/// <summary>
/// Assemble the input sequences into the largest possible contigs.
/// </summary>
/// <param name="referenceSequence">The sequence used as backbone for assembly.</param>
/// <param name="queryParser">The parser to load the sequences to assemble.</param>
/// <returns>IComparativeAssembly instance which contains list of assembled sequences.</returns>
public IEnumerable <ISequence> Assemble(IEnumerable <ISequence> referenceSequence, FastASequencePositionParser queryParser)
{
this._progressTimer = new Timer(ProgressTimerInterval);
this._progressTimer.Elapsed += this.ProgressTimerElapsed;
if (queryParser == null)
{
throw new ArgumentNullException("queryParser");
}
string readAlignmentOutputFilename = null;
string unsortedRepeatResolutionOutputFilename = null;
string repeateResolutionOutputFilename = null;
string unsortedLayoutRefinmentOutputFilename = null;
string layoutRefinmentOutputFileName = null;
try
{
// Converting to list to avoid multiple parse of the reference file if its a yield return
var refSequences = referenceSequence.ToList();
// CacheSequencesForRandomAccess will ignore the call if called more than once.
queryParser.CacheSequencesForRandomAccess();
IEnumerable <ISequence> reads = queryParser.Parse();
// Comparative Assembly Steps
// 1) Read Alignment (Calling NUCmer for aligning reads to reference sequence)
this.StatusEventStart(Properties.Resources.ReadAlignmentStarted);
IEnumerable <DeltaAlignment> alignmentBetweenReferenceAndReads = this.ReadAlignment(refSequences,
reads.Where(a => a.Count >= this.LengthOfMum));
readAlignmentOutputFilename = Path.GetTempFileName();
WriteDelta(alignmentBetweenReferenceAndReads, readAlignmentOutputFilename);
this.StatusEventEnd(Properties.Resources.ReadAlignmentEnded);
// 2) Repeat Resolution
this.StatusEventStart(Properties.Resources.RepeatResolutionStarted);
DeltaAlignmentSorter sorter;
unsortedRepeatResolutionOutputFilename = Path.GetTempFileName();
using (DeltaAlignmentCollection deltaAlignmentFromReadAlignment = new DeltaAlignmentCollection(readAlignmentOutputFilename, queryParser))
{
IEnumerable <DeltaAlignment> repeatResolvedDeltas = RepeatResolution(deltaAlignmentFromReadAlignment);
sorter = new DeltaAlignmentSorter(refSequences[0].Count);
WriteUnsortedDelta(repeatResolvedDeltas, sorter, unsortedRepeatResolutionOutputFilename);
}
this.StatusEventEnd(Properties.Resources.RepeatResolutionEnded);
this.StatusEventStart(Properties.Resources.SortingResolvedDeltasStarted);
repeateResolutionOutputFilename = Path.GetTempFileName();
WriteSortedDelta(sorter, unsortedRepeatResolutionOutputFilename, queryParser, repeateResolutionOutputFilename);
this.StatusEventEnd(Properties.Resources.SortingResolvedDeltasEnded);
// 3) Layout Refinement
this.StatusEventStart(Properties.Resources.LayoutRefinementStarted);
layoutRefinmentOutputFileName = Path.GetTempFileName();
using (DeltaAlignmentCollection unsortedDeltaCollectionForLayoutRefinment = new DeltaAlignmentCollection(repeateResolutionOutputFilename, queryParser))
{
unsortedLayoutRefinmentOutputFilename = Path.GetTempFileName();
IEnumerable <DeltaAlignment> layoutRefinedDeltas = LayoutRefinment(unsortedDeltaCollectionForLayoutRefinment);
sorter = new DeltaAlignmentSorter(refSequences[0].Count);
WriteUnsortedDelta(layoutRefinedDeltas, sorter, unsortedLayoutRefinmentOutputFilename);
WriteSortedDelta(sorter, unsortedLayoutRefinmentOutputFilename, queryParser, layoutRefinmentOutputFileName);
}
this.StatusEventEnd(Properties.Resources.LayoutRefinementEnded);
// 4) Consensus Generation
this.StatusEventStart(Properties.Resources.ConsensusGenerationStarted);
IList <ISequence> contigs;
using (var delta = new DeltaAlignmentCollection(layoutRefinmentOutputFileName, queryParser))
{
contigs = this.ConsensusGenerator(delta).ToList();
}
this.StatusEventEnd(Properties.Resources.ConsensusGenerationEnded);
if (this.ScaffoldingEnabled)
{
// 5) Scaffold Generation
this.StatusEventStart(Properties.Resources.ScaffoldGenerationStarted);
IEnumerable <ISequence> scaffolds = this.ScaffoldsGenerator(contigs, reads);
this.StatusEventEnd(Properties.Resources.ScaffoldGenerationEnded);
return(scaffolds);
}
else
{
return(contigs);
}
}
finally
{
this._progressTimer.Stop();
// Cleanup temp files.
