Issue 55042, Issue 55036: Peptide Map Excel cross-link rows and early-stage PTM report improvements

resources/queries/targetedms/UngroupedMissedCleavagePeptides.query.xml

@@ -0,0 +1,3 @@
+<query xmlns="http://labkey.org/data/xml/query">
+    <description>Missed-cleavage peptides lacking a grouping. PTM reports show these as separate rows instead of combining them per site; set a shared Attribute Group ID in Skyline to merge them.</description>
+</query>

resources/queries/targetedms/UngroupedMissedCleavagePeptides.sql

@@ -0,0 +1,40 @@
+/*
+ * Copyright (c) 2026 LabKey Corporation
+ *
+ * Licensed under the Apache License, Version 2.0: http://www.apache.org/licenses/LICENSE-2.0
+ */
+-- Missed-cleavage peptides whose Attribute Group ID doesn't match any shorter peptide they contain, so PTM reports split the site across rows
+SELECT
+  l.RunId,
+  l.ProteinId,
+  l.Protein,
+  l.Sequence AS MissedCleavageSequence,
+  l.StartIndex,
+  l.EndIndex,
+  l.NumMissedCleavages,
+  l.GroupKey AS CurrentGroup,
+  GROUP_CONCAT(s.Sequence, ', ') AS ContainedPeptides
+FROM (
+  SELECT DISTINCT p.PeptideGroupId.RunId AS RunId, p.PeptideGroupId AS ProteinId, p.PeptideGroupId.Label AS Protein,
+    p.Sequence, p.StartIndex, p.EndIndex, p.NumMissedCleavages,
+    -- Skyline groups by unmodified sequence unless Attribute Group ID is set
+    COALESCE(NULLIF(p.AttributeGroupId, ''), p.Sequence) AS GroupKey
+  FROM Peptide p
+  WHERE p.Decoy IS NOT TRUE
+    AND p.NumMissedCleavages > 0
+    AND p.PeptideGroupId.RunId.ReplacedByRun IS NULL
+) l
+INNER JOIN (
+  SELECT DISTINCT p.PeptideGroupId AS ProteinId, p.Sequence, p.StartIndex, p.EndIndex, p.NumMissedCleavages,
+    COALESCE(NULLIF(p.AttributeGroupId, ''), p.Sequence) AS GroupKey
+  FROM Peptide p
+  WHERE p.Decoy IS NOT TRUE
+) s
+  ON s.ProteinId = l.ProteinId
+ AND s.Sequence <> l.Sequence
+ AND s.StartIndex >= l.StartIndex
+ AND s.EndIndex <= l.EndIndex
+ AND s.NumMissedCleavages < l.NumMissedCleavages
+GROUP BY l.RunId, l.ProteinId, l.Protein, l.Sequence, l.StartIndex, l.EndIndex, l.NumMissedCleavages, l.GroupKey
+-- A missed-cleavage peptide spans several cleaved peptides but can only join one group
+HAVING SUM(CASE WHEN s.GroupKey = l.GroupKey THEN 1 ELSE 0 END) = 0

src/org/labkey/targetedms/TargetedMSController.java

@@ -249,6 +249,7 @@
 import org.labkey.targetedms.query.SampleFileTable;
 import org.labkey.targetedms.query.SkylineListManager;
 import org.labkey.targetedms.query.SkylineListSchema;
+import org.labkey.targetedms.query.TargetedMSQueryView;
 import org.labkey.targetedms.query.TargetedMSTable;
 import org.labkey.targetedms.query.TransitionManager;
 import org.labkey.targetedms.search.ModificationSearchWebPart;
@@ -5353,7 +5354,12 @@ protected QueryView createQueryView(RunDetailsForm form, BindException errors, b
         {
             QuerySettings settings = new QuerySettings(getViewContext(), _dataRegionName, "PeptideIds");
             settings.setBaseFilter(new SimpleFilter(FieldKey.fromParts("PeptideGroupId", "RunId"), form.getId()));
-            return createViewWithNoContainerFilterOptions(settings, errors);
+            QueryView result = createViewWithNoContainerFilterOptions(settings, errors);
+            if (result instanceof TargetedMSQueryView view)
+            {
+                view.setSplitCrossLinkedPeptides(true);
+            }
+            return result;
         }
     }
 

src/org/labkey/targetedms/query/CDRConditionalFormattingDisplayColumnFactory.java

