Merge branch 'DockerFileUpdate' of https://github.com/Alexander-Sol/MetaMorpheus into DockerFileUpdate

Author: Alexander-Sol

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/GUI/MainWindow.xaml.cs

@@ -21,6 +21,7 @@
 using TaskLayer;
 using System.Text.RegularExpressions;
 using Readers.InternalResults;
+using System.Diagnostics;
 
 namespace MetaMorpheusGUI
 {
@@ -1263,8 +1264,35 @@ private void MenuItem_GuiSettings_Click(object sender, RoutedEventArgs e)
 
         private void MenuItem_MetaDraw_Click(object sender, RoutedEventArgs e)
         {
-            MetaDraw metaDrawGui = new MetaDraw();
-            metaDrawGui.Show();
+            string[]? filesToLoad = null;
+            try
+            {
+                // get completed search tasks from in progress tasks. 
+                var completedSearchTasks = InProgressTasks.Where(p => p.Task is SearchTask && p.Progress >= 100).ToList();
+
+                // find last search task by the task ID
+                var finalSearchTask = completedSearchTasks.MaxBy(p => p.Id.Split('-')[0].Replace("Task", "").ToNullableInt());
+
+                // Get Spectra files
+                var prose = MetaMorpheusProseFile.LocateInDirectory(finalSearchTask.Task.OutputFolder);
+                var spectraFiles = prose!.SpectraFilePaths;
+
+                // Get search results
+                var searchResult = Directory.GetFiles(finalSearchTask.Task.OutputFolder)
+                    .First(p => p.EndsWith(".psmtsv") || p.EndsWith(".osmtsv"));
+
+                filesToLoad = spectraFiles.Append(searchResult).ToArray();
+            }
+            catch (Exception ex)
+            {
+                // Something went wrong in trying to find a completed search task. 
+                // This is expected when we do not have a completed search to parse. 
+            }
+            finally
+            {
+                MetaDraw metaDrawGui = new MetaDraw(filesToLoad);
+                metaDrawGui.Show();
+            }
         }
 
         private void MenuItem_ResetDefaults_Click(object sender, RoutedEventArgs e)

MetaMorpheus/GUI/MetaDraw/MetaDraw.xaml.cs

@@ -43,7 +43,7 @@ public partial class MetaDraw : Window
         private FragmentationReanalysisViewModel FragmentationReanalysisViewModel;
         public ChimeraAnalysisTabViewModel ChimeraAnalysisTabViewModel { get; set; }
 
-        public MetaDraw()
+        public MetaDraw(string[]? filesToLoad = null)
         {
             InitializeComponent();
 
@@ -77,6 +77,12 @@ public MetaDraw()
             plotTypes = new ObservableCollection<string>();
             SetUpPlots();
             plotsListBox.ItemsSource = plotTypes;
+
+            // if files piped in from a MetaMorpheus search, load them immediately
+            if (filesToLoad is not null)
+            {
+                filesToLoad.ForEach(AddFile);
+            }
         }
 
         private void Window_Drop(object sender, DragEventArgs e)

MetaMorpheus/GUI/Views/ProgressBarIndeterminate.xaml

@@ -6,7 +6,8 @@
         xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006"
         xmlns:local="clr-namespace:MetaMorpheusGUI"
         mc:Ignorable="d"
-        Title="Progress" Height="115" Width="300">
+        Title="Progress" Height="115" Width="300"
+        Topmost="True">
     <StackPanel Margin="3">
         <Grid Height="20">
             <ProgressBar Minimum="0" Maximum="100" Name="pbStatus" IsIndeterminate="True"/>

MetaMorpheus/TaskLayer/CalibrationTask/CalibrationTask.cs

@@ -268,7 +268,7 @@ private DataPointAquisitionResults GetDataAcquisitionResults(MsDataFile myMsData
                 new SingleAbsoluteAroundZeroSearchMode(initPrecTol.Value);
 
             Ms2ScanWithSpecificMass[] listOfSortedms2Scans = GetMs2Scans(myMsDataFile, currentDataFile, combinedParameters).OrderBy(b => b.PrecursorMass).ToArray();
-            SpectralMatch[] allPsmsArray = new PeptideSpectralMatch[listOfSortedms2Scans.Length];
+            SpectralMatch[] allPsmsArray = new SpectralMatch[listOfSortedms2Scans.Length];
 
             Log("Searching with searchMode: " + searchMode, new List<string> { taskId, "Individual Spectra Files", fileNameWithoutExtension });
             Log("Searching with productMassTolerance: " + initProdTol, new List<string> { taskId, "Individual Spectra Files", fileNameWithoutExtension });

MetaMorpheus/TaskLayer/MetaMorpheusTask.cs

@@ -139,7 +139,8 @@ public abstract class MetaMorpheusTask
 
         protected readonly StringBuilder ProseCreatedWhileRunning = new StringBuilder();
 
-        protected string OutputFolder { get; private set; }
+        [TomlIgnore]
+        public string OutputFolder { get; private set; }
 
         protected MyTaskResults MyTaskResults;
 

MetaMorpheus/Test/Test.csproj

@@ -652,6 +652,9 @@
     <None Update="TopDownTestData\TopDownSearchToml.toml">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
     </None>
+    <None Update="Transcriptomics\TestData\16mer2.fasta">
+      <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
+    </None>
     <None Update="Transcriptomics\TestData\22mer-Am.fasta">
       <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
     </None>
@@ -676,6 +679,9 @@
     <None Update="Transcriptomics\TestData\GUACUG_NegativeMode_Sliced.mzML">
       <CopyToOutputDirectory>Always</CopyToOutputDirectory>
     </None>
+    <None Update="Transcriptomics\TestData\RnaStandard_Subset.mzML">
+      <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
+    </None>
     <None Update="Transcriptomics\TestData\TruncationAndVariantMods.xml">
       <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
     </None>