if (!string.IsNullOrEmpty(readAlignmentOutputFilename))
{
File.Delete(readAlignmentOutputFilename);
}
if (!string.IsNullOrEmpty(unsortedRepeatResolutionOutputFilename))
{
File.Delete(unsortedRepeatResolutionOutputFilename);
}
if (!string.IsNullOrEmpty(repeateResolutionOutputFilename))
{
File.Delete(repeateResolutionOutputFilename);
}
if (!string.IsNullOrEmpty(unsortedLayoutRefinmentOutputFilename))
{
File.Delete(unsortedLayoutRefinmentOutputFilename);
}
if (!string.IsNullOrEmpty(layoutRefinmentOutputFileName))
{
File.Delete(layoutRefinmentOutputFileName);
}
}
}
/// <summary>
/// Parses the file.
/// </summary>
/// <param name="fileName">The FileName.</param>
/// <returns>List of sequence.</returns>
private static IEnumerable <ISequence> ParseWithPosition(string fileName)
{
FastASequencePositionParser parser = new FastASequencePositionParser(fileName);
return(parser.Parse());
}
/// <summary>
/// Writes delta for query sequences.
/// </summary>
/// <param name="sorter">Sorter instance.</param>
/// <param name="unsortedDeltaStream">Unsorted Delta Filename.</param>
/// <param name="queryParser">Query/read sequences parser.</param>
/// <param name="outputStream">Output file name.</param>
public static void WriteSortedDelta(DeltaAlignmentSorter sorter, Stream unsortedDeltaStream, FastASequencePositionParser queryParser, Stream outputStream)
{
if (sorter == null)
{
throw new ArgumentNullException("sorter");
}
using (DeltaAlignmentParser unsortedDeltaParser = new DeltaAlignmentParser(unsortedDeltaStream, queryParser))
{
using (StreamWriter writer = new StreamWriter(outputStream))
{
long deltaPositionInFile = 0;
foreach (long id in sorter.GetSortedIds())
{
DeltaAlignment deltaAlignment = unsortedDeltaParser.GetDeltaAlignmentAt(id);
deltaAlignment.Id = deltaPositionInFile;
string deltaString = Helper.GetString(deltaAlignment);
deltaPositionInFile += deltaString.Length;
writer.Write(deltaString);
}
writer.Flush();
}
}
}
/// <summary>
/// Assemble the input sequences into the largest possible contigs.
/// </summary>
/// <param name="referenceSequence">The sequence used as backbone for assembly.</param>
/// <param name="queryParser">The parser to load the sequences to assemble.</param>
/// <returns>IComparativeAssembly instance which contains list of assembled sequences.</returns>
public IEnumerable <ISequence> Assemble(IEnumerable <ISequence> referenceSequence, FastASequencePositionParser queryParser)
{
if (referenceSequence == null)
{
throw new ArgumentNullException("referenceSequence");
}
if (queryParser == null)
{
throw new ArgumentNullException("queryParser");
}
Stream readAlignmentOutputStream = null;
Stream unsortedRepeatResolutionOutputStream = null;
Stream repeatResolutionOutputStream = null;
Stream unsortedLayoutRefinmentOutputStream = null;
Stream layoutRefinmentOutputStream = null;
try
{
// Converting to list to avoid multiple parse of the reference file if its a yield return
var refSequences = referenceSequence.ToList();
// CacheSequencesForRandomAccess will ignore the call if called more than once.