@@ -39,22 +39,25 @@ class CDRConditionalFormattingDisplayColumnFactory implements DisplayColumnFacto
     private final Function<Long, Protein> _proteinGetter;
     private final Map<Pair<String, Long>, Pair<Boolean, String>> _stressedSamples;
     private final long _runId;
+    private final Set<String> _terminalModifications;
 
-    public CDRConditionalFormattingDisplayColumnFactory(Function<Long, Protein> proteinGetter, Map<Pair<String, Long>, Pair<Boolean, String>> stressedSamples, long runId)
+    public CDRConditionalFormattingDisplayColumnFactory(Function<Long, Protein> proteinGetter, Map<Pair<String, Long>, Pair<Boolean, String>> stressedSamples, long runId, Set<String> terminalModifications)
     {
         _proteinGetter = proteinGetter;
         _stressedSamples = stressedSamples;
         _runId = runId;
+        _terminalModifications = terminalModifications;
     }
 
-    public static RiskLevel getRiskLevel(Number value, boolean inCDR, boolean stressed, String modification, String modifiedSequence)
+    public static RiskLevel getRiskLevel(Number value, boolean inCDR, boolean stressed, boolean terminalModification, String modifiedSequence)
     {
         if (value == null)
         {
             return null;
         }
 
-        if ("Gln->pyro-Glu (N-term Q)".equalsIgnoreCase(modification))
+        // Issue 55036: N- and C-terminal modifications aren't risk-scored
+        if (terminalModification)
         {
             return null;
         }
@@ -120,7 +123,7 @@ public static boolean isInCDR(FieldKey parentFieldKey, RenderContext ctx, Functi
     @Override
     public DisplayColumn createRenderer(ColumnInfo boundCol)
     {
-        return new Col(boundCol, _proteinGetter, _stressedSamples, _runId);
+        return new Col(boundCol, _proteinGetter, _stressedSamples, _runId, _terminalModifications);
     }
 
     public enum RiskLevel
@@ -150,13 +153,15 @@ private static class Col extends DataColumn
         private final Function<Long, Protein> _proteinGetter;
         private final Map<Pair<String, Long>, Pair<Boolean, String>> _stressedSamples;
         private final long _runId;
+        private final Set<String> _terminalModifications;
 
-        public Col(ColumnInfo colInfo, Function<Long, Protein> proteinGetter, Map<Pair<String, Long>, Pair<Boolean, String>> stressedSamples, long runId)
+        public Col(ColumnInfo colInfo, Function<Long, Protein> proteinGetter, Map<Pair<String, Long>, Pair<Boolean, String>> stressedSamples, long runId, Set<String> terminalModifications)
         {
             super(colInfo);
             _proteinGetter = proteinGetter;
             _stressedSamples = stressedSamples;
             _runId = runId;
+            _terminalModifications = terminalModifications;
         }
 
         @Override
@@ -173,7 +178,7 @@ public void addQueryFieldKeys(Set<FieldKey> keys)
         @Nullable
         private ConditionalFormat createConditionalFormat(Number value, boolean inCDR, boolean stressed, String modification, String modifiedSequence)
         {
-            RiskLevel level = getRiskLevel(value, inCDR, stressed, modification, modifiedSequence);
+            RiskLevel level = getRiskLevel(value, inCDR, stressed, modification != null && _terminalModifications.contains(modification), modifiedSequence);
             if (level == null)
             {
                 return null;

src/org/labkey/targetedms/query/CrossLinkedPeptideDisplayColumn.java

@@ -17,6 +17,7 @@
 
 import org.apache.commons.lang3.StringUtils;
 import org.jetbrains.annotations.NotNull;
+import org.jetbrains.annotations.Nullable;
 import org.junit.Assert;
 import org.junit.Test;
 import org.labkey.api.collections.LongHashMap;
@@ -34,6 +35,7 @@
 import java.util.List;
 import java.util.Map;
 import java.util.Set;
+import java.util.stream.Collectors;
 
