private void RunMultiplePeptidesSameMz(RefinementSettings.ConvertToSmallMoleculesMode asSmallMolecules)
{
if (asSmallMolecules != RefinementSettings.ConvertToSmallMoleculesMode.none)
{
TestDirectoryName = asSmallMolecules.ToString();
}
TestSmallMolecules = false; // Don't need the magic test node, we have an explicit test
var testFilesDir = new TestFilesDir(TestContext, ZIP_FILE);
string docPath;
SrmDocument document = InitMultiplePeptidesSameMzDocument(testFilesDir, out docPath);
document = (new RefinementSettings()).ConvertToSmallMolecules(document, asSmallMolecules);
var docContainer = new ResultsTestDocumentContainer(document, docPath);
var doc = docContainer.Document;
var listChromatograms = new List <ChromatogramSet>();
var path = MsDataFileUri.Parse(@"AMultiplePeptidesSameMz\ljz_20131201k_Newvariant_standards_braf.mzML");
listChromatograms.Add(AssertResult.FindChromatogramSet(doc, path) ??
new ChromatogramSet(path.GetFileName().Replace('.', '_'), new[] { path }));
var docResults = doc.ChangeMeasuredResults(new MeasuredResults(listChromatograms));
Assert.IsTrue(docContainer.SetDocument(docResults, doc, true));
docContainer.AssertComplete();
document = docContainer.Document;
float tolerance = (float)document.Settings.TransitionSettings.Instrument.MzMatchTolerance;
var results = document.Settings.MeasuredResults;
foreach (var pair in document.MoleculePrecursorPairs)
{
ChromatogramGroupInfo[] chromGroupInfo;
Assert.IsTrue(results.TryLoadChromatogram(0, pair.NodePep, pair.NodeGroup,
tolerance, true, out chromGroupInfo));
Assert.AreEqual(1, chromGroupInfo.Length); // without the fix, only the first pair will have a chromatogram
}
// now drill down for specific values
int nPeptides = 0;
foreach (var nodePep in document.Molecules.Where(nodePep => nodePep.Results[0] != null))
{
// expecting three peptide result in this small data set
if (nodePep.Results[0].Sum(chromInfo => chromInfo.PeakCountRatio > 0 ? 1 : 0) > 0)
{
Assert.AreEqual(34.2441024780273, (double)nodePep.GetMeasuredRetentionTime(0), .0001);
nPeptides++;
}
}
Assert.AreEqual(3, nPeptides); // without the fix this will give just one result
// Release file handles
docContainer.Release();
testFilesDir.Dispose();
}
private SrmDocument InitRefineDocument(RefinementSettings.ConvertToSmallMoleculesMode mode)
{
TestFilesDir testFilesDir = new TestFilesDir(TestContext, @"TestData\Refine.zip", mode.ToString());
if (mode == RefinementSettings.ConvertToSmallMoleculesMode.none)
{
var doc = ResultsUtil.DeserializeDocument(testFilesDir.GetTestPath("SRM_mini.sky"));
AssertEx.IsDocumentState(doc, null, 4, 36, 38, 334);
return(doc);
}
var docPath = testFilesDir.GetTestPath("SRM_mini_single_replicate.sky");
var dataPaths = new[] { testFilesDir.GetTestPath("worm1.mzML") };
var converted = ConvertToSmallMolecules(null, ref docPath, dataPaths, mode);
AssertEx.IsDocumentState(converted, null, 4, 36, 38, 334);
return(converted);
}
public void RunTestFindNode(RefinementSettings.ConvertToSmallMoleculesMode asSmallMolecules)
{
if (asSmallMolecules != RefinementSettings.ConvertToSmallMoleculesMode.none)
{
TestDirectoryName = asSmallMolecules.ToString();
}
SrmDocument doc = CreateStudy7Doc();
doc = new RefinementSettings().ConvertToSmallMolecules(doc, TestDirectoryName, asSmallMolecules);
var displaySettings = new DisplaySettings(null, false, 0, 0); //, ProteinDisplayMode.ByName);
// Find every other transition, searching down.
