/// <summary>
/// Find the matches for given searchStrings in sequence and returns
/// the matched strings with indices found at.
/// </summary>
/// <param name="sequence">Input sequence.</param>
/// <param name="searchPatterns">Strings to be searched.</param>
/// <returns>Matches found in sequence.</returns>
public IDictionary<string, IList<int>> FindMatch(ISequence sequence, IList<string> searchPatterns)
{
if (sequence == null)
{
throw new ArgumentNullException("sequence");
}
if (searchPatterns == null)
{
throw new ArgumentNullException("searchPatterns");
}
// Create tasks
IList<Task<KeyValuePair<string, IList<int>>>> tasks = searchPatterns.Select(
searchString => Task<KeyValuePair<string, IList<int>>>.Factory.StartNew(
t => new KeyValuePair<string, IList<int>>(searchString, FindMatch(sequence, searchString)),
TaskCreationOptions.None)).ToList();
// Wait for all the task
Task.WaitAll(tasks.ToArray());
IDictionary<string, IList<int>> results = new Dictionary<string, IList<int>>();
foreach (Task<KeyValuePair<string, IList<int>>> task in tasks)
{
results.Add(task.Result.Key, task.Result.Value);
}
return results;
}
/// <summary>
/// <see cref="IWaitStrategy.WaitFor"/>
/// </summary>
public long WaitFor(long sequence, Sequence cursor, ISequence dependentSequence, ISequenceBarrier barrier)
{
long startTime = 0;
int counter = _spinTries;
do
{
long availableSequence;
if ((availableSequence = dependentSequence.Value) >= sequence)
return availableSequence;
if (0 == --counter)
{
if (0 == startTime)
{
startTime = GetSystemTimeTicks();
}
else
{
var timeDelta = GetSystemTimeTicks() - startTime;
if (timeDelta > _yieldTimeoutTicks)
{
return _fallbackStrategy.WaitFor(sequence, cursor, dependentSequence, barrier);
}
if (timeDelta > _spinTimeoutTicks)
{
Thread.Yield();
}
}
counter = _spinTries;
}
}
while (true);
}
/// <summary>
/// <see cref="IWaitStrategy.WaitFor"/>
/// </summary>
public long WaitFor(long sequence, Sequence cursor, ISequence dependentSequence, ISequenceBarrier barrier)
{
var timeSpan = _timeout;
if (cursor.Value < sequence)
{
lock (_gate)
{
while (cursor.Value < sequence)
{
barrier.CheckAlert();
if (!Monitor.Wait(_gate, timeSpan))
{
throw TimeoutException.Instance;
}
}
}
}
long availableSequence;
while ((availableSequence = dependentSequence.Value) < sequence)
{
barrier.CheckAlert();
}
return availableSequence;
}
///<summary>
/// Initializes a new instance of the DeltaAlignment class
/// </summary>
/// <param name="referenceSequence">Reference Sequence</param>
/// <param name="querySequence">Query Sequence</param>
public DeltaAlignment(ISequence referenceSequence, ISequence querySequence)
{
this.internalDeltas = new List<long>();
ReferenceSequence = referenceSequence;
QuerySequence = querySequence;
QueryDirection = Cluster.ForwardDirection;
}
private void _UseSequenceData(ISequence sequence)
{
if (!_persistPreFilterCache) {
_intentPreFilter.ClearCache();
}
_intentPreFilter.Data = sequence.SequenceData.EffectData;
}
public SequenceStartedEventArgs(ISequence sequence, ITiming timingSource, TimeSpan startTime, TimeSpan endTime)
{
Sequence = sequence;
TimingSource = timingSource;
StartTime = startTime;
EndTime = endTime;
}
public virtual RelationalTypeMapping GetTypeMapping(ISequence sequence)
=> GetTypeMapping(
/*specifiedType:*/ null,
sequence.Name,
sequence.Type,
isKey: false,
isConcurrencyToken: false);
/// <summary>
/// <see cref="IWaitStrategy.WaitFor"/>.
/// </summary>
public long WaitFor(long sequence, Sequence cursor, ISequence dependentSequence, ISequenceBarrier barrier)
{
var milliseconds = _timeoutInMilliseconds;
long availableSequence;
if (cursor.Value < sequence)
{
lock (_lock)
{
while (cursor.Value < sequence)
{
Interlocked.Exchange(ref _signalNeeded, 1);
barrier.CheckAlert();
if (!Monitor.Wait(_lock, milliseconds))
{
throw TimeoutException.Instance;
}
}
}
}
while ((availableSequence = dependentSequence.Value) < sequence)
{
barrier.CheckAlert();
}
return availableSequence;
}
/// <summary>
/// For input sequence, constructs k-mers by sliding
/// a frame of size kmerLength along the input sequence.
/// Track positions of occurance for each kmer in sequence.
/// Constructs KmersOfSequence for sequence and associated k-mers.
/// </summary>
/// <param name="sequence">Input sequence.</param>
/// <param name="kmerLength">K-mer length.</param>
/// <returns>KmersOfSequence constructed from sequence and associated k-mers.</returns>
public static KmerPositionDictionary BuildKmerDictionary(ISequence sequence, int kmerLength)
{
if (sequence == null)
{
throw new ArgumentNullException("sequence");
}
if (kmerLength > sequence.Count)
{
throw new ArgumentException(Properties.Resource.KmerLengthIsTooLong);
}
// kmers maintains the map between k-mer strings to list of positions in sequence.
KmerPositionDictionary kmers = new KmerPositionDictionary();
// Sequence 'kmer' stores the k-mer in each window.
// Construct each k-mer using range from sequence.
for (long i = 0; i <= sequence.Count - kmerLength; ++i)
{
ISequence kmerString = sequence.GetSubSequence(i, kmerLength);
if (kmers.ContainsKey(kmerString))
{
kmers[kmerString].Add(i);
}
else
{
kmers[kmerString] = new List<long>() { i };
}
}
return kmers;
}
public Form1()
{
this.displayModel = new DisplayModel(Form1.NLeds);
this.sequence = new PowerOnStartupTest();
InitializeComponent();
}
/// <summary>
/// Pre-optimized Construct constructor.
/// </summary>
/// <param name="fragList">Fragment list.</param>
public Construct(List<Overlap> overlaps, ISequence sequence, DesignerSettings settings)
: base()
{
this.Overlaps = overlaps;
this.Sequence = sequence;
this.Settings = settings;
}
public ExportDialog(ISequence sequence)
{
InitializeComponent();
ForeColor = ThemeColorTable.ForeColor;
BackColor = ThemeColorTable.BackgroundColor;
ThemeUpdateControls.UpdateControls(this);
Icon = Resources.Icon_Vixen3;
_sequence = sequence;
_exportOps = new Export();
_exportOps.SequenceNotify += SequenceNotify;
_sequenceFileName = _sequence.FilePath;
IEnumerable<string> mediaFileNames =
(from media in _sequence.SequenceData.Media
where media.GetType().ToString().Contains("Audio")
where media.MediaFilePath.Length != 0
select media.MediaFilePath);
_audioFileName = "";
if (mediaFileNames.Count() > 0)
{
_audioFileName = mediaFileNames.First();
}
exportProgressBar.Visible = false;
currentTimeLabel.Visible = false;
_cancelled = false;
backgroundWorker1.DoWork += new DoWorkEventHandler(backgroundWorker1_DoWork);
backgroundWorker1.ProgressChanged += new ProgressChangedEventHandler(backgroundWorker1_ProgressChanged);
}
/// <summary>
/// Transcribes a DNA sequence into an RNA sequence. The length
/// of the resulting sequence will equal the length of the source
/// sequence. Gap and ambiguous characters will also be transcribed.