 /**
  * Renders the cross-linked peptide info in a few different variants. For unlinked peptides, passes through the primary
@@ -87,34 +89,41 @@ public Object getValue(RenderContext ctx)
             CrossLinkedPeptideInfo i = new CrossLinkedPeptideInfo(modifiedSequence);
             List<Protein> proteins = getProteinsFromRun(ctx);
             List<CrossLinkedPeptideInfo.PeptideSequence> sequences = i.getAllSequences();
-            return render(sequences, proteins, modification);
+            Integer peptideIndex = ctx.get(CrossLinkedPeptideExcelWriter.PEPTIDE_INDEX) instanceof Integer index ? index : null;
+            return render(sequences, proteins, modification, peptideIndex);
         }
         return defaultValue;
     }
 
-    protected @NotNull String render(List<CrossLinkedPeptideInfo.PeptideSequence> sequences, List<Protein> proteins, String modification)
+    /** @param peptideIndex non-null when the Excel export is writing each cross-linked peptide on its own row */
+    protected @NotNull String render(List<CrossLinkedPeptideInfo.PeptideSequence> sequences, List<Protein> proteins, String modification, @Nullable Integer peptideIndex)
     {
-        StringBuilder result = new StringBuilder();
-        String outerSeparator = "";
+        return formatAll(sequences, proteins, modification).stream()
+                .filter(value -> !value.isEmpty())
+                .collect(Collectors.joining("\n"));
+    }
 
-        for (CrossLinkedPeptideInfo.PeptideSequence sequence : sequences)
-        {
-            List<CrossLinkedPeptideInfo.Match> matches = sequence.findMatches(proteins);
-            if (!matches.isEmpty())
-            {
-                result.append(outerSeparator);
-                outerSeparator = "\n";
+    /** @return the formatted value for each peptide, blank when it matches no protein */
+    protected @NotNull List<String> formatAll(List<CrossLinkedPeptideInfo.PeptideSequence> sequences, List<Protein> proteins, String modification)
+    {
+        return sequences.stream().map(sequence -> formatSequence(sequence, proteins, modification)).toList();
+    }
 
-                String innerSeparator = "";
-                for (CrossLinkedPeptideInfo.Match match : matches)
-                {
-                    result.append(innerSeparator);
-                    innerSeparator = "; ";
-                    result.append(_formatter.format(match, sequence, modification));
-                }
-            }
+    protected @NotNull List<String> getPeptideValues(RenderContext ctx)
+    {
+        String modifiedSequence = ctx.get(getModifiedSequenceFieldKey(), String.class);
+        if (modifiedSequence == null)
+        {
+            return List.of();
         }
-        return result.toString();
+        return formatAll(new CrossLinkedPeptideInfo(modifiedSequence).getAllSequences(), getProteinsFromRun(ctx), ctx.get(getModificationFieldKey(), String.class));
+    }
+
+    protected @NotNull String formatSequence(CrossLinkedPeptideInfo.PeptideSequence sequence, List<Protein> proteins, String modification)
+    {
+        return sequence.findMatches(proteins).stream()
+                .map(match -> _formatter.format(match, sequence, modification))
+                .collect(Collectors.joining("; "));
     }
 
     private List<Protein> getProteinsFromRun(RenderContext ctx)
@@ -127,7 +136,7 @@ private List<Protein> getProteinsFromRun(RenderContext ctx)
         return Collections.emptyList();
     }
 
-    private FieldKey getModifiedSequenceFieldKey()
+    protected FieldKey getModifiedSequenceFieldKey()
     {
         return FieldKey.fromString(getColumnInfo().getFieldKey().getParent(), "PeptideModifiedSequence");
     }
@@ -151,12 +160,69 @@ public void addQueryFieldKeys(Set<FieldKey> keys)
         keys.add(getRunIdFieldKey());
     }
 