List <TransitionDocNode> listTransitions = doc.MoleculeTransitions.ToList();
var pathFound = doc.GetPathTo(0, 0);
int i;
for (i = 0; i < doc.MoleculeTransitionCount; i += 2)
{
pathFound = doc.SearchDocumentForString(pathFound, String.Format("{0:F04}", listTransitions[i].Mz), displaySettings, false, false);
Assert.AreEqual(doc.GetPathTo((int)SrmDocument.Level.Transitions, i), pathFound);
}
// Test wrapping in search down.
pathFound = doc.SearchDocumentForString(pathFound, String.Format("{0:F04}", listTransitions[0].Mz), displaySettings, false, false);
Assert.AreEqual(doc.GetPathTo((int)SrmDocument.Level.Transitions, 0), pathFound);
// Find every other peptide searching up while for each finding one of its children searching down.
pathFound = doc.LastNodePath;
List <PeptideDocNode> listPeptides = new List <PeptideDocNode>();
listPeptides.AddRange(doc.Molecules);
List <TransitionGroupDocNode> listTransitionGroups = new List <TransitionGroupDocNode>();
listTransitionGroups.AddRange(doc.MoleculeTransitionGroups);
for (int x = doc.MoleculeCount; x > 0; x -= 2)
{
// Test case insensitivity.
pathFound = doc.SearchDocumentForString(pathFound, listPeptides[x - 1].ToString().ToLower(), displaySettings, true, false);
Assert.AreEqual(doc.GetPathTo((int)SrmDocument.Level.Molecules, x - 1), pathFound);
// Test parents can find children.
pathFound = doc.SearchDocumentForString(pathFound, String.Format("{0:F04}", listTransitionGroups[x * 2 - 1].PrecursorMz.Value), displaySettings,
false, true);
Assert.AreEqual(doc.GetPathTo((int)SrmDocument.Level.TransitionGroups, x * 2 - 1), pathFound);
// Test Children can find parents.
pathFound = doc.SearchDocumentForString(pathFound, listPeptides[x - 1].ToString().ToLower(), displaySettings, true, false);
Assert.AreEqual(doc.GetPathTo((int)SrmDocument.Level.Molecules, x - 1), pathFound);
}
// Test wrapping in search up.
pathFound = doc.SearchDocumentForString(pathFound, String.Format("{0:F04}", listTransitionGroups[listTransitionGroups.Count - 1].PrecursorMz.Value),
displaySettings, false, true);
Assert.AreEqual(doc.GetPathTo((int)SrmDocument.Level.TransitionGroups, listTransitionGroups.Count - 1), pathFound);
// Test children can find other parents.
pathFound = doc.SearchDocumentForString(pathFound, listPeptides[0].ToString().ToLowerInvariant(), displaySettings, true, false);
Assert.AreEqual(doc.GetPathTo((int)SrmDocument.Level.Molecules, 0), pathFound);
// Test forward and backward searching in succession
const string heavyText = "heavy";
int countHeavyForward = CountOccurrances(doc, heavyText, displaySettings, false, true);
Assert.IsTrue(countHeavyForward > 0);
Assert.AreEqual(countHeavyForward, CountOccurrances(doc, heavyText, displaySettings, true, true));
// More tests of case insensitive searching
Assert.AreEqual(0, CountOccurrances(doc, heavyText.ToUpperInvariant(), displaySettings, false, true));
Assert.AreEqual(countHeavyForward, CountOccurrances(doc, heavyText.ToUpperInvariant(), displaySettings, false, false));
if (asSmallMolecules != RefinementSettings.ConvertToSmallMoleculesMode.masses_only)
{
Assert.AreEqual(1, CountOccurrances(doc, "hgflpr", displaySettings, true, false));
}
// Test mismatched transitions finder
var missmatchFinder = new FindOptions().ChangeCustomFinders(new[] { new MismatchedIsotopeTransitionsFinder() });