/// For example:
/// Sequence dna = new Sequence(Alphabets.DNA, "TACCGC");
/// Sequence rna = Transcription.Transcribe(dna);
/// rna.ToString() would produce "AUGGCG"
/// </summary>
/// <param name="dnaSource">The dna source sequence to be transcribed.</param>
/// <returns>The transcribed RNA sequence.</returns>
public static ISequence Transcribe(ISequence dnaSource)
{
if (dnaSource == null)
{
throw new ArgumentNullException("dnaSource");
}
if (dnaSource.Alphabet != Alphabets.DNA && dnaSource.Alphabet != Alphabets.AmbiguousDNA)
{
throw new InvalidOperationException(Properties.Resource.InvalidAlphabetType);
}
byte[] transcribedResult = new byte[dnaSource.Count];
long counter = 0;
foreach (byte n in dnaSource)
{
transcribedResult[counter] = GetRnaComplement(n);
counter++;
}
var alphabet = dnaSource.Alphabet == Alphabets.DNA ? Alphabets.RNA : Alphabets.AmbiguousRNA;
Sequence result = new Sequence(alphabet, transcribedResult);
result.ID = "Complement: " + dnaSource.ID;
return result;
}
public static bool RemoveSequence(ref ISequence[] sequences, ISequence sequence)
{
int numToRemove;
ISequence[] oldSequences;
ISequence[] newSequences;
do
{
oldSequences = Volatile.Read(ref sequences);
numToRemove = CountMatching(oldSequences, sequence);
if (numToRemove == 0)
break;
var oldSize = oldSequences.Length;
newSequences = new ISequence[oldSize - numToRemove];
for (int i = 0, pos = 0; i < oldSize; i++)
{
var testSequence = oldSequences[i];
if (sequence != testSequence)
{
newSequences[pos++] = testSequence;
}
}
}
while (Interlocked.CompareExchange(ref sequences, newSequences, oldSequences) != oldSequences);
return numToRemove != 0;
}
public static void AddSequences(ref ISequence[] sequences, ICursored cursor, params ISequence[] sequencesToAdd)
{
long cursorSequence;
ISequence[] updatedSequences;
ISequence[] currentSequences;
do
{
currentSequences = Volatile.Read(ref sequences);
updatedSequences = new ISequence[currentSequences.Length + sequencesToAdd.Length];
Array.Copy(currentSequences, updatedSequences, currentSequences.Length);
cursorSequence = cursor.Cursor;
var index = currentSequences.Length;
foreach (var sequence in sequencesToAdd)
{
sequence.SetValue(cursorSequence);
updatedSequences[index++] = sequence;
}
}
while (Interlocked.CompareExchange(ref sequences, updatedSequences, currentSequences) != currentSequences);
cursorSequence = cursor.Cursor;
foreach (var sequence in sequencesToAdd)
{
sequence.SetValue(cursorSequence);
}
}
/// <summary>
/// Convert fasta sequence id to a string array
/// </summary>
public static string[,] SequenceIDHeaderToRange(ISequence sequence)
{
var formattedData = new string[1,2];
formattedData[0, 0] = Properties.Resources.Sequence_ID;
formattedData[0, 1] = (sequence != null) ? sequence.ID : "";
return formattedData;
}
/// <summary>
/// Fragment constructor.
/// </summary>
/// <param name="source">Filename or URL.</param>
/// <param name="name">Fragment name.</param>
public Fragment(String source, String name, ISequence sequence, bool vector = false)
{
this.Source = source;
this.Name = name;
this.Sequence = sequence;
this.Length = sequence.Count;
this.IsVector = vector;
}
/// <summary>
/// Method to align two sequences, this sample code uses NeedlemanWunschAligner.
/// </summary>
/// <param name="referenceSequence">Reference sequence for alignment</param>
/// <param name="querySequence">Query sequence for alignment</param>
/// <returns>List of IPairwiseSequenceAlignment</returns>
public static IList<IPairwiseSequenceAlignment> AlignSequences(ISequence referenceSequence, ISequence querySequence)
{
// Initialize the Aligner
NeedlemanWunschAligner aligner = new NeedlemanWunschAligner();
// Calling align method to do the alignment.
return aligner.Align(referenceSequence, querySequence);
}
public void SaveToFile(ISequence obj, string filePath)
{
IFileWriter fileWriter = FileWriterFactory.CreateFileWriter();
IObjectContentReader contentReader = ObjectContentReaderFactory.Instance.CreateSequenceContentReader(filePath);
ObjectPersistorService.Instance.SaveToFile(obj, fileWriter, contentReader, filePath);
if (obj != null) obj.FilePath = filePath;
}
private EnumeratorSequence(IEnumerator enumerator, object first, ISequence rest, IPersistentMap metadata)
: base(metadata)
{
this.enumerator = enumerator;
this.first = first;
this.rest = rest;
this.restWasCalculated = true;
}
public void Save(ISequence sequence, string filePath)
{
if (sequence == null) throw new ArgumentNullException("sequence");
if (string.IsNullOrWhiteSpace(filePath))
throw new InvalidOperationException("File path must be a file name or path.");
FileService.Instance.SaveSequenceFile(sequence, filePath);
}
/// <summary>
/// Initializes a new instance of the Synteny class
/// </summary>
/// <param name="referenceSequence">Reference sequence</param>
/// <param name="querySequence">Query sequence</param>
public Synteny(
ISequence referenceSequence,
ISequence querySequence)
{
internalReferenceSequence = referenceSequence;
internalQuerySequence = querySequence;
internalClusters = new List<Cluster>();
}
/// <summary>
/// Get the minimum sequence from an array of <see cref="Sequence"/>s.
