Micro optimizations

mzLib/MassSpectrometry/Deconvolution/Algorithms/ClassicDeconvolutionAlgorithm.cs

@@ -32,7 +32,7 @@ internal override IEnumerable<IsotopicEnvelope> Deconvolute(MzSpectrum spectrumT
                 yield break;
             }
 
-            var isolatedMassesAndCharges = new List<IsotopicEnvelope>();
+            var isolatedMassesAndCharges = new List<IsotopicEnvelope>(10);
 
             (int start, int end) indexes = ExtractIndices(range.Minimum, range.Maximum);
 
@@ -48,6 +48,7 @@ internal override IEnumerable<IsotopicEnvelope> Deconvolute(MzSpectrum spectrumT
 
             //go through each peak in the selected range and assume it is the most intense peak of its isotopic envelope (if it's not, it will hopefully get a low score)
             //cycle through possible charge states and select the one that has the best score (fit) with the averagine model
+            HashSet<int> allPossibleChargeStates = new HashSet<int>();
             for (int candidateForMostIntensePeakIndex = indexes.start;
                  candidateForMostIntensePeakIndex < indexes.end;
                  candidateForMostIntensePeakIndex++)
@@ -61,7 +62,7 @@ internal override IEnumerable<IsotopicEnvelope> Deconvolute(MzSpectrum spectrumT
                     double candidateForMostIntensePeakMz = spectrum.XArray[candidateForMostIntensePeakIndex];
 
                     //Find what charge states this peak might be based on the spacing of nearby peaks (assumes isotopic resolution)
-                    HashSet<int> allPossibleChargeStates = new HashSet<int>();
+                    allPossibleChargeStates.Clear();
                     for (int i = candidateForMostIntensePeakIndex + 1;
                          i < spectrum.XArray.Length;
                          i++) //look at peaks of higher m/z
@@ -169,8 +170,9 @@ private IsotopicEnvelope FindIsotopicEnvelope(int massIndex, double candidateFor
             double[] theoreticalMasses = allMasses[massIndex];
             double[] theoreticalIntensities = allIntensities[massIndex];
             //add "most intense peak"
-            var listOfObservedPeaks = new List<(double, double)> { (candidateForMostIntensePeakMz, candidateForMostIntensePeakIntensity) };
-            var listOfRatios = new List<double> { theoreticalIntensities[0] / candidateForMostIntensePeakIntensity }; // theoreticalIntensities and theoreticalMasses are sorted by intensity, so first is most intense
+            int estimatedSize = theoreticalIntensities.Length;
+            var listOfObservedPeaks = new List<(double, double)>(estimatedSize) { (candidateForMostIntensePeakMz, candidateForMostIntensePeakIntensity) };
+            var listOfRatios = new List<double>(estimatedSize) { theoreticalIntensities[0] / candidateForMostIntensePeakIntensity }; // theoreticalIntensities and theoreticalMasses are sorted by intensity, so first is most intense
             // Assuming the test peak is most intense...
             // Try to find the rest of the isotopes!
             double differenceBetweenTheorAndActualMass = testMostIntenseMass - theoreticalMasses[0]; //mass difference actual-theoretical for the tallest peak (not necessarily the monoisotopic)

mzLib/MzLibUtil/PpmTolerance.cs

@@ -25,40 +25,35 @@ namespace MzLibUtil
     /// </summary>
     public class PpmTolerance : Tolerance
     {
+        private readonly double _factor;
+
         /// <summary>
-        /// Creates a new tolerance given a unit, value, and whether the tolerance is ±
+        /// Creates a new tolerance given value
         /// </summary>
-        /// <param name="unit">The units for this tolerance</param>
         /// <param name="value">The numerical value of the tolerance</param>
-        public PpmTolerance(double value) 
+        public PpmTolerance(double value)
             : base(value)
         {
+            _factor = value / 1e6;
         }
 
-        public override string ToString()
-        {
-            return $"{"±"}{Value.ToString("f4", System.Globalization.CultureInfo.InvariantCulture)} PPM";
-        }
+        public override string ToString() => $"\u00b1{Value.ToString("f4", System.Globalization.CultureInfo.InvariantCulture)} PPM";
 
         public override DoubleRange GetRange(double mean)
         {
-            double tol = Value * mean / 1e6;
+            double tol = _factor * mean;
             return new DoubleRange(mean - tol, mean + tol);
         }
 
-        public override double GetMinimumValue(double mean)
-        {
-            return mean * (1 - (Value / 1e6));
-        }
+        public override double GetMinimumValue(double mean) => mean * (1 - _factor);
 
