Set qcTraceName on ChromatogramGroupId (#982)

* Set qcTraceName on ChromatogramGroupId. * Fix error importing Skyline documents with Spectrum Filters (#921) "Spectrum Filter" gets parsed from the Skyline document light:heavy ratios are not calculated for precursors that have a spectrum filter The spectrum filter is used when deciding which chromatogram to read from the .skyd file for a particular precursor. # Conflicts: # src/org/labkey/targetedms/parser/ChromatogramGroupId.java * Added constructor for ChromatogramGroupId that takes both qcTraceName and spectrumFilter. --------- Co-authored-by: nickshulman <[email protected]>
LabKey · Nov 19, 2024 · e4c1920 · e4c1920
1 parent d886075
commit e4c1920
Show file tree

Hide file tree

Showing 13 changed files with 394 additions and 19 deletions.
diff --git a/resources/schemas/dbscripts/postgresql/targetedms-24.004-24.005.sql b/resources/schemas/dbscripts/postgresql/targetedms-24.004-24.005.sql
@@ -0,0 +1,2 @@
+ALTER TABLE targetedms.GeneralPrecursor
+    ADD COLUMN SpectrumFilter BYTEA;
diff --git a/resources/schemas/dbscripts/sqlserver/targetedms-24.004-24.005.sql b/resources/schemas/dbscripts/sqlserver/targetedms-24.004-24.005.sql
@@ -0,0 +1 @@
+ALTER TABLE targetedms.GeneralPrecursor ADD SpectrumFilter VARBINARY(MAX);
diff --git a/resources/schemas/targetedms.xml b/resources/schemas/targetedms.xml
@@ -438,6 +438,7 @@
             <column columnName="Modified">
                 <formatString>DateTime</formatString>
             </column>
+            <column columnName="SpectrumFilter" />
         </columns>
     </table>
     <table tableDbType="TABLE" tableName="MoleculePrecursor">

diff --git a/src/org/labkey/targetedms/SkylineDocImporter.java b/src/org/labkey/targetedms/SkylineDocImporter.java
@@ -1521,6 +1521,8 @@ else if (generalMolecule instanceof Molecule molecule)
                         {
                             if (precursor.getIsotopeLabelId() != numLabelId)
                                 continue;
+                            if (precursor.getSpectrumFilter() != null)
+                                continue;
 
                             PrecursorAreaRatio pRatio = areaRatioCalculator.getPrecursorAreaRatio(sampleFile.getId(),
                                     precursor,

diff --git a/src/org/labkey/targetedms/TargetedMSModule.java b/src/org/labkey/targetedms/TargetedMSModule.java
@@ -69,6 +69,7 @@
 import org.labkey.targetedms.folderImport.QCFolderWriterFactory;
 import org.labkey.targetedms.parser.Protein;
 import org.labkey.targetedms.parser.SampleFile;
+import org.labkey.targetedms.parser.SpectrumFilter;
 import org.labkey.targetedms.parser.skyaudit.SkylineAuditLogParser;
 import org.labkey.targetedms.passport.PassportController;
 import org.labkey.targetedms.pipeline.TargetedMSPipelineProvider;
@@ -222,7 +223,7 @@ public String getName()
     @Override
     public Double getSchemaVersion()
     {
-        return 24.004;
+        return 24.005;
     }
 
     @Override
@@ -668,7 +669,8 @@ public Set<Class> getUnitTests()
             TargetedMSController.TestCase.class,
             PrecursorManager.TestCase.class,
             CrossLinkedPeptideInfo.TestCase.class,
-            Protein.TestCase.class
+            Protein.TestCase.class,
+            SpectrumFilter.TestCase.class
         );
     }
 

diff --git a/src/org/labkey/targetedms/parser/ChromatogramGroupId.java b/src/org/labkey/targetedms/parser/ChromatogramGroupId.java
@@ -15,22 +15,32 @@
  */
 package org.labkey.targetedms.parser;
 