/// </summary>
/// <param name="sequences">sequences to compare.</param>
/// <param name="minimum">an initial default minimum. If the array is empty this value will returned.</param>
/// <returns>the minimum sequence found or lon.MaxValue if the array is empty.</returns>
public static long GetMinimumSequence(ISequence[] sequences, long minimum = long.MaxValue)
{
for (var i = 0; i < sequences.Length; i++)
{
var sequence = sequences[i].Value;
minimum = Math.Min(minimum, sequence);
}
return minimum;
}
public static float PairWiseScoreFunction(ISequence sequenceA, ISequence sequenceB, SimilarityMatrix similarityMatrix,
int gapOpenPenalty, int gapExtensionPenalty)
{
if (sequenceA.Count != sequenceB.Count)
{
throw new Exception("Unaligned sequences");
}
float result = 0;
bool isGapA = false;
bool isGapB = false;
for (int i = 0; i < sequenceA.Count; ++i)
{
if (sequenceA.Alphabet.CheckIsGap(sequenceA[i]) && sequenceB.Alphabet.CheckIsGap(sequenceB[i]))
{
continue;
}
if (sequenceA.Alphabet.CheckIsGap(sequenceA[i]) && !sequenceB.Alphabet.CheckIsGap(sequenceB[i]))
{
if (isGapB)
{
isGapB = false;
}
if (isGapA)
{
result += gapExtensionPenalty;
}
else
{
result += gapOpenPenalty;
isGapA = true;
}
continue;
}
if (!sequenceA.Alphabet.CheckIsGap(sequenceA[i]) && sequenceB.Alphabet.CheckIsGap(sequenceB[i]))
{
if (isGapA)
{
isGapA = false;
}
if (isGapB)
{
result += gapExtensionPenalty;
}
else
{
result += gapOpenPenalty;
isGapB = true;
}
continue;
}
result += similarityMatrix[sequenceA[i], sequenceB[i]];
}
return result;
}
public ITiming GetTimingSource(ISequence sequence, string sourceName)
{
IMediaModuleInstance mediaModule =
sequence.GetAllMedia().FirstOrDefault(x => x.TimingSource != null && x.MediaFilePath == sourceName);
if (mediaModule != null) {
return mediaModule.TimingSource;
}
return null;
}
public bool Equals(int i, ISequence other, int j)
{
StringSequence seq = other as StringSequence;
if (seq == null)
return false;
return content[i] == seq.content[j];
}
public bool Equals(int thisIdx, ISequence other, int otherIdx)
{
DocumentSequence seq = other as DocumentSequence;
if (seq == null)
return false;
return hashes[thisIdx] == seq.hashes[otherIdx];
}
/// <summary>
/// Converts a portion of an ISequence to a string. Intended to match the behaviour of Sequence.ConvertToString.
/// </summary>
/// <returns>A string </returns>
/// <param name="sequence">Sequence.</param>
/// <param name="startIndex">Start index.</param>
/// <param name="length">Length.</param>
private static string ConvertToString(ISequence sequence, long startIndex, long length)
{
StringBuilder sb = new StringBuilder((int)length);
for (long i = startIndex; i < startIndex + length; i++)
{
sb.Append(sequence[i]);
}
return sb.ToString();
}
/// <summary>
/// Lookup an amino acid within a sequence starting a certain offset.
/// </summary>
/// <param name="sequence">The source sequence to lookup from.</param>
/// <param name="offset">
/// The offset within the sequence from which to look at the next three
/// nucleotides. Note that this offset begins its count at zero. Thus
/// looking at a sequence described by "AUGGCG" and using a offset of 0
/// would lookup the amino acid for codon "AUG" while using a offset of 1
/// would lookup the amino acid for codon "UGG".
/// </param>
/// <returns>An amino acid from the protein alphabet</returns>
public static byte Lookup(ISequence sequence, int offset)
{
byte value;
if (TryLookup(sequence, offset, out value))
return value;
throw new InvalidOperationException(
string.Format(null, "({0},{1},{2}) not found.", (char)sequence[offset], (char)sequence[offset + 1], (char)sequence[offset + 2]));
}
public ITiming GetTimingSource(ISequence sequence, string sourceName)
{
IModuleDescriptor moduleDescriptor =
Modules.GetDescriptors<ITimingModuleInstance>().FirstOrDefault(x => x.TypeName == sourceName);
if (moduleDescriptor != null) {
return Modules.ModuleManagement.GetTiming(moduleDescriptor.TypeId);
}
return null;
}
public void SplitterTestMultiThread()
{
var context = TestExecuter.GetContext();
var splitter = new Splitter <DefaultRowQueue>(context)
{
InputProcess = new EnumerableImporter(context)
{
InputGenerator = caller => TestData.Person(context).TakeRowsAndReleaseOwnership(caller),
},
};
var processes = new ISequence[4];
for (var i = 0; i < 3; i++)
{
processes[i] = new CustomMutator(context)
{
Input = splitter,
Action = row =>
{
Thread.Sleep(new Random().Next(10));
row["ThreadIndex"] = i;
return(true);
},
};
}
var threads = new List <Thread>();
var results = new List <ISlimRow> [3];
for (var i = 0; i < 3; i++)
{
var threadIndex = i;
var thread = new Thread(() =>
{
var rows = processes[threadIndex].TakeRowsAndReleaseOwnership(null);
results[threadIndex] = new List <ISlimRow>(rows);
});
thread.Start();
threads.Add(thread);
}
for (var i = 0; i < 3; i++)
{
threads[i].Join();
}
Assert.AreEqual(7, results[0].Count + results[1].Count + results[2].Count);
foreach (var p in TestData.Person(context).TakeRowsAndReleaseOwnership(null))
{
Assert.IsTrue(
results[0].Any(m => m.GetAs <int>("id") == p.GetAs <int>("id")) ||
results[1].Any(m => m.GetAs <int>("id") == p.GetAs <int>("id")) ||
results[2].Any(m => m.GetAs <int>("id") == p.GetAs <int>("id")));
}
var exceptions = context.GetExceptions();
Assert.AreEqual(0, exceptions.Count);
}
/// <summary>
/// Calculates all measures needed for plotting a sequence within the given space constraints
/// Adds sequence items (TextBlocks) to this seuqnnce line depending on the available width
/// </summary>
/// <param name="constraint"></param>
/// <returns></returns>
protected override Size MeasureOverride(Size constraint)
{
long sequencePlotStartColumn = AlignStartIndex - ReadAlignStartIndex + BasePaddingLeft;
long sequencePlotEndColumn = sequencePlotStartColumn + sequence.Count;
if (!(sequencePlotStartColumn < DisplayEndIndex && sequencePlotEndColumn + 1 > DisplayStartIndex))
{
return(new Size(0, containerGrid.RowDefinitions[0].Height.Value + containerGrid.RowDefinitions[1].Height.Value));
}
long startCharIndex = 0, endCharIndex;
double leftMargin = 0;
sequenceItemsPanel.Children.Clear();
if (System.ComponentModel.DesignerProperties.GetIsInDesignMode(this))
{
sequence = new Sequence(Alphabets.DNA, "ACGT");
}
if (DisplayStartIndex > sequencePlotStartColumn)
{
startCharIndex = DisplayStartIndex - sequencePlotStartColumn;
endCharIndex = DisplayEndIndex - DisplayStartIndex + startCharIndex;
// Keep metadata text left aligned with the sequence start point.