Assert.AreEqual(4, CountOccurrances(doc, missmatchFinder, displaySettings));
var docRemoved = (SrmDocument)doc.RemoveChild(doc.Children[1]).RemoveChild(doc.Children[2]);
Assert.AreEqual(0, CountOccurrances(docRemoved, missmatchFinder, displaySettings));
var refineRemoveHeavy = new RefinementSettings {
RefineLabelType = IsotopeLabelType.heavy
};
var docLight = refineRemoveHeavy.Refine(doc);
Assert.AreEqual(0, CountOccurrances(docLight, missmatchFinder, displaySettings));
var refineRemoveLight = new RefinementSettings {
RefineLabelType = IsotopeLabelType.light
};
var docHeavy = refineRemoveLight.Refine(doc);
Assert.AreEqual(0, CountOccurrances(docHeavy, missmatchFinder, displaySettings));
var docMulti = ResultsUtil.DeserializeDocument("MultiLabel.sky", GetType());
docMulti = (new RefinementSettings()).ConvertToSmallMolecules(docMulti, TestContext.TestDir, asSmallMolecules);
Assert.AreEqual(0, CountOccurrances(docMulti, missmatchFinder, displaySettings));
var pathTranMultiRemove = docMulti.GetPathTo((int)SrmDocument.Level.Transitions, 7);
var tranMultiRemove = docMulti.FindNode(pathTranMultiRemove);
var docMultiRemoved = (SrmDocument)docMulti.RemoveChild(pathTranMultiRemove.Parent, tranMultiRemove);
Assert.AreEqual(2, CountOccurrances(docMultiRemoved, missmatchFinder, displaySettings));
var tranGroupMultiRemove = docMulti.FindNode(pathTranMultiRemove.Parent);
var docMultiGroupRemoved = (SrmDocument)
docMulti.RemoveChild(pathTranMultiRemove.Parent.Parent, tranGroupMultiRemove);
Assert.AreEqual(0, CountOccurrances(docMultiGroupRemoved, missmatchFinder, displaySettings));
}
private void WatersImsMseChromatogramTest(DriftFilterType mode,
IonMobilityWindowWidthCalculator.IonMobilityPeakWidthType driftPeakWidthCalcType,
RefinementSettings.ConvertToSmallMoleculesMode asSmallMolecules = RefinementSettings.ConvertToSmallMoleculesMode.none)
{
if (asSmallMolecules != RefinementSettings.ConvertToSmallMoleculesMode.none && !RunSmallMoleculeTestVersions)
{
Console.Write(MSG_SKIPPING_SMALLMOLECULE_TEST_VERSION);
return;
}
string subdir = (asSmallMolecules == RefinementSettings.ConvertToSmallMoleculesMode.none) ? null : asSmallMolecules.ToString();
var testFilesDir = new TestFilesDir(TestContext, ZIP_FILE, subdir);
TestSmallMolecules = false; // Don't need that extra magic node
bool withDriftTimePredictor = (mode == DriftFilterType.predictor); // Load the doc that has a drift time predictor?
bool withDriftTimeFilter = (mode != DriftFilterType.none); // Perform drift time filtering? (either with predictor, or with bare times in blib file)
string docPath;
SrmDocument document = InitWatersImsMseDocument(testFilesDir, driftPeakWidthCalcType, withDriftTimeFilter, withDriftTimePredictor, out docPath);
AssertEx.IsDocumentState(document, null, 1, 1, 1, 8); // Drift time lib load bumps the doc version, so does small mol conversion
var listChromatograms = new List <ChromatogramSet>();
// A small subset of the QC_HDMSE_02_UCA168_3495_082213 data set (RT 21.5-22.5) from Will Thompson
string mz5Path = "waters-mobility" + ExtensionTestContext.ExtMz5;
string testModeStr = withDriftTimePredictor ? "with drift time predictor" : "without drift time info";
if (withDriftTimeFilter && !withDriftTimePredictor)
{
testModeStr = "with drift times from spectral library";
}
listChromatograms.Add(AssertResult.FindChromatogramSet(document, new MsDataFilePath(mz5Path)) ??