+import org.apache.commons.lang3.StringUtils;
 import org.labkey.targetedms.parser.proto.ChromatogramGroupDataOuterClass;
 
 import java.util.ArrayList;
 import java.util.List;
+import java.util.stream.Collectors;
 
 public class ChromatogramGroupId
 {
     private Target _target;
     private String _qcTraceName;
+    private SpectrumFilter _spectrumFilter;
 
     private ChromatogramGroupId()
     {
     }
-    public ChromatogramGroupId(Target target)
+    public ChromatogramGroupId(Target target, String qcTraceName, SpectrumFilter spectrumFilter)
     {
         _target = target;
+        _qcTraceName = StringUtils.isEmpty(qcTraceName) ? null : qcTraceName;
+        _spectrumFilter = spectrumFilter;
+    }
+
+    public ChromatogramGroupId(Target target, SpectrumFilter spectrumFilter)
+    {
+        this(target, null, spectrumFilter);
     }
 
     public Target getTarget()
@@ -43,6 +53,11 @@ public String getQcTraceName()
         return _qcTraceName;
     }
 
+    public SpectrumFilter getSpectrumFilter()
+    {
+        return _spectrumFilter;
+    }
+
     public static ChromatogramGroupId forQcTraceName(String qcTraceName)
     {
         ChromatogramGroupId chromatogramGroupId = new ChromatogramGroupId();
@@ -55,14 +70,24 @@ public static List<ChromatogramGroupId> fromProtos(ChromatogramGroupDataOuterCla
         List<Target> targets = new ArrayList<>();
         // Make one-based lookups easy
         targets.add(null);
+
+        List<SpectrumFilter.FilterClause> filterClauses = new ArrayList<>();
+
         for (ChromatogramGroupDataOuterClass.ChromatogramGroupIdsProto.Target target : proto.getTargetsList())
         {
             targets.add(new Target(target));
         }
+        for (ChromatogramGroupDataOuterClass.ChromatogramGroupIdsProto.SpectrumFilter spectrumFilter : proto.getFiltersList()) 
+        {
+            filterClauses.add(SpectrumFilter.FilterClause.fromProtocolMessage(spectrumFilter));
+        }
         List<ChromatogramGroupId> list = new ArrayList<>();
         for (ChromatogramGroupDataOuterClass.ChromatogramGroupIdsProto.ChromatogramGroupId chromatogramGroupId : proto.getChromatogramGroupIdsList())
         {
-            list.add(new ChromatogramGroupId(targets.get(chromatogramGroupId.getTargetIndex())));
+            SpectrumFilter spectrumFilter = SpectrumFilter.fromFilterClauses(
+                    chromatogramGroupId.getFilterIndexesList().stream()
+                    .map(filterClauses::get).collect(Collectors.toList())).orElse(null);
+            list.add(new ChromatogramGroupId(targets.get(chromatogramGroupId.getTargetIndex()), chromatogramGroupId.getQcTraceName(), spectrumFilter));
         }
         return list;
     }

diff --git a/src/org/labkey/targetedms/parser/GeneralPrecursor.java b/src/org/labkey/targetedms/parser/GeneralPrecursor.java
@@ -38,6 +38,7 @@ public class GeneralPrecursor<TransitionType extends GeneralTransition> extends
     private Double _explicitCcsSqa;
     private Double _explicitCompensationVoltage;
     private Double _precursorConcentration;
+    private byte[] _spectrumFilter;
 
     public long getGeneralMoleculeId()
     {
@@ -213,4 +214,14 @@ public void setPrecursorConcentration(Double precursorConcentration)
     {
         _precursorConcentration = precursorConcentration;
     }
+
+    public byte[] getSpectrumFilter()
+    {
+        return _spectrumFilter;
+    }
+
+    public void setSpectrumFilter(byte[] spectrumFilter)
+    {
+        _spectrumFilter = spectrumFilter;
+    }
 }
diff --git a/src/org/labkey/targetedms/parser/PeakAreaRatioCalculator.java b/src/org/labkey/targetedms/parser/PeakAreaRatioCalculator.java
@@ -58,6 +58,13 @@ public void init(Map<SkylineDocImporter.SampleFileKey, SampleFile> skylineIdSamp
 