metadataBlock.Margin = new Thickness((startCharIndex * CharWidth * -1), 0, 0, 0);
}
else
{
endCharIndex = DisplayEndIndex - DisplayStartIndex - (sequencePlotStartColumn - DisplayStartIndex);
leftMargin = (sequencePlotStartColumn - DisplayStartIndex) * CharWidth;
}
if (endCharIndex > sequence.Count)
{
endCharIndex = sequence.Count;
}
this.Margin = new Thickness(leftMargin, 0, 0, 0);
sequenceItemsBorder.Width = (endCharIndex - startCharIndex) * CharWidth;
// Color code sequences if complemented , reversed or reverse complemented
if (IsComplemented && IsReversed)
{
sequenceItemsBorder.BorderBrush = Brushes.Crimson;
}
else if (IsComplemented)
{
sequenceItemsBorder.BorderBrush = Brushes.LightSeaGreen;
}
else if (IsReversed)
{
sequenceItemsBorder.BorderBrush = Brushes.BurlyWood;
}
// If font size is less than 4, skip this so that the sequences will appear as a line.
// Any font size below 4 is not readable so no point in displaying it.
if (FontSize > 4)
{
TextBlock block = null;
for (long pos = startCharIndex; pos < endCharIndex; pos++)
{
block = new TextBlock();
if (pos < ReadAlignStartIndex || pos >= ReadAlignStartIndex + AlignmentLength)
{
string symbol = new string(new char[] { (char)sequence[pos] });
block.Text = symbol.ToLower();
block.Opacity = .6;
}
else
{
string symbol = new string(new char[] { (char)sequence[pos] });
block.Text = symbol.ToString();
}
block.Width = CharWidth;
block.Height = CharHeight;
block.TextAlignment = TextAlignment.Center;
block.UseLayoutRounding = true;
Brush colorBrush;
if (ColorMap != null && ColorMap.TryGetValue(char.ToUpper((char)sequence[pos]), out colorBrush))
{
block.Background = colorBrush;
}
else if (ColorMap != null && ColorMap.TryGetValue('*', out colorBrush))
{
block.Background = colorBrush;
}
sequenceItemsPanel.Children.Add(block);
}
metadataBlock.Text = sequence.ID;
}
else
{
metadataRow.Height = new GridLength(0);
sequenceRow.Height = new GridLength(sequenceItemsBorder.BorderThickness.Top + sequenceItemsBorder.BorderThickness.Bottom + (isReferenceSequence ? 1 : 0));
}
return(base.MeasureOverride(constraint));
}
public OracleSequenceValueGeneratorState([NotNull] ISequence sequence)
: base(Check.NotNull(sequence, nameof(sequence)).IncrementBy)
{
Sequence = sequence;
}
private static void WriteLocus(ISequence sequence, TextWriter txtWriter)
{
// determine molecule and sequence type
GenBankMetadata metadata = (GenBankMetadata)sequence.Metadata[Helper.GenBankMetadataKey];
GenBankLocusInfo locusInfo = null;
string molType = sequence.Alphabet.Name;
if (metadata != null)
{
locusInfo = metadata.Locus;
molType = locusInfo.MoleculeType.ToString();
}
string seqType;
if (sequence.Alphabet.Name != null)
{
if (molType == Alphabets.Protein.Name)
{
seqType = "aa";
molType = string.Empty; // protein files don't use molecule type
}
else
{
seqType = "bp";
}
}
else
{
if (sequence.Alphabet == Alphabets.Protein)
{
seqType = "aa";
molType = string.Empty; // protein files don't use molecule type
}
else
{
seqType = "bp";
if (sequence.Alphabet == Alphabets.DNA)
{
molType = Alphabets.DNA.Name;
}
else
{
molType = Alphabets.RNA.Name;
}
}
}
// retrieve metadata fields
string strandType = string.Empty;
string strandTopology = string.Empty;
string division = string.Empty;
DateTime date = DateTime.Now;
if (locusInfo != null)
{
strandType = Helper.GetStrandType(locusInfo.Strand);
strandTopology = Helper.GetStrandTopology(locusInfo.StrandTopology);
if (locusInfo.DivisionCode != SequenceDivisionCode.None)
{
division = locusInfo.DivisionCode.ToString();
}
date = locusInfo.Date;
}
txtWriter.WriteLine("{0,-12}{1,-16} {2,11} {3} {4,3}{5,-6} {6,-8} {7,3} {8}",
"LOCUS",
sequence.ID,
sequence.Count,
seqType,
strandType,
molType,
strandTopology,
division,
date.ToString("dd-MMM-yyyy", CultureInfo.InvariantCulture).ToUpperInvariant());
}
private void _UseSequenceFilters(ISequence sequence)
{
_intentPreFilter.Filters = sequence.GetAllSequenceFilters();
}
public ActionResult Index(
bool importGenes,
bool importPartial,
bool filterMinLength,
int minLength,
bool filterMaxLength,
int maxLength)
{
return(CreateTask(() =>
{
var genBankSearchResultsStream = FileHelper.GetFileStream(Request.Files[0]);
string searchResults = FileHelper.ReadSequenceFromStream(genBankSearchResultsStream);
string[] accessions;
if (filterMinLength)
{
accessions = filterMaxLength ?
NcbiHelper.GetIdsFromNcbiSearchResults(searchResults, importPartial, minLength, maxLength) :
NcbiHelper.GetIdsFromNcbiSearchResults(searchResults, importPartial, minLength);
}
else
{
accessions = filterMaxLength ?
NcbiHelper.GetIdsFromNcbiSearchResults(searchResults, importPartial, 1, maxLength) :
NcbiHelper.GetIdsFromNcbiSearchResults(searchResults, importPartial);
}
accessions = accessions.Distinct().Select(a => a.Split('.')[0]).ToArray();
var importResults = new List <MatterImportResult>(accessions.Length);
using (var db = new LibiadaWebEntities())
{
var matterRepository = new MatterRepository(db);
var dnaSequenceRepository = new GeneticSequenceRepository(db);
var(existingAccessions, accessionsToImport) = dnaSequenceRepository.SplitAccessionsIntoExistingAndNotImported(accessions);
importResults.AddRange(existingAccessions.ConvertAll(existingAccession => new MatterImportResult
{
MatterName = existingAccession,
Result = "Sequence already exists",
Status = "Exists"
}));
foreach (string accession in accessionsToImport)
{
var importResult = new MatterImportResult()
{
MatterName = accession
};
try
{
ISequence bioSequence = NcbiHelper.DownloadGenBankSequence(accession);
GenBankMetadata metadata = NcbiHelper.GetMetadata(bioSequence);
importResult.MatterName = metadata.Version.CompoundAccession;
Matter matter = matterRepository.CreateMatterFromGenBankMetadata(metadata);
importResult.SequenceType = matter.SequenceType.GetDisplayValue();
importResult.Group = matter.Group.GetDisplayValue();
importResult.MatterName = matter.Name;
importResult.AllNames = $"Common name = {metadata.Source.CommonName}, "
+ $"Species = {metadata.Source.Organism.Species}, "
+ $"Definition = {metadata.Definition}, "
+ $"Saved matter name = {importResult.MatterName}";
var sequence = new CommonSequence
{
Matter = matter,
Notation = Notation.Nucleotides,
RemoteDb = RemoteDb.GenBank,
RemoteId = metadata.Version.CompoundAccession
};
bool partial = metadata.Definition.ToLower().Contains("partial");
dnaSequenceRepository.Create(sequence, bioSequence, partial);
(importResult.Result, importResult.Status) = importGenes ?
