, kind, glycanIons.OrderBy(p => p.IonMass).ToList(), false, "Nxs", glycanType);
+                N_glycan_Nxs.GlyId = id;
+                Glycan N_glycan_Nxt = new Glycan(theGlycanStruct, mass, kind, glycanIons.OrderBy(p => p.IonMass).ToList(), false, "Nxt", glycanType);
+                N_glycan_Nxt.GlyId = id+1;
+
+                glycans.Add(N_glycan_Nxs);
+                glycans.Add(N_glycan_Nxt);
                 return glycans;
             }
 

MetaMorpheus/EngineLayer/GlycoSearch/GlycanDatabase.cs

@@ -91,14 +91,24 @@ public static IEnumerable<Glycan> LoadKindGlycan(string filePath, bool ToGenerat
                     }
                     else // Load the N-glycan with one motif : N
                     {
-                        var nGlycan = new Glycan(kind, "N", GlycanType.N_glycan); // Use the kind[] to create a glycan object.
-                        nGlycan.GlyId = id;
+                        var nGlycan_Nxs = new Glycan(kind, "Nxs", GlycanType.N_glycan); // Use the kind[] to create a glycan object.
+                        nGlycan_Nxs.GlyId = id;
                         id++;
                         if (ToGenerateIons)
                         {
-                            nGlycan.Ions = OGlycanCompositionCombinationChildIons(kind);
+                            nGlycan_Nxs.Ions = OGlycanCompositionCombinationChildIons(kind);
                         }
-                        yield return nGlycan;
+                        yield return nGlycan_Nxs;
+
+
+                        var nGlycan_Nxt = new Glycan(kind, "Nxt", GlycanType.N_glycan); // Use the kind[] to create a glycan object.
+                        nGlycan_Nxt.GlyId = id;
+                        id++;
+                        if (ToGenerateIons)
+                        {
+                            nGlycan_Nxt.Ions = OGlycanCompositionCombinationChildIons(kind);
+                        }
+                        yield return nGlycan_Nxt;
                     }
                 }
             }
@@ -138,20 +148,14 @@ public static IEnumerable<Glycan> LoadStructureGlycan(string filePath, bool IsOG
                 {
                     string line = glycans.ReadLine();   // Read the line from the database file. Ex. (N(H(A))(A))
 
-                    // For each O-glycan, two versions will be generated: one modified on serine (S), and the other on threonine (T).
-                    if (IsOGlycan)
-                    {
-                        foreach (var glycan in Glycan.Struct2Glycan(line, id, IsOGlycan)) // Modify the line to handle multiple Glycan objects returned by Struct2Glycan.  
-                        {
-                            yield return glycan;
-                        }
-                        id = id +2; // Each line will generate two glycan objects
-                    }
-                    //For N-glycan, we only generate one kind of N-glycan.
-                    else
+                    // For each glycan, two versions will be generated:
+                    // For O-glycan, one modified on serine (S), and the other on threonine (T).
+                    // For N-glycan, one modified on N-glycosylation on motif Asn-X-Ser(Nxs), and the other on Asn-X-Thr(Nxt).
+                    foreach (var glycan in Glycan.Struct2Glycan(line, id, IsOGlycan)) // Modify the line to handle multiple Glycan objects returned by Struct2Glycan.  
                     {
-                        yield return Glycan.Struct2Glycan(line, id++, IsOGlycan).FirstOrDefault(); // For N-Glycan, we only return the first Glycan object.
+                        yield return glycan;
                     }
+                    id = id + 2; // Each line will generate two glycan objects
                 }
             }
         }

MetaMorpheus/EngineLayer/GlycoSearch/GlycoSearchEngine.cs

@@ -481,20 +481,29 @@ private void FindOGlycan(Ms2ScanWithSpecificMass theScan, int scanIndex, int sco
 
         private void FindNGlycan(Ms2ScanWithSpecificMass theScan, int scanIndex, int scoreCutOff, PeptideWithSetModifications theScanBestPeptide, int ind, double possibleGlycanMassLow, double[] oxoniumIonIntensities, ref List<GlycoSpectralMatch> possibleMatches)
         {
-            List<int> modPos = GlycoSpectralMatch.GetPossibleModSites(theScanBestPeptide, new string[] { "Nxt", "Nxs" }).Select(p => p.Key).ToList();
-            if (modPos.Count < 1)
+            List<int> modPos_Nxs = GlycoSpectralMatch.GetPossibleModSites(theScanBestPeptide, new string[] { "Nxs" }).Select(p => p.Key).ToList();
+            List<int> modPos_Nxt = GlycoSpectralMatch.GetPossibleModSites(theScanBestPeptide, new string[] { "Nxt" }).Select(p => p.Key).ToList();
+            if (modPos_Nxs.Count < 1 && modPos_Nxt.Count < 1) // if there is no possible glycosylation site, we can skip this peptide.
             {
                 return;
             }
 
             int iDLow = GlycoPeptides.BinarySearchGetIndex(NGlycans.Select(p => (double)p.Mass / 1E5).ToArray(), possibleGlycanMassLow);
-
             while (iDLow < NGlycans.Length && PrecusorSearchMode.Within(theScan.PrecursorMass, theScanBestPeptide.MonoisotopicMass + (double)NGlycans[iDLow].Mass / 1E5))
             {
                 double bestLocalizedScore = scoreCutOff;
                 int bestSite = 0;
                 List<MatchedFragmentIon> bestMatchedIons = new List<MatchedFragmentIon>();
                 PeptideWithSetModifications[] peptideWithSetModifications = new PeptideWithSetModifications[1];
+
+                // Get the correct modification position based on the glycan target type
+                List<int> modPos = NGlycans[iDLow].Target.ToString() == "Nxs" ? modPos_Nxs : modPos_Nxt;
+                if (modPos.Count < 1)
+                {
+                    iDLow++;
+                    continue; // if there is no possible glycosylation site, we can skip this glycan.
+                }
+
                 foreach (int possibleSite in modPos)
                 {
                     var testPeptide = GlycoPeptides.GenerateGlycopeptide(possibleSite, theScanBestPeptide, NGlycans[iDLow]);