         for(Precursor precursor: _peptide.getPrecursorList())
         {
+            if  (precursor.getSpectrumFilter() != null)
+            {
+                // Ideally, ratios would be calculated separately between Precursors with the same Spectrum Filter.
+                // However, we only store one ratio in the database for each Peptide, so we just skip precursors
+                // which have a Spectrum Filter.
+                continue;
+            }
             for(PrecursorChromInfo precursorChromInfo : precursor.getChromInfoList())
             {
                 if(precursorChromInfo.isOptimizationPeak())

diff --git a/src/org/labkey/targetedms/parser/SkylineBinaryParser.java b/src/org/labkey/targetedms/parser/SkylineBinaryParser.java
@@ -383,7 +383,7 @@ public ChromatogramGroupId getTextId(ChromGroupHeaderInfo chromGroupHeaderInfo)
         {
             return null;
         }
-        return new ChromatogramGroupId(target);
+        return new ChromatogramGroupId(target, null);
     }
 
     public String getFilePath(ChromGroupHeaderInfo chromGroupHeaderInfo) {

diff --git a/src/org/labkey/targetedms/parser/SkylineDocumentParser.java b/src/org/labkey/targetedms/parser/SkylineDocumentParser.java
@@ -17,6 +17,7 @@
 package org.labkey.targetedms.parser;
 
 import com.google.common.collect.Iterables;
+import com.google.protobuf.InvalidProtocolBufferException;
 import org.apache.commons.io.FilenameUtils;
 import org.apache.commons.lang3.StringUtils;
 import org.apache.logging.log4j.Logger;
@@ -63,6 +64,7 @@
 import java.util.LinkedList;
 import java.util.List;
 import java.util.Map;
+import java.util.Objects;
 import java.util.concurrent.atomic.AtomicInteger;
 import java.util.function.Function;
 import java.util.regex.Pattern;
@@ -157,6 +159,7 @@ public class SkylineDocumentParser implements AutoCloseable
     private static final String CHARGE = "charge" ;
     public static final String TRANSITION_DATA = "transition_data";
     private static final String LINKED_FRAGMENT_ION = "linked_fragment_ion";
+    private static final String SPECTRUM_FILTER = "spectrum_filter";
 
     private static final double MIN_SUPPORTED_VERSION = 1.2;
     public static final double MAX_SUPPORTED_VERSION = 23.1;
@@ -1860,7 +1863,7 @@ private MoleculePrecursor readMoleculePrecursor(XMLStreamReader reader, Molecule
         String charge = reader.getAttributeValue(null, CHARGE);
         if(null != charge)
             moleculePrecursor.setCharge(Integer.parseInt(charge));
-
+        List<SpectrumFilter.FilterClause> spectrumFilterClauses = new ArrayList<>();
         while(reader.hasNext())
         {
             int evtType = reader.next();
@@ -1879,8 +1882,11 @@ else if (XmlUtil.isStartElement(reader, evtType, ANNOTATION))
                 annotations.add(readAnnotation(reader, new PrecursorAnnotation()));
             else if (XmlUtil.isStartElement(reader, evtType, NOTE))
                 moleculePrecursor.setNote(readNote(reader));
+            else if (XmlUtil.isStartElement(reader, evtType, SPECTRUM_FILTER))
+                spectrumFilterClauses.add(SpectrumFilter.FilterClause.parse(reader));
         }
-
+        moleculePrecursor.setSpectrumFilter(SpectrumFilter.fromFilterClauses(spectrumFilterClauses)
+                .map(SpectrumFilter::toByteArray).orElse(null));
         List<ChromGroupHeaderInfo> chromatograms = tryLoadChromatogram(moleculeTransitionList, molecule, moleculePrecursor, _matchTolerance);
         populateChromInfoChromatograms(moleculePrecursor, chromatograms);
 