ImportFeatures(metadata, sequence) :
("Successfully imported sequence", "Success");
}
catch (Exception exception)
{
importResult.Status = "Error";
importResult.Result = $"Error: {exception.Message}";
while (exception.InnerException != null)
{
exception = exception.InnerException;
importResult.Result += $" {exception.Message}";
}
foreach (var dbEntityEntry in db.ChangeTracker.Entries())
{
if (dbEntityEntry.Entity != null)
{
dbEntityEntry.State = EntityState.Detached;
}
}
}
finally
{
importResults.Add(importResult);
}
}
string[] names = importResults.Select(r => r.MatterName).ToArray();
// removing matters for which adding of sequence failed
Matter[] orphanMatters = db.Matter
.Include(m => m.Sequence)
.Where(m => names.Contains(m.Name) && m.Sequence.Count == 0)
.ToArray();
if (orphanMatters.Length > 0)
{
db.Matter.RemoveRange(orphanMatters);
db.SaveChanges();
}
}
var result = new Dictionary <string, object> {
{ "result", importResults }
};
return new Dictionary <string, string> {
{ "data", JsonConvert.SerializeObject(result) }
};
}));
}
/// <summary>
/// Parses all test cases related to ParseOne() method based on the
/// parameters passed and validates the same.
/// </summary>
/// <param name="nodeName">Xml Node name to be read.</param>
/// <param name="isFilePath">Is file path passed as parameter?</param>
void ValidateParseOneGeneralTestCases(string nodeName,
bool isFilePath)
{
// Gets the expected sequence from the Xml
string filePath = _utilityObj._xmlUtil.GetTextValue(nodeName,
Constants.FilePathNode);
Assert.IsTrue(File.Exists(filePath));
// Logs information to the log file
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"Gff Parser BVT : File Exists in the Path '{0}'.", filePath));
ISequence originalSeq = null;
GffParser parserObj = new GffParser();
if (isFilePath)
{
originalSeq = parserObj.ParseOne(filePath);
}
else
{
using (StreamReader reader = File.OpenText(filePath))
{
originalSeq = parserObj.ParseOne(reader);
}
}
Assert.IsNotNull(originalSeq);
Assert.IsTrue(ValidateFeatures(originalSeq, nodeName));
ApplicationLog.WriteLine(
"Gff Parser BVT : Successfully validated all the Features for a give Sequence in GFF File.");
Console.WriteLine(
"Gff Parser BVT : Successfully validated all the Features for a give Sequence in GFF File.");
string expectedSequence = _utilityObj._xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequenceNode);
Sequence seq = (Sequence)originalSeq;
Assert.IsNotNull(seq);
Assert.AreEqual(expectedSequence, seq.ToString());
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"Gff Parser BVT: The Gff sequence '{0}' validation after ParseOne() is found to be as expected.",
seq.ToString()));
// Logs to the NUnit GUI (Console.Out) window
Console.WriteLine(string.Format((IFormatProvider)null,
"Gff Parser BVT: The Gff sequence '{0}' validation after ParseOne() is found to be as expected.",
seq.ToString()));
Assert.AreEqual(expectedSequence.Length, seq.EncodedValues.Length);
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"Gff Parser BVT: The Gff Length sequence '{0}' is as expected.",
expectedSequence.Length));
string expectedAlphabet = _utilityObj._xmlUtil.GetTextValue(nodeName,
Constants.AlphabetNameNode).ToLower(CultureInfo.CurrentCulture);
Assert.IsNotNull(seq.Alphabet);
Assert.AreEqual(seq.Alphabet.Name.ToLower(CultureInfo.CurrentCulture),
expectedAlphabet);
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"Gff Parser BVT: The Sequence Alphabet is '{0}' and is as expected.",
seq.Alphabet.Name));
string expectedSequenceId = _utilityObj._xmlUtil.GetTextValue(nodeName,
Constants.SequenceIdNode);
Assert.AreEqual(expectedSequenceId, seq.DisplayID);
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"Gff Parser BVT: The Sequence ID is '{0}' and is as expected.",
seq.DisplayID));
// Logs to the NUnit GUI (Console.Out) window
Console.WriteLine(string.Format((IFormatProvider)null,
"Gff Parser BVT: The Sequence ID is '{0}' and is as expected.",
seq.DisplayID));
}
/// <summary>
/// Performs initializations and validations required
/// before carrying out sequence alignment.
/// Initializes only gap open penalty. Initialization for
/// gap extension, if required, has to be done seperately.
/// </summary>
/// <param name="similarityMatrix">Scoring matrix.</param>
/// <param name="gapPenalty">Gap open penalty (by convention, use a negative number for this.)</param>
/// <param name="aInput">First input sequence.</param>
/// <param name="bInput">Second input sequence.</param>
private void SimpleAlignPrimer(SimilarityMatrix similarityMatrix, int gapPenalty, ISequence aInput, ISequence bInput)
{
InitializeAlign(aInput);
ResetSpecificAlgorithmMemberVariables();
// Set Gap Penalty and Similarity Matrix
_gapOpenCost = gapPenalty;
// note that _gapExtensionCost is not used for simple gap penalty
_similarityMatrix = similarityMatrix;
ValidateAlignInput(aInput, bInput); // throws exception if input not valid
// Convert input strings to 0-based int arrays using similarity matrix mapping
_a = similarityMatrix.ToByteArray(aInput.ToString());
_b = similarityMatrix.ToByteArray(bInput.ToString());
}
/// <summary>
/// Pairwise alignment of two sequences using a single gap penalty. The various algorithms in derived classes (NeedlemanWunsch,
/// SmithWaterman, and PairwiseOverlap) all use this general engine for alignment with a single gap penalty.
/// </summary>
/// <param name="similarityMatrix">Scoring matrix.</param>
/// <param name="gapPenalty">Gap penalty (by convention, use a negative number for this.)</param>
/// <param name="aInput">First input sequence.</param>
/// <param name="bInput">Second input sequence.</param>
/// <returns>A list of sequence alignments.</returns>
public IList <IPairwiseSequenceAlignment> AlignSimple(SimilarityMatrix similarityMatrix, int gapPenalty, ISequence aInput, ISequence bInput)
{
// Initialize and perform validations for simple alignment
SimpleAlignPrimer(similarityMatrix, gapPenalty, aInput, bInput);
FillMatrixSimple();
////DumpF(); // Writes F-matrix to application log, used for development and testing
List <byte[]> alignedSequences;
List <int> offsets;
List <int> endOffsets;
List <int> insertions;
List <int> startOffsets;
int optScore = Traceback(out alignedSequences, out offsets, out startOffsets, out endOffsets, out insertions);
return(CollateResults(aInput, bInput, alignedSequences, offsets, optScore, startOffsets, endOffsets, insertions));
}
/// <summary>
/// Aligns two sequences using the affine gap metric, a gap open penalty and a gap extension penalty.