queryParser.CacheSequencesForRandomAccess();
IEnumerable <ISequence> reads = queryParser.Parse();
// Comparative Assembly Steps
// 1) Read Alignment (Calling NUCmer for aligning reads to reference sequence)
this.StatusEventStart(Properties.Resource.ReadAlignmentStarted);
IEnumerable <DeltaAlignment> alignmentBetweenReferenceAndReads = this.ReadAlignment(refSequences,
reads.Where(a => a.Count >= this.LengthOfMum));
readAlignmentOutputStream = PlatformManager.Services.CreateTempStream();
WriteDelta(alignmentBetweenReferenceAndReads, readAlignmentOutputStream);
this.StatusEventEnd(Properties.Resource.ReadAlignmentEnded);
// 2) Repeat Resolution
this.StatusEventStart(Properties.Resource.RepeatResolutionStarted);
DeltaAlignmentSorter sorter;
unsortedRepeatResolutionOutputStream = PlatformManager.Services.CreateTempStream();
using (var deltaAlignmentFromReadAlignment = new DeltaAlignmentCollection(readAlignmentOutputStream, queryParser))
{
IEnumerable <DeltaAlignment> repeatResolvedDeltas = RepeatResolution(deltaAlignmentFromReadAlignment);
sorter = new DeltaAlignmentSorter(refSequences[0].Count);
WriteUnsortedDelta(repeatResolvedDeltas, sorter, unsortedRepeatResolutionOutputStream);
}
this.StatusEventEnd(Properties.Resource.RepeatResolutionEnded);
this.StatusEventStart(Properties.Resource.SortingResolvedDeltasStarted);
repeatResolutionOutputStream = PlatformManager.Services.CreateTempStream();
WriteSortedDelta(sorter, unsortedRepeatResolutionOutputStream, queryParser, repeatResolutionOutputStream);
this.StatusEventEnd(Properties.Resource.SortingResolvedDeltasEnded);
// 3) Layout Refinement
this.StatusEventStart(Properties.Resource.LayoutRefinementStarted);
layoutRefinmentOutputStream = PlatformManager.Services.CreateTempStream();
using (var unsortedDeltaCollectionForLayoutRefinment = new DeltaAlignmentCollection(repeatResolutionOutputStream, queryParser))
{
unsortedLayoutRefinmentOutputStream = PlatformManager.Services.CreateTempStream();
IEnumerable <DeltaAlignment> layoutRefinedDeltas = LayoutRefinment(unsortedDeltaCollectionForLayoutRefinment);
sorter = new DeltaAlignmentSorter(refSequences[0].Count);
WriteUnsortedDelta(layoutRefinedDeltas, sorter, unsortedLayoutRefinmentOutputStream);
WriteSortedDelta(sorter, unsortedLayoutRefinmentOutputStream, queryParser, layoutRefinmentOutputStream);
}
this.StatusEventEnd(Properties.Resource.LayoutRefinementEnded);
// 4) Consensus Generation
this.StatusEventStart(Properties.Resource.ConsensusGenerationStarted);
IList <ISequence> contigs;
using (var delta = new DeltaAlignmentCollection(layoutRefinmentOutputStream, queryParser))
{
contigs = this.ConsensusGenerator(delta).ToList();
}
this.StatusEventEnd(Properties.Resource.ConsensusGenerationEnded);
if (this.ScaffoldingEnabled)
{
// 5) Scaffold Generation
this.StatusEventStart(Properties.Resource.ScaffoldGenerationStarted);
IEnumerable <ISequence> scaffolds = this.ScaffoldsGenerator(contigs, reads);
this.StatusEventEnd(Properties.Resource.ScaffoldGenerationEnded);
return(scaffolds);
}
else
{
return(contigs);
}
}
finally
{
// Cleanup temp files.
if (readAlignmentOutputStream != null)
{
readAlignmentOutputStream.Dispose();
}
if (unsortedRepeatResolutionOutputStream != null)
{
unsortedRepeatResolutionOutputStream.Dispose();
}
if (repeatResolutionOutputStream != null)
{
repeatResolutionOutputStream.Dispose();
}
if (unsortedLayoutRefinmentOutputStream != null)
{
unsortedLayoutRefinmentOutputStream.Dispose();
}
if (layoutRefinmentOutputStream != null)
{
layoutRefinmentOutputStream.Dispose();
}
}
}
/// <summary>
/// It assembles the sequences.