-        public override double GetMaximumValue(double mean)
-        {
-            return mean * (1 + (Value / 1e6));
-        }
+        public override double GetMaximumValue(double mean) => mean * (1 + _factor);
 
         public override bool Within(double experimental, double theoretical)
         {
-            return Math.Abs((experimental - theoretical) / theoretical * 1e6) <= Value;
+            double diff = experimental - theoretical;
+            double scaledTolerance = theoretical * _factor;
+            return -scaledTolerance <= diff && diff <= scaledTolerance;
         }
     }
-}
+}

mzLib/Omics/BioPolymerWithSetModsExtensions.cs

@@ -112,7 +112,8 @@ public static string EssentialSequence(this IBioPolymerWithSetMods withSetMods,
     /// <returns></returns>
     public static string DetermineFullSequence(this IBioPolymerWithSetMods withSetMods)
     {
-        var subSequence = new StringBuilder();
+        // start string builder with initial capacity to avoid resizing costs. 
+        var subSequence = new StringBuilder(withSetMods.BaseSequence.Length + withSetMods.AllModsOneIsNterminus.Count * 20);
 
         // modification on peptide N-terminus
         if (withSetMods.AllModsOneIsNterminus.TryGetValue(1, out Modification mod))

mzLib/Omics/Digestion/DigestionProduct.cs

@@ -52,8 +52,17 @@ protected static IEnumerable<Dictionary<int, Modification>> GetVariableModificat
                 var possible_variable_modifications = new Dictionary<int, List<Modification>>(possibleVariableModifications);
 
                 int[] base_variable_modification_pattern = new int[peptideLength + 4];
-                var totalAvailableMods = possible_variable_modifications.Sum(b => b.Value == null ? 0 : b.Value.Count);
-                for (int variable_modifications = 0; variable_modifications <= Math.Min(totalAvailableMods, maxModsForPeptide); variable_modifications++)
+                int totalAvailableMods = 0;
+                foreach (var kvp in possible_variable_modifications)
+                {
+                    if (kvp.Value != null)
+                    {
+                        totalAvailableMods += kvp.Value.Count;
+                    }
+                }
+
+                int maxVariableMods = Math.Min(totalAvailableMods, maxModsForPeptide);
+                for (int variable_modifications = 0; variable_modifications <= maxVariableMods; variable_modifications++)
                 {
                     foreach (int[] variable_modification_pattern in GetVariableModificationPatterns(new List<KeyValuePair<int, List<Modification>>>(possible_variable_modifications),
                         possible_variable_modifications.Count - variable_modifications, base_variable_modification_pattern, 0))
@@ -76,18 +85,17 @@ protected Dictionary<int, Modification> GetFixedModsOneIsNorFivePrimeTerminus(in
                     case "Oligo 5'-terminal.":
                     case "N-terminal.":
                     case "Peptide N-terminal.":
+
                         //the modification is protease associated and is applied to the n-terminal cleaved residue, not at the beginign of the protein
-                        if (mod.ModificationType == "Protease" && ModificationLocalization.ModFits(mod, Parent.BaseSequence, 1, length, OneBasedStartResidue))
+                        if (ModificationLocalization.ModFits(mod, Parent.BaseSequence, 1, length, OneBasedStartResidue))
                         {
-                            if (OneBasedStartResidue != 1)
+                            if (mod.ModificationType == "Protease")
                             {
-                                fixedModsOneIsNterminus[2] = mod;
+                                if (OneBasedStartResidue != 1)
+                                    fixedModsOneIsNterminus[2] = mod;
                             }
-                        }
-                        //Normal N-terminal peptide modification
-                        else if (ModificationLocalization.ModFits(mod, Parent.BaseSequence, 1, length, OneBasedStartResidue))
-                        {
-                            fixedModsOneIsNterminus[1] = mod;
+                            else //Normal N-terminal peptide modification
+                                fixedModsOneIsNterminus[1] = mod;
                         }
                         break;
 
@@ -106,17 +114,15 @@ protected Dictionary<int, Modification> GetFixedModsOneIsNorFivePrimeTerminus(in
                     case "C-terminal.":
                     case "Peptide C-terminal.":
                         //the modification is protease associated and is applied to the c-terminal cleaved residue, not if it is at the end of the protein
-                        if (mod.ModificationType == "Protease" && ModificationLocalization.ModFits(mod, Parent.BaseSequence, length, length, OneBasedStartResidue + length - 1))
+                        if (ModificationLocalization.ModFits(mod, Parent.BaseSequence, length, length, OneBasedStartResidue + length - 1))
                         {
-                            if (OneBasedEndResidue != Parent.Length)
+                            if (mod.ModificationType == "Protease")
                             {
-                                fixedModsOneIsNterminus[length + 1] = mod;
+                                if (OneBasedEndResidue != Parent.Length)
+                                    fixedModsOneIsNterminus[length + 1] = mod;
                             }
-                        }
-                        //Normal C-terminal peptide modification 
-                        else if (ModificationLocalization.ModFits(mod, Parent.BaseSequence, length, length, OneBasedStartResidue + length - 1))
-                        {
-                            fixedModsOneIsNterminus[length + 2] = mod;
+                            else //Normal C-terminal peptide modification 
+                                fixedModsOneIsNterminus[length + 2] = mod;
                         }
                         break;
 