new ChromatogramSet(Path.GetFileName(mz5Path).Replace('.', '_'), new[] { mz5Path }));
using (var docContainer = new ResultsTestDocumentContainer(document, docPath))
{
var doc = docContainer.Document;
var docResults = doc.ChangeMeasuredResults(new MeasuredResults(listChromatograms));
Assume.IsTrue(docContainer.SetDocument(docResults, doc, true));
docContainer.AssertComplete();
document = docContainer.Document;
}
document = ConvertToSmallMolecules(document, ref docPath, new[] { mz5Path }, asSmallMolecules);
using (var docContainer = new ResultsTestDocumentContainer(document, docPath))
{
float tolerance = (float)document.Settings.TransitionSettings.Instrument.MzMatchTolerance;
double maxHeight = 0;
var results = document.Settings.MeasuredResults;
Assume.AreEqual(1, document.MoleculePrecursorPairs.Count());
foreach (var pair in document.MoleculePrecursorPairs)
{
ChromatogramGroupInfo[] chromGroupInfo;
Assume.IsTrue(results.TryLoadChromatogram(0, pair.NodePep, pair.NodeGroup,
tolerance, true, out chromGroupInfo));
Assume.AreEqual(1, chromGroupInfo.Length, testModeStr);
var chromGroup = chromGroupInfo[0];
int expectedPeaks;
if (withDriftTimeFilter)
{
expectedPeaks = 3;
}
else if (asSmallMolecules != RefinementSettings.ConvertToSmallMoleculesMode.masses_only)
{
expectedPeaks = 5;
}
else
{
expectedPeaks = 6; // No libraries
}
Assume.AreEqual(expectedPeaks, chromGroup.NumPeaks, testModeStr); // This will be higher if we don't filter on DT
foreach (var tranInfo in chromGroup.TransitionPointSets)
{
maxHeight = Math.Max(maxHeight, tranInfo.MaxIntensity);
}
}
Assume.AreEqual(withDriftTimeFilter ? 5226 : 20075, maxHeight, 1, testModeStr); // Without DT filtering, this will be much greater
// now drill down for specific values
int nPeptides = 0;
foreach (var nodePep in document.Molecules.Where(nodePep => !nodePep.Results[0].IsEmpty))
{
// expecting just one peptide result in this small data set
if (nodePep.Results[0].Sum(chromInfo => chromInfo.PeakCountRatio > 0 ? 1 : 0) > 0)
{
Assume.AreEqual(21.94865, (double)nodePep.GetMeasuredRetentionTime(0), .0001, testModeStr);
Assume.AreEqual(1.0, (double)nodePep.GetPeakCountRatio(0), 0.0001, testModeStr);
nPeptides++;
}
}
Assume.AreEqual(1, nPeptides);
if (withDriftTimePredictor || withDriftTimeFilter)
{
// Verify that the .imdb pr .blib file goes out in the share zipfile
for (int complete = 0; complete <= 1; complete++)
{
var sharePath =
testFilesDir.GetTestPath(complete == 1 ? "share_complete.zip" : "share_minimized.zip");
var share = new SrmDocumentSharing(document, docPath, sharePath,
new ShareType(complete == 1, SkylineVersion.CURRENT)); // Explicitly declaring version number forces a save before zip
share.Share(new SilentProgressMonitor());
var files = share.ListEntries().ToArray();
var imdbFile = withDriftTimePredictor ? "scaled.imdb" : "waters-mobility.filtered-scaled.blib";
if (asSmallMolecules != RefinementSettings.ConvertToSmallMoleculesMode.none)
{
var ext = "." + imdbFile.Split('.').Last();
imdbFile = imdbFile.Replace(ext, BiblioSpecLiteSpec.DotConvertedToSmallMolecules + ext);
}
Assume.IsTrue(files.Contains(imdbFile));
// And round trip it to make sure we haven't left out any new features in minimized imdb or blib files
share.Extract(new SilentProgressMonitor());
using (var cmdline = new CommandLine())
{
Assume.IsTrue(cmdline.OpenSkyFile(share.DocumentPath)); // Handles any path shifts in database files, like our .imdb file
var document2 = cmdline.Document;
Assume.IsNotNull(document2);
Assume.IsTrue(docContainer.SetDocument(document2, docContainer.Document, true));
docContainer.AssertComplete();
document2 = docContainer.Document;