/// </summary>
public virtual void AssembleSequences()
{
if (this.FilePath.Length != 2)
{
Output.WriteLine(OutputLevel.Error, "Error: A reference file and 1 query file are required.");
return;
}
TimeSpan timeSpan = new TimeSpan();
Stopwatch runAlgorithm = new Stopwatch();
FileInfo inputFileinfo = new FileInfo(this.FilePath[0]);
long inputFileLength = inputFileinfo.Length;
inputFileinfo = null;
if (!string.IsNullOrEmpty(this.CloneLibraryName))
{
CloneLibrary.Instance.AddLibrary(this.CloneLibraryName, (float)this.MeanLengthOfInsert, (float)this.StandardDeviationOfInsert);
}
runAlgorithm.Restart();
// Parse input files
IEnumerable <ISequence> referenceSequences = ParseFile(this.FilePath[0]);
runAlgorithm.Stop();
if (this.Verbose)
{
Output.WriteLine(OutputLevel.Verbose);
Output.WriteLine(OutputLevel.Verbose, "Processed reference file: {0}", Path.GetFullPath(this.FilePath[0]));
Output.WriteLine(OutputLevel.Verbose, " Read/Processing time : {0}", runAlgorithm.Elapsed);
Output.WriteLine(OutputLevel.Verbose, " File Size : {0}", inputFileLength);
}
inputFileinfo = new FileInfo(this.FilePath[1]);
inputFileLength = inputFileinfo.Length;
runAlgorithm.Restart();
FastASequencePositionParser queryParser;
using (var stream = File.OpenRead(this.FilePath[1]))
{
// Parse and cache the sequences.
queryParser = new FastASequencePositionParser(stream, true);
queryParser.CacheSequencesForRandomAccess();
}
// Check the input
var reads = queryParser.Parse();
if (reads.Any(s => s.Alphabet.HasAmbiguity))
{
throw new ArgumentException(Resources.AmbiguousReadsNotSupported);
}
runAlgorithm.Stop();
if (this.Verbose)
{
Output.WriteLine(OutputLevel.Verbose);
Output.WriteLine(OutputLevel.Verbose, "Processed reads file : {0}", Path.GetFullPath(this.FilePath[1]));
Output.WriteLine(OutputLevel.Verbose, " Read/Processing time: {0}", runAlgorithm.Elapsed);
Output.WriteLine(OutputLevel.Verbose, " File Size : {0}", inputFileLength);
}
runAlgorithm.Restart();
runAlgorithm.Stop();
if (this.Verbose)
{
Output.WriteLine(OutputLevel.Verbose);
Output.WriteLine(OutputLevel.Verbose, "Time taken for Validating reads: {0}", runAlgorithm.Elapsed);
Output.WriteLine(OutputLevel.Verbose);
}
runAlgorithm.Restart();
ComparativeGenomeAssembler assembler = new ComparativeGenomeAssembler();
assembler.StatusChanged += this.AssemblerStatusChanged;
assembler.ScaffoldingEnabled = this.Scaffold;
assembler.KmerLength = this.KmerLength;
assembler.LengthOfMum = this.MumLength;
IEnumerable <ISequence> assemblerResult = assembler.Assemble(referenceSequences, queryParser);
runAlgorithm.Stop();
timeSpan = timeSpan.Add(runAlgorithm.Elapsed);
runAlgorithm.Restart();
this.WriteContigs(assemblerResult);
runAlgorithm.Stop();
if (this.Verbose)
{
Output.WriteLine(OutputLevel.Verbose, "Assemble time: {0}", timeSpan);
Output.WriteLine(OutputLevel.Verbose, "Write time: {0}", runAlgorithm.Elapsed);
}
}
/// <summary>
/// Writes delta for query sequences.
/// </summary>
/// <param name="sorter">Sorter instance.</param>
/// <param name="unsortedDeltaFilename">Unsorted Delta Filename.</param>
/// <param name="queryParser">Query/Read parser</param>
/// <param name="outputfilename">Output file name.</param>
private static void WriteSortedDelta(DeltaAlignmentSorter sorter, string unsortedDeltaFilename, FastASequencePositionParser queryParser, string outputfilename)
{
using (DeltaAlignmentParser unsortedDeltaParser = new DeltaAlignmentParser(unsortedDeltaFilename, queryParser))
{
TextWriter textWriterConsoleOutSave = Console.Out;
StreamWriter streamWriterConsoleOut = null;
try
{
if (!string.IsNullOrEmpty(outputfilename))
{
Output.WriteLine(OutputLevel.Required);
streamWriterConsoleOut = new StreamWriter(outputfilename);
Console.SetOut(streamWriterConsoleOut);
}
long deltaPositionInFile = 0;
foreach (long id in sorter.GetSortedIds())
{
DeltaAlignment deltaAlignment = unsortedDeltaParser.GetDeltaAlignmentAt(id);
deltaAlignment.Id = deltaPositionInFile;
string deltaString = Helper.GetString(deltaAlignment);
deltaPositionInFile += deltaString.Length;
Console.Write(deltaString);
}
Console.Out.Flush();
}
finally
{
if (streamWriterConsoleOut != null)
{
streamWriterConsoleOut.Dispose();
}
Console.SetOut(textWriterConsoleOutSave);
}
}
}