@@ -2035,7 +2041,7 @@ private Precursor readPrecursor(XMLStreamReader reader, Peptide peptide, boolean
         precursor.setExplicitCcsSqa(XmlUtil.readDoubleAttribute(reader, "explicit_ccs_sqa"));
         precursor.setExplicitCompensationVoltage(XmlUtil.readDoubleAttribute(reader, "explicit_compensation_voltage"));
         precursor.setPrecursorConcentration(XmlUtil.readDoubleAttribute(reader, "precursor_concentration"));
-
+        List<SpectrumFilter.FilterClause> spectrumFilterClauses = new ArrayList<>();
         while (reader.hasNext()) {
 
             int evtType = reader.next();
@@ -2083,8 +2089,15 @@ else if (XmlUtil.isStartElement(reader, evtType, NOTE))
             {
                 precursor.setNote(readNote(reader));
             }
+            else if (XmlUtil.isStartElement(reader, evtType, SPECTRUM_FILTER))
+            {
+                spectrumFilterClauses.add(SpectrumFilter.FilterClause.parse(reader));
+            }
         }
 
+        precursor.setSpectrumFilter(SpectrumFilter.fromFilterClauses(spectrumFilterClauses)
+                .map(SpectrumFilter::toByteArray).orElse(null));
+
         // Boolean type annotations are not listed in the .sky file if their value was false.
         // We would still like to store them in the database.
         List<String> missingBooleanAnnotations = _dataSettings.getMissingBooleanAnnotations(annotations,
@@ -3113,7 +3126,6 @@ public List<SampleFileChromInfo> getSampleFileChromInfos(Map<String, SampleFile>
         {
             return Collections.emptyList();
         }
-        int traceMetricIndex = 1;
         for (ChromGroupHeaderInfo chromatogram : _binaryParser.getChromatograms())
         {
             // Sample-scoped chromatograms have a magic precursor MZ value
@@ -3145,14 +3157,7 @@ public List<SampleFileChromInfo> getSampleFileChromInfos(Map<String, SampleFile>
                     ChromatogramGroupId chromatogramGroupId = _binaryParser.getTextId(chromatogram);
                     if (chromatogramGroupId != null)
                     {
-                        if (chromatogramGroupId.getQcTraceName() == null && chromatogram.getFlagValues().contains(ChromGroupHeaderInfo.FlagValues.extracted_qc_trace))
-                        {
-                            info.setTextId("QC Trace " + traceMetricIndex++);
-                        }
-                        else
-                        {
-                            info.setTextId(chromatogramGroupId.getQcTraceName());
-                        }
+                        info.setTextId(chromatogramGroupId.getQcTraceName());
                     }
                     info.setChromatogramFormat(chromatogram.getChromatogramBinaryFormat().ordinal());
                     info.setChromatogramOffset(chromatogram.getLocationPoints());
@@ -3204,8 +3209,24 @@ private List<ChromGroupHeaderInfo> tryLoadChromatogram(
                 ChromGroupHeaderInfo chrom = _binaryParser.getChromatograms()[i++];
                 // Sequence matching for extracted chromatogram data added in v1.5
                 ChromatogramGroupId chromTextId = _binaryParser.getTextId(chrom);
-                if (chromTextId != null && !molecule.targetMatches(chromTextId.getTarget()))
-                    continue;
+                if (chromTextId != null) 
+                {
+                    if (!molecule.targetMatches(chromTextId.getTarget()))
+                        continue;
+                    try
+                    {
+                        SpectrumFilter spectrumFilter = SpectrumFilter.fromByteArray(precursor.getSpectrumFilter());
+                        if (!Objects.equals(spectrumFilter, chromTextId.getSpectrumFilter())) 
+                        {
+                            continue;
+                        }
+                    }
+                    catch (InvalidProtocolBufferException e)
+                    {
+                        _log.warn("Error parsing spectrum filter {}", e);
+                        return Collections.emptyList();
+                    }
+                }
 
                 // If explicit retention time info is available, use that to discard obvious mismatches
                 if (explicitRT == null || !chrom.excludesTime(explicitRT))
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		ALTER TABLE targetedms.GeneralPrecursor
		ADD COLUMN SpectrumFilter BYTEA;
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		ALTER TABLE targetedms.GeneralPrecursor ADD SpectrumFilter VARBINARY(MAX);