/// This method uses the existing gap open and extension penalties and similarity matrix.
/// Set these using GapOpenCost, GapExtensionCost and SimilarityMatrix properties before calling this method.
/// </summary>
/// <param name="sequence1">First input sequence.</param>
/// <param name="sequence2">Second input sequence.</param>
/// <returns>A list of sequence alignments.</returns>
public IList <IPairwiseSequenceAlignment> Align(ISequence sequence1, ISequence sequence2)
{
return(Align(_similarityMatrix, _gapOpenCost, _gapExtensionCost, sequence1, sequence2));
}
/// <summary>
/// This method is considered as main execute method which defines the
/// step by step algorithm. Derived class flows the defined flow by this
/// method.
/// </summary>
/// <param name="referenceSequenceList">Reference sequence.</param>
/// <param name="originalQuerySequences">List of input sequences.</param>
/// <returns>A list of sequence alignment.</returns>
private IEnumerable <IPairwiseSequenceAlignment> Alignment(IEnumerable <ISequence> referenceSequenceList, IEnumerable <ISequence> originalQuerySequences)
{
ConsensusResolver = new SimpleConsensusResolver(referenceSequenceList.ElementAt(0).Alphabet);
IEnumerable <ISequence> querySequenceList =
ForwardOnly ? originalQuerySequences
: (ReverseOnly
? ReverseComplementSequenceList(originalQuerySequences)
: AddReverseComplementsToSequenceList(originalQuerySequences));
IList <IPairwiseSequenceAlignment> results = new List <IPairwiseSequenceAlignment>();
var deltas = new List <DeltaAlignment>();
foreach (ISequence refSequence in referenceSequenceList)
{
this.nucmerAlgo = new NUCmer(refSequence);
if (GapOpenCost != DefaultGapOpenCost)
{
this.nucmerAlgo.GapOpenCost = GapOpenCost;
}
if (GapExtensionCost != DefaultGapExtensionCost)
{
this.nucmerAlgo.GapExtensionCost = GapExtensionCost;
}
if (LengthOfMUM != DefaultLengthOfMUM)
{
this.nucmerAlgo.LengthOfMUM = LengthOfMUM;
}
// Set the ClusterBuilder properties to defaults
if (FixedSeparation != ClusterBuilder.DefaultFixedSeparation)
{
this.nucmerAlgo.FixedSeparation = FixedSeparation;
}
if (MaximumSeparation != ClusterBuilder.DefaultMaximumSeparation)
{
this.nucmerAlgo.MaximumSeparation = MaximumSeparation;
}
if (MinimumScore != ClusterBuilder.DefaultMinimumScore)
{
this.nucmerAlgo.MinimumScore = MinimumScore;
}
if (SeparationFactor != ClusterBuilder.DefaultSeparationFactor)
{
this.nucmerAlgo.SeparationFactor = SeparationFactor;
}
if (BreakLength != ModifiedSmithWaterman.DefaultBreakLength)
{
this.nucmerAlgo.BreakLength = BreakLength;
}
this.nucmerAlgo.ConsensusResolver = ConsensusResolver;
if (SimilarityMatrix != null)
{
this.nucmerAlgo.SimilarityMatrix = SimilarityMatrix;
}
foreach (ISequence querySequence in querySequenceList)
{
// Check for parameters that would prevent an alignment from being returned.
if (Math.Min(querySequence.Count, refSequence.Count) < MinimumScore)
{
var msg = "Bad parameter settings for NucmerPairwiseAligner. " +
"Tried to align a reference of length " + refSequence.Count.ToString() +
" to a sequence of length " + querySequence.Count.ToString() +
" while requiring a minimum score of MinimumScore = " + MinimumScore +
". This will prevent any alignments from being returned.";
throw new ArgumentException(msg);
}
IEnumerable <DeltaAlignment> deltaAlignment = this.nucmerAlgo.GetDeltaAlignments(querySequence, !MaxMatch, querySequence.IsMarkedAsReverseComplement());
deltas.AddRange(deltaAlignment);
}
}
if (deltas.Count > 0)
{
ISequence concatReference = referenceSequenceList.ElementAt(0);
//// concat all the sequences into one sequence
if (referenceSequenceList.Count() > 1)
{
concatReference = ConcatSequence(referenceSequenceList);
}
foreach (ISequence querySequence in querySequenceList)
{
List <DeltaAlignment> qDelta = deltas.Where(d => d.QuerySequence.Equals(querySequence)).ToList();
IPairwiseSequenceAlignment sequenceAlignment = new PairwiseSequenceAlignment(concatReference, querySequence);
// Convert delta alignments to sequence alignments
IList <PairwiseAlignedSequence> alignments = ConvertDeltaToAlignment(qDelta);
if (alignments.Count > 0)
{
foreach (PairwiseAlignedSequence align in alignments)
{
// Calculate the score of alignment
align.Score = CalculateScore(
align.FirstSequence,
align.SecondSequence);
// Make Consensus
align.Consensus = MakeConsensus(
align.FirstSequence,
align.SecondSequence);
sequenceAlignment.PairwiseAlignedSequences.Add(align);
}
}
results.Add(sequenceAlignment);
}
}
return(results);
}
/// <summary>
/// Calculate the score of alignment.
/// </summary>
/// <param name="referenceSequence">Reference sequence.</param>
/// <param name="querySequence">Query sequence.</param>
/// <returns>Score of the alignment.</returns>
protected int CalculateScore(
ISequence referenceSequence,
ISequence querySequence)
{
if (referenceSequence == null)
{
throw new ArgumentNullException("referenceSequence");
}
if (querySequence == null)
{
throw new ArgumentNullException("querySequence");
}
int index;
int score = 0;
// For each pair of symbols (characters) in reference and query sequence
// 1. If the character are different and not alignment character "-",
// then find the cost form Similarity Matrix
// 2. If Gap Extension cost needs to be used
// a. Find how many gaps exists in appropriate sequence (reference / query)
// and calculate the score
// 3. Add the gap open cost
for (index = 0; index < referenceSequence.Count; index++)
{
byte referenceCharacter = referenceSequence[index];
byte queryCharacter = querySequence[index];
if (DnaAlphabet.Instance.Gap != referenceCharacter &&
DnaAlphabet.Instance.Gap != queryCharacter)
{
score += SimilarityMatrix[referenceCharacter, queryCharacter];
}
else
{
if (IsAlign)
{
int gapCount;
if (DnaAlphabet.Instance.Gap == referenceCharacter)
{
gapCount = FindExtensionLength(referenceSequence, index);
}
else
{
gapCount = FindExtensionLength(querySequence, index);
}
score += GapOpenCost + (gapCount * GapExtensionCost);
// move the index pointer to end of extension
index = index + gapCount - 1;
}
else
{
score += GapOpenCost;
}
}
}
return(score);
}
/// <summary>
/// Validates the Metadata Features of a Gff Sequence for the sequence and node name specified.