var im = document2.Settings.GetIonMobilities(new MsDataFilePath(mz5Path));
var pep = document2.Molecules.First();
foreach (TransitionGroupDocNode nodeGroup in pep.Children)
{
double windowDT;
var centerDriftTime = document2.Settings.GetIonMobility(
pep, nodeGroup, null, im, null, driftTimeMax, out windowDT);
Assume.AreEqual(3.86124, centerDriftTime.IonMobility.Mobility.Value, .0001, testModeStr);
Assume.AreEqual(0.077224865797235934, windowDT, .0001, testModeStr);
}
}
}
}
}
}
private string InitRefineDocument(RefinementSettings.ConvertToSmallMoleculesMode mode)
{
TestFilesDir testFilesDir = new TestFilesDir(TestContext, @"TestData\CommandLineRefine.zip", mode.ToString());
string docPath = testFilesDir.GetTestPath("SRM_mini_single_replicate.sky");
// if (mode != RefinementSettings.ConvertToSmallMoleculesMode.none)
// {
// var dataPaths = new[] { testFilesDir.GetTestPath("worm1.mzML") };
// doc = ConvertToSmallMolecules(null, ref docPath, dataPaths, mode);
// }
IsDocumentState(docPath, _initProt, _initList, _initPep, _initMol, _initPrec, _initTran);
return(docPath);
}
/* TODO bspratt drift time libs for small molecules
*
* [TestMethod]
* public void WatersImsMsePredictedDriftTimesChromatogramTestAsSmallMolecules()
* {
* WatersImsMseChromatogramTest(DriftFilterType.predictor, true);
* }
*
* [TestMethod]
* public void WatersImsMseLibraryDriftTimesChromatogramTestAsSmallMolecules()
* {
* WatersImsMseChromatogramTest(DriftFilterType.library, true);
* }
*
*/
private void WatersImsMseChromatogramTest(DriftFilterType mode,
RefinementSettings.ConvertToSmallMoleculesMode asSmallMolecules = RefinementSettings.ConvertToSmallMoleculesMode.none)
{
string subdir = (asSmallMolecules == RefinementSettings.ConvertToSmallMoleculesMode.none) ? null : asSmallMolecules.ToString();
var testFilesDir = new TestFilesDir(TestContext, ZIP_FILE, subdir);
TestSmallMolecules = false; // Don't need that extra magic node
bool withDriftTimePredictor = (mode == DriftFilterType.predictor); // Load the doc that has a drift time predictor?
bool withDriftTimeFilter = (mode != DriftFilterType.none); // Perform drift time filtering? (either with predictor, or with bare times in blib file)
string docPath;
SrmDocument document = InitWatersImsMseDocument(testFilesDir, withDriftTimePredictor ? "single_with_driftinfo.sky" : "single_no_driftinfo.sky", asSmallMolecules, out docPath);
AssertEx.IsDocumentState(document, (withDriftTimePredictor || (asSmallMolecules != RefinementSettings.ConvertToSmallMoleculesMode.none)) ? 1 : 0, 1, 1, 1, 8); // Drift time lib load bumps the doc version
var docContainer = new ResultsTestDocumentContainer(document, docPath);
var doc = docContainer.Document;
var docOriginal = doc;
string testModeStr = withDriftTimePredictor ? "with drift time predictor" : "without drift time info";
if (withDriftTimeFilter && !withDriftTimePredictor)
{
// Use the bare drift times in the spectral library
var librarySpec = new BiblioSpecLiteSpec("drift test",
testFilesDir.GetTestPath("mse-mobility.filtered-scaled.blib"));
doc = doc.ChangeSettings(
doc.Settings.ChangePeptideLibraries(lib => lib.ChangeLibrarySpecs(new[] { librarySpec })).
ChangePeptidePrediction(p => p.ChangeLibraryDriftTimesResolvingPower(100)).
ChangePeptidePrediction(p => p.ChangeUseLibraryDriftTimes(true))
);
testModeStr = "with drift times from spectral library";
}
var listChromatograms = new List <ChromatogramSet>();
// A small subset of the QC_HDMSE_02_UCA168_3495_082213 data set (RT 21.5-22.5) from Will Thompson
const string path = @"waters-mobility.mz5";
listChromatograms.Add(AssertResult.FindChromatogramSet(doc, new MsDataFilePath(path)) ??