@@ -188,7 +194,5 @@ private static Dictionary<int, Modification> GetNewVariableModificationPattern(i
 
             return modification_pattern;
         }
-
-
     }
 }

mzLib/Omics/Modifications/ModificationLocalization.cs

@@ -2,45 +2,42 @@
 {
     public static class ModificationLocalization
     {
+        // This method is called a ton in MetaMorpheus. If changes are made, ensure they are efficient. 
         public static bool ModFits(Modification attemptToLocalize, string sequence, int digestionProductOneBasedIndex, int digestionProductLength, int bioPolymerOneBasedIndex)
         {
             // First find the capital letter...
-            var motif = attemptToLocalize.Target;
-            var motifStartLocation = motif.ToString().IndexOf(motif.ToString().First(b => char.IsUpper(b)));
+            var motif = attemptToLocalize.Target.ToString();
+            var motifStartLocation = motif.IndexOf(motif.First(char.IsUpper));
 
             // Look up starting at and including the capital letter
             var proteinToMotifOffset = bioPolymerOneBasedIndex - motifStartLocation - 1;
-            var indexUp = 0;
-            while (indexUp < motif.ToString().Length)
+            var motifLength = motif.Length;
+
+            for (int indexUp = 0; indexUp < motifLength; indexUp++)
             {
-                if (indexUp + proteinToMotifOffset < 0 || indexUp + proteinToMotifOffset >= sequence.Length
-                    || !MotifMatches(motif.ToString()[indexUp], sequence[indexUp + proteinToMotifOffset]))
+                int sequenceIndex = indexUp + proteinToMotifOffset;
+                if (sequenceIndex < 0 || sequenceIndex >= sequence.Length || !MotifMatches(motif[indexUp], sequence[sequenceIndex]))
                 {
                     return false;
                 }
-                indexUp++;
             }
-            switch (attemptToLocalize.LocationRestriction)
-            {
-                case "N-terminal." when bioPolymerOneBasedIndex > 2:
-                case "Peptide N-terminal." when digestionProductOneBasedIndex > 1:
-                case "C-terminal." when bioPolymerOneBasedIndex < sequence.Length:
-                case "Peptide C-terminal." when digestionProductOneBasedIndex < digestionProductLength:
-                case "5'-terminal." when bioPolymerOneBasedIndex > 2:
-                // first residue in oligo but not first in nucleic acid
-                case "Oligo 5'-terminal." when digestionProductOneBasedIndex > 1
-                                               || bioPolymerOneBasedIndex == 1:
-                case "3'-terminal." when bioPolymerOneBasedIndex < sequence.Length:
-                // not the last residue in oligo but not in nucleic acid
-                case "Oligo 3'-terminal." when digestionProductOneBasedIndex < digestionProductLength
-                                               || bioPolymerOneBasedIndex == sequence.Length:
-                    return false;
 
-                default:
+            return attemptToLocalize.LocationRestriction switch
+            {
+                "N-terminal." when bioPolymerOneBasedIndex > 2 => false,
+                "Peptide N-terminal." when digestionProductOneBasedIndex > 1 => false,
+                "C-terminal." when bioPolymerOneBasedIndex < sequence.Length => false,
+                "Peptide C-terminal." when digestionProductOneBasedIndex < digestionProductLength => false,
+                "5'-terminal." when bioPolymerOneBasedIndex > 2 => false,
+                    // first residue in oligo but not first in nucleic acid
+                "Oligo 5'-terminal." when digestionProductOneBasedIndex > 1 || bioPolymerOneBasedIndex == 1 => false,
+                "3'-terminal." when bioPolymerOneBasedIndex < sequence.Length => false,
+                    // last residue in oligo but not in nucleic acid
+                "Oligo 3'-terminal." when digestionProductOneBasedIndex < digestionProductLength || bioPolymerOneBasedIndex == sequence.Length => false,
                     // I guess Anywhere. and Unassigned. are true since how do you localize anywhere or unassigned.
-
-                    return true;
-            }
+                    
+                _ => true,
+            };
         }
 
         public static bool UniprotModExists(IBioPolymer bioPolymer, int i, Modification attemptToLocalize)
@@ -56,11 +53,19 @@ public static bool UniprotModExists(IBioPolymer bioPolymer, int i, Modification
         private static bool MotifMatches(char motifChar, char sequenceChar)
         {
             char upperMotifChar = char.ToUpper(motifChar);
-            return upperMotifChar.Equals('X')
-                || upperMotifChar.Equals(sequenceChar)
-                || upperMotifChar.Equals('B') && new[] { 'D', 'N' }.Contains(sequenceChar)
-                || upperMotifChar.Equals('J') && new[] { 'I', 'L' }.Contains(sequenceChar)
-                || upperMotifChar.Equals('Z') && new[] { 'E', 'Q' }.Contains(sequenceChar);
+            switch (upperMotifChar)
+            {
+                case 'X':
+                    return true;
+                case 'B':
+                    return sequenceChar is 'D' or 'N';
+                case 'J':
+                    return sequenceChar is 'I' or 'L';
+                case 'Z':
+                    return sequenceChar is 'E' or 'Q';
+                default:
+                    return upperMotifChar == sequenceChar;
+            }
         }
     }
 }