/// </summary>
/// <param name="seq">Sequence that needs to be validated.</param>
/// <param name="nodeName">Xml Node name to be read.</param>
/// <returns>True/false</returns>
bool ValidateFeatures(ISequence seq, string nodeName)
{
// Gets all the Features from the Sequence for Validation
List <MetadataListItem <List <string> > > featureList =
(List <MetadataListItem <List <string> > >)seq.Metadata[Constants.Features];
// Gets all the xml values for validation
string[] sequenceNames = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.SequenceNameNodeName);
string[] sources = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.SourceNodeName);
string[] featureNames = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.FeatureNameNodeName);
string[] startValues = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.StartNodeName);
string[] endValues = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.EndNodeName);
string[] scoreValues = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.ScoreNodeName);
string[] strandValues = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.StrandNodeName);
string[] frameValues = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.FrameNodeName);
string[] attributeValues = _utilityObj._xmlUtil.GetTextValues(nodeName,
Constants.AttributesNodeName);
int i = 0;
// Loop through each and every feature and validate the same.
foreach (MetadataListItem <List <string> > feature in featureList)
{
Dictionary <string, List <string> > itemList = feature.SubItems;
// Read specific feature Item and validate
// Validate Start
try
{
List <string> st = itemList[Constants.FeatureStart];
foreach (string sin in st)
{
if (0 != string.Compare(startValues[i], sin,
true, CultureInfo.CurrentCulture))
{
return(false);
}
}
}
catch (KeyNotFoundException) { }
// Validate Score
try
{
List <string> st = itemList[Constants.FeatureScore];
foreach (string sin in st)
{
if (0 != string.Compare(scoreValues[i],
sin, true, CultureInfo.CurrentCulture))
{
return(false);
}
}
}
catch (KeyNotFoundException) { }
// Validate Strand
try
{
List <string> st = itemList[Constants.FeatureStrand];
foreach (string sin in st)
{
if (0 != string.Compare(strandValues[i],
sin, true, CultureInfo.CurrentCulture))
{
return(false);
}
}
}
catch (KeyNotFoundException) { }
// Validate Source
try
{
List <string> st = itemList[Constants.FeatureSource];
foreach (string sin in st)
{
if (0 != string.Compare(sources[i],
sin, true, CultureInfo.CurrentCulture))
{
return(false);
}
}
}
catch (KeyNotFoundException) { }
// Validate End
try
{
List <string> st = itemList[Constants.FeatureEnd];
foreach (string sin in st)
{
if (0 != string.Compare(endValues[i],
sin, true, CultureInfo.CurrentCulture))
{
return(false);
}
}
}
catch (KeyNotFoundException) { }
// Validate Frame
try
{
List <string> st = itemList[Constants.FeatureFrame];
foreach (string sin in st)
{
if (0 != string.Compare(frameValues[i],
sin, true, CultureInfo.CurrentCulture))
{
return(false);
}
}
}
catch (KeyNotFoundException) { }
if (0 != string.Compare(feature.FreeText, attributeValues[i],
true, CultureInfo.CurrentCulture))
{
return(false);
}
if (0 != string.Compare(feature.Key, featureNames[i],
true, CultureInfo.CurrentCulture))
{
return(false);
}
if (0 != string.Compare(seq.DisplayID, sequenceNames[i],
true, CultureInfo.CurrentCulture))
{
return(false);
}
i++;
}
return(true);
}
/// <summary>
/// Returns a sequence which contains bases from the specified sequence as specified by the location.
/// If the location contains accession then the sequence from the referredSequences which matches the
/// accession of the location will be considered.
///
/// For example,
/// if location is "join(100..200, J00089.1:10..50, J00090.2:30..40)"
/// then bases from 100 to 200 will be considered from the sequence parameter and referredSequences will
/// be searched for the J00089.1 and J00090.2 accession if found then those sequences will be considered
/// for constructing the output sequence.
/// If the referred sequence is not found in the referredSequences then an exception will occur.
/// </summary>
/// <param name="location">Location instance.</param>
/// <param name="sequence">Sequence instance from which the sub sequence has to be returned.</param>
/// <param name="referredSequences">A dictionary containing Accession numbers as keys and Sequences as values, this will be used when
/// the location or sub-locations contains accession.</param>
public ISequence GetSubSequence(ILocation location, ISequence sequence, Dictionary <string, ISequence> referredSequences)
{
if (location == null)
{
throw new ArgumentNullException(Properties.Resource.ParameterNameLocation);
}
if (sequence == null)
{
throw new ArgumentNullException(Properties.Resource.ParameterNameSequence);
}
DerivedSequence basicDerSeq;
if (location.Operator == LocationOperator.Complement)
{
if (location.SubLocations.Count > 1)
{
throw new ArgumentException(Properties.Resource.ComplementWithMorethanOneSubLocs);
}
if (location.SubLocations.Count > 0)
{
basicDerSeq = new DerivedSequence(location.SubLocations[0].GetSubSequence(sequence, referredSequences), false, true);
}
else
{
basicDerSeq = new DerivedSequence(GetSubSequence(location.LocationStart, location.LocationEnd, location.Accession, location.Separator, sequence, referredSequences), false, true);
}
byte[] tempSeqData = new byte[basicDerSeq.Count];
for (int i = 0; i < basicDerSeq.Count; i++)
{
tempSeqData[i] = basicDerSeq[i];
}
return(new Sequence(sequence.Alphabet, tempSeqData));
}
if (location.Operator == LocationOperator.Order)
{
List <ISequence> subSequences = new List <ISequence>();
if (location.SubLocations.Count > 0)
{
foreach (ILocation loc in location.SubLocations)
{
subSequences.Add(loc.GetSubSequence(sequence, referredSequences));
}
}
else
{
basicDerSeq = new DerivedSequence(GetSubSequence(location.LocationStart, location.LocationEnd, location.Accession, location.Separator, sequence, referredSequences), false, false);
byte[] seqData = new byte[basicDerSeq.Count];
for (long i = 0; i < basicDerSeq.Count; i++)
{
seqData[i] = basicDerSeq[i];
}
subSequences.Add(new Sequence(sequence.Alphabet, seqData));
}
long totalSubSequenceLength = 0;
long j = 0;
foreach (ISequence seq in subSequences)
{
totalSubSequenceLength += seq.Count;
}
byte[] tempSeqData = new byte[totalSubSequenceLength];
totalSubSequenceLength = 0;
IAlphabet alphabet = null;
int m = 0;
foreach (ISequence seq in subSequences)
{
totalSubSequenceLength += seq.Count;
while (j < totalSubSequenceLength)
{
tempSeqData[j] = seq[m];
j++;
m++;
}
m = 0;
alphabet = seq.Alphabet;
}
//return Segmented sequence.