new ChromatogramSet(Path.GetFileName(path).Replace('.', '_'), new[] { path }));
var docResults = doc.ChangeMeasuredResults(new MeasuredResults(listChromatograms));
Assert.IsTrue(docContainer.SetDocument(docResults, docOriginal, true));
docContainer.AssertComplete();
document = docContainer.Document;
float tolerance = (float)document.Settings.TransitionSettings.Instrument.MzMatchTolerance;
double maxHeight = 0;
var results = document.Settings.MeasuredResults;
Assert.AreEqual(1, document.MoleculePrecursorPairs.Count());
foreach (var pair in document.MoleculePrecursorPairs)
{
ChromatogramGroupInfo[] chromGroupInfo;
Assert.IsTrue(results.TryLoadChromatogram(0, pair.NodePep, pair.NodeGroup,
tolerance, true, out chromGroupInfo));
Assert.AreEqual(1, chromGroupInfo.Length, testModeStr);
var chromGroup = chromGroupInfo[0];
var expectedPeaks = ((asSmallMolecules == RefinementSettings.ConvertToSmallMoleculesMode.masses_only) ? 6 : 5);
Assert.AreEqual(withDriftTimeFilter ? 3 : expectedPeaks, chromGroup.NumPeaks, testModeStr); // This will be higher if we don't filter on DT
foreach (var tranInfo in chromGroup.TransitionPointSets)
{
maxHeight = Math.Max(maxHeight, tranInfo.MaxIntensity);
}
}
Assert.AreEqual(withDriftTimeFilter? 5226 : 20075, maxHeight, 1, testModeStr); // Without DT filtering, this will be much greater
// now drill down for specific values
int nPeptides = 0;
foreach (var nodePep in document.Molecules.Where(nodePep => nodePep.Results[0] != null))
{
// expecting just one peptide result in this small data set
if (nodePep.Results[0].Sum(chromInfo => chromInfo.PeakCountRatio > 0 ? 1 : 0) > 0)
{
Assert.AreEqual(21.94865, (double)nodePep.GetMeasuredRetentionTime(0), .0001, testModeStr);
Assert.AreEqual(1.0, (double)nodePep.GetPeakCountRatio(0), 0.0001, testModeStr);
nPeptides++;
}
}
Assert.AreEqual(1, nPeptides);
if (withDriftTimePredictor || withDriftTimeFilter)
{
// Verify that the .imdb pr .blib file goes out in the share zipfile
for (int complete = 0; complete <= 1; complete++)
{
var sharePath = testFilesDir.GetTestPath(complete == 1?"share_complete.zip":"share_minimized.zip");
var share = new SrmDocumentSharing(document, docPath, sharePath, complete == 1);
using (var longWaitDlg = new LongWaitDlg
{
// ReSharper disable once LocalizableElement
Text = "unit test WatersImsTest -- sharing document",
})
{
longWaitDlg.PerformWork(null, 1000, share.Share);
Assert.IsFalse(longWaitDlg.IsCanceled);
}
var files = share.ListEntries().ToArray();
Assert.IsTrue(files.Contains(withDriftTimePredictor ? "scaled.imdb" : "mse-mobility.filtered-scaled.blib"));
// And round trip it to make sure we haven't left out any new features in minimized imdb or blib files
using (var longWaitDlg = new LongWaitDlg
{
// ReSharper disable once LocalizableElement
Text = "unit test WatersImsTest",
})
{
longWaitDlg.PerformWork(null, 1000, share.Extract);
Assert.IsFalse(longWaitDlg.IsCanceled);
}
using (TextReader reader = new StreamReader(share.DocumentPath))
{
XmlSerializer documentSerializer = new XmlSerializer(typeof(SrmDocument));
var document2 = (SrmDocument)documentSerializer.Deserialize(reader);
Assert.IsNotNull(document2);
var im = document.Settings.GetIonMobilities(new MsDataFilePath(path));
var pep = document2.Molecules.First();
foreach (TransitionGroupDocNode nodeGroup in pep.Children)
{
double windowDT;
var centerDriftTime = document.Settings.PeptideSettings.Prediction.GetDriftTime(
pep, nodeGroup, im, out windowDT);
Assert.AreEqual(3.86124, centerDriftTime.DriftTimeMsec(false) ?? 0, .0001, testModeStr);
Assert.AreEqual(0.077224865797235934, windowDT, .0001, testModeStr);
}
}
}
}
// Release file handles
docContainer.Release();
testFilesDir.Dispose();
string cachePath = ChromatogramCache.FinalPathForName(docPath, null);
FileEx.SafeDelete(cachePath);
}