MetaMorpheus/Test/Transcriptomics/TestCalibration.cs

@@ -0,0 +1,118 @@
+using EngineLayer;
+using MassSpectrometry;
+using MzLibUtil;
+using NUnit.Framework;
+using System;
+using System.Collections.Generic;
+using System.IO;
+using System.Linq;
+using TaskLayer;
+using Transcriptomics.Digestion;
+using UsefulProteomicsDatabases;
+using Readers;
+
+namespace Test.Transcriptomics;
+
+[TestFixture]
+public class TestCalibration
+{
+    public static RnaSearchParameters SearchParameters;
+    public static CommonParameters CommonParameters;
+
+    [OneTimeSetUp]
+    public static void Setup()
+    {
+        SearchParameters = new RnaSearchParameters
+        {
+            DecoyType = DecoyType.Reverse,
+            MassDiffAcceptorType = MassDiffAcceptorType.Exact,
+            DisposeOfFileWhenDone = true
+        };
+        CommonParameters = new CommonParameters
+        (
+            dissociationType: DissociationType.CID,
+            deconvolutionMaxAssumedChargeState: -20,
+            deconvolutionIntensityRatio: 3,
+            deconvolutionMassTolerance: new PpmTolerance(20),
+            precursorMassTolerance: new PpmTolerance(10),
+            productMassTolerance: new PpmTolerance(5),
+            scoreCutoff: 5,
+            totalPartitions: 1,
+            maxThreadsToUsePerFile: 1,
+            doPrecursorDeconvolution: true,
+            useProvidedPrecursorInfo: false,
+            digestionParams: new RnaDigestionParams()
+        );
+    }
+
+    [Test]
+    public void CalibrationTask_ReducesMassError()
+    {
+        // Arrange
+        string testDir = TestContext.CurrentContext.TestDirectory;
+        List<string> dbPaths = 
+            [Path.Combine(testDir, "Transcriptomics", "TestData", "16mer2.fasta")];
+        List<string> spectraPaths = 
+            [Path.Combine(testDir, "Transcriptomics", "TestData", "RnaStandard_Subset.mzML")];
+        string outputDir = Path.Combine(testDir, "CalibrationTest");
+
+        if (Directory.Exists(outputDir))
+            Directory.Delete(outputDir, true); // Clean up previous test run
+        Directory.CreateDirectory(outputDir);
+
+        // Set up search-cali-search task for RNA
+        var dbs = dbPaths.Select(dbPath => new DbForTask(dbPath, false)).ToList();
+        var searchTask = new SearchTask()
+        {
+            SearchParameters = SearchParameters,
+            CommonParameters = CommonParameters
+        };
+        var cali = new CalibrationTask()
+        {
+            CommonParameters = CommonParameters
+        };
+        var searchTask2 = new SearchTask()
+        {
+            SearchParameters = SearchParameters,
+            CommonParameters = CommonParameters
+        };
+        var runner = new EverythingRunnerEngine(new List<(string, MetaMorpheusTask)>()
+        {
+            ("Search1", searchTask),
+            ("Calibration", cali),
+            ("Search2", searchTask2)
+        }, spectraPaths, dbs, outputDir);
+        runner.Run();
+
+        var firstSearchDir = Path.Combine(outputDir, "Search1");
+        var firstOsmPath = Directory.GetFiles(firstSearchDir, "*OSMs.osmtsv", SearchOption.AllDirectories).First();
+        var firstOsms = SpectrumMatchTsvReader.ReadTsv(firstOsmPath, out var warnings);
+        Assert.That(warnings.Count, Is.EqualTo(0));
+
+        var secondSearchDir = Path.Combine(outputDir, "Search2");
+        var secondOsmPath = Directory.GetFiles(secondSearchDir, "*OSMs.osmtsv", SearchOption.AllDirectories).First();
+        var secondOsms = SpectrumMatchTsvReader.ReadTsv(secondOsmPath, out warnings);
+        Assert.That(warnings.Count, Is.EqualTo(0));
+
+        // Assert that calibration increased search results
+        Assert.That(secondOsms.Count, Is.GreaterThan(firstOsms.Count), "Expected more OligoSpectralMatches after calibration.");
+        Assert.That(secondOsms.Count(p => p.QValue <= 0.01), Is.GreaterThan(firstOsms.Count(p => p.QValue <= 0.01)),
+            "Expected more OligoSpectralMatches with QValue <= 0.01 after calibration.");
+
+        // Assert that the mass error is reduced after calibration
+        var firstErrors = firstOsms.Select(p => p.MassDiffDa)
+            .Select(double.Parse)
+            .ToList();
+        var secondErrors = secondOsms.Select(p => p.MassDiffDa)
+            .Select(double.Parse)
+            .ToList();
+
+        var firstAverage = firstErrors.Average();
+        var secondAverage = secondErrors.Average();
+
+        Assert.That(secondAverage, Is.LessThan(firstAverage),
+            "Expected average mass error to be reduced after calibration.");
+
+        Directory.Delete(outputDir, true); // Clean up after test
+    }
+}

MetaMorpheus/Test/Transcriptomics/TestData/16mer2.fasta

@@ -0,0 +1,2 @@
+>id:9|Name:16mer1|SOterm:16mer1|Type:tRNA|Subtype:Ala|Feature:VGC|Cellular_Localization:freezer|Species:standard
+AUCCAGUGCAGUACUG