+    /** Renders one value per cross-linked peptide, so the Excel export puts each on its own row */
+    private static class PerPeptideColumn extends CrossLinkedPeptideDisplayColumn implements CrossLinkedPeptideExcelWriter.PerPeptideColumn
+    {
+        private PerPeptideColumn(ColumnInfo col, Formatter formatter)
+        {
+            super(col, formatter);
+        }
+
+        @Override
+        public int getPeptideCount(RenderContext ctx)
+        {
+            String modifiedSequence = ctx.get(getModifiedSequenceFieldKey(), String.class);
+            return modifiedSequence == null ? 1 : new CrossLinkedPeptideInfo(modifiedSequence).getAllSequences().size();
+        }
+
+        @Override
+        protected @NotNull String render(List<CrossLinkedPeptideInfo.PeptideSequence> sequences, List<Protein> proteins, String modification, @Nullable Integer peptideIndex)
+        {
+            if (peptideIndex == null)
+            {
+                return super.render(sequences, proteins, modification, null);
+            }
+            return peptideIndex < sequences.size() ? formatSequence(sequences.get(peptideIndex), proteins, modification) : "";
+        }
+    }
+
+    /** Splits across the Excel rows only when the cross-linked peptides are on different chains */
+    private static class ChainColumn extends PerPeptideColumn
+    {
+        private ChainColumn(ColumnInfo col)
+        {
+            super(col, (match, _, _) -> match.protein().getLabel() == null ? match.protein().getName() : match.protein().getLabel());
+        }
+
+        @Override
+        public boolean isSplit(RenderContext ctx)
+        {
+            return getDistinctChains(getPeptideValues(ctx)).size() > 1;
+        }
+
+        @Override
+        protected @NotNull String render(List<CrossLinkedPeptideInfo.PeptideSequence> sequences, List<Protein> proteins, String modification, @Nullable Integer peptideIndex)
+        {
+            List<String> chains = getDistinctChains(formatAll(sequences, proteins, modification));
+            if (peptideIndex == null || chains.size() > 1)
+            {
+                return super.render(sequences, proteins, modification, peptideIndex);
+            }
+            return String.join("\n", chains);
+        }
+
+        private static List<String> getDistinctChains(List<String> values)
+        {
+            return values.stream().filter(value -> !value.isEmpty()).distinct().toList();
+        }
+    }
+
     public static class ChainFactory implements DisplayColumnFactory
     {
         @Override
         public DisplayColumn createRenderer(ColumnInfo colInfo)
         {
-            return new CrossLinkedPeptideDisplayColumn(colInfo, (match, sequence, modification) -> match.protein().getLabel() == null ? match.protein().getName() : match.protein().getLabel());
+            return new ChainColumn(colInfo);
         }
     }
 
@@ -165,7 +231,7 @@ public static class PeptideLocationFactory implements DisplayColumnFactory
         @Override
         public DisplayColumn createRenderer(ColumnInfo colInfo)
         {
-            return new CrossLinkedPeptideDisplayColumn(colInfo, (match, sequence, modification) ->
+            return new PerPeptideColumn(colInfo, (match, sequence, modification) ->
                 (match.index() + 1) + "-" + (match.index() + getPeptideLength(sequence.getUnmodified(), modification)));
         }
     }
@@ -185,7 +251,7 @@ public static class PeptideIdentityFactory implements DisplayColumnFactory
         @Override
         public DisplayColumn createRenderer(ColumnInfo colInfo)
         {
-            return new CrossLinkedPeptideDisplayColumn(colInfo, (match, sequence, modification) ->
+            return new PerPeptideColumn(colInfo, (match, sequence, modification) ->
             {
                 String label = match.protein().getLabel();
                 String prefix = getChainPrefix(label);
@@ -224,7 +290,7 @@ public DisplayColumn createRenderer(ColumnInfo colInfo)
             return new CrossLinkedPeptideDisplayColumn(colInfo, null)
             {
                 @Override
-                protected @NotNull String render(List<CrossLinkedPeptideInfo.PeptideSequence> sequences, List<Protein> proteins, String modification)
+                protected @NotNull String render(List<CrossLinkedPeptideInfo.PeptideSequence> sequences, List<Protein> proteins, String modification, @Nullable Integer peptideIndex)
                 {
                     List<List<String>> allBonds = new ArrayList<>();