return(new Sequence(alphabet, tempSeqData));
}
if (location.Operator == LocationOperator.Join || location.Operator == LocationOperator.Bond)
{
if (location.SubLocations.Count > 0)
{
List <ISequence> subSequences = new List <ISequence>();
foreach (ILocation loc in location.SubLocations)
{
subSequences.Add(loc.GetSubSequence(sequence, referredSequences));
}
long i = 0;
long subSeqLength = 0;
foreach (ISequence subSeq in subSequences)
{
subSeqLength += subSeq.Count;
}
byte[] seqData = new byte[subSeqLength];
subSeqLength = 0;
int m = 0;
foreach (ISequence subSeq in subSequences)
{
subSeqLength += subSeq.Count;
while (i < subSeqLength)
{
seqData[i] = subSeq[m];
i++;
m++;
}
m = 0;
}
Sequence seq = new Sequence(sequence.Alphabet, seqData);
return(seq);
}
else
{
return(GetSubSequence(location.LocationStart, location.LocationEnd, location.Accession, location.Separator, sequence, referredSequences));
}
}
if (location.SubLocations.Count > 0)
{
throw new ArgumentException(Properties.Resource.NoneWithSubLocs);
}
return(GetSubSequence(location.LocationStart, location.LocationEnd, location.Accession, location.Separator, sequence, referredSequences));
}
/// <summary>
/// Returns the sequence for the specified start and end positions.
/// If the accession is null or empty then the source sequence is used to construct the output sequence,
/// otherwise appropriate sequence from the referred sequence is used to construct output sequence.
/// </summary>
/// <param name="start">Start position.</param>
/// <param name="end">End position.</param>
/// <param name="accession">Accession number.</param>
/// <param name="sepataror">Start and End separator.</param>
/// <param name="source">Source sequence.</param>
/// <param name="referredSequences">Referred Sequences.</param>
private static ISequence GetSubSequence(int start, int end, string accession, string sepataror, ISequence source, Dictionary <string, ISequence> referredSequences)
{
if (string.Compare(sepataror, "^", StringComparison.OrdinalIgnoreCase) == 0)
{
return(new Sequence(source.Alphabet, string.Empty));
}
if (string.Compare(sepataror, "..", StringComparison.OrdinalIgnoreCase) != 0 &&
string.Compare(sepataror, ".", StringComparison.OrdinalIgnoreCase) != 0 &&
!string.IsNullOrEmpty(sepataror))
{
string str = string.Format(CultureInfo.CurrentCulture, Properties.Resource.InvalidSeparator, sepataror);
throw new ArgumentException(str);
}
if (!string.IsNullOrEmpty(accession) && (referredSequences == null || !referredSequences.ContainsKey(accession)))
{
string str = string.Format(CultureInfo.CurrentCulture, Properties.Resource.AccessionSequenceNotFound, accession);
throw new ArgumentException(str);
}
if (!string.IsNullOrEmpty(accession))
{
if (source.Alphabet != referredSequences[accession].Alphabet)
{
string str = string.Format(CultureInfo.CurrentCulture, Properties.Resource.InvalidReferredAlphabet, accession);
throw new ArgumentException(str);
}
source = referredSequences[accession];
}
// as location.start is one based where as Range accepts zero based index.
start = start - 1;
int length = end - start;
if (string.IsNullOrEmpty(sepataror) || string.Compare(sepataror, ".", StringComparison.OrdinalIgnoreCase) == 0)
{
length = 1;
}
ISequence newSequence = source.GetSubSequence(start, length);
byte[] seqData = new byte[newSequence.Count];
for (long i = 0; i < newSequence.Count; i++)
{
seqData[i] = newSequence[i];
}
return(new Sequence(source.Alphabet, seqData));
}
/// <summary> /// Averages the <paramref name="values"/>. /// </summary> /// <typeparam name="TEnumerator">The type of the enumerator of the <paramref name="values"/>.</typeparam> /// <param name="values">The values to find the mean of.</param> /// <returns>The mean of the <paramref name="values"/>.</returns> public static Double GeometricMean <TEnumerator>([AllowNull] this ISequence <Single, TEnumerator> values) where TEnumerator : notnull, ICount, ICurrent <Single>, IMoveNext, IReset => Math.Pow(values.Product(), 1.0 / (values?.Count ?? 0));
/// <summary>
/// Align the Gap by executing pairwise alignment.
/// </summary>
/// <param name="referenceSequence">Reference sequence.</param>
/// <param name="querySequence">Query Sequence.</param>
/// <param name="sequenceResult1">Editable sequence containing alignment first result.</param>
/// <param name="sequenceResult2">Editable sequence containing alignment second result.</param>
/// <param name="consensusResult">Editable sequence containing consensus sequence.</param>
/// <param name="mum1">First MUM of Gap.</param>
/// <param name="mum2">Second MUM of Gap.</param>
/// <param name="insertions">Insertions made to the aligned sequences.</param>
/// <returns>Score of alignment.</returns>
private long AlignGap(
ISequence referenceSequence,
ISequence querySequence,
List <byte> sequenceResult1,
List <byte> sequenceResult2,
List <byte> consensusResult,
Match mum1,
Match mum2,
out List <long> insertions)
{
long score = 0;
ISequence sequence1 = null;
ISequence sequence2 = null;
IList <IPairwiseSequenceAlignment> sequenceAlignment = null;
byte[] mum1String;
byte[] mum2String;
insertions = new List <long>(2);
insertions.Add(0);
insertions.Add(0);
long mum1ReferenceStartIndex = 0;
long mum1QueryStartIndex = 0;
long mum1Length = 0;
long mum2ReferenceStartIndex = 0;
long mum2QueryStartIndex = 0;
long mum2Length = 0;
if (mum1.Length != 0)
{
mum1ReferenceStartIndex = mum1.ReferenceSequenceOffset;
mum1QueryStartIndex = mum1.QuerySequenceOffset;
mum1Length = mum1.Length;
}
if (mum2.Length != 0)
{
mum2ReferenceStartIndex = mum2.ReferenceSequenceOffset;
mum2QueryStartIndex = mum2.QuerySequenceOffset;
mum2Length = mum2.Length;
}
else
{
mum2ReferenceStartIndex = referenceSequence.Count;
mum2QueryStartIndex = querySequence.Count;
}
long referenceGapStartIndex = mum1ReferenceStartIndex + mum1Length;
long queryGapStartIndex = mum1QueryStartIndex + mum1Length;
if (mum2ReferenceStartIndex > referenceGapStartIndex &&
mum2QueryStartIndex > queryGapStartIndex)
{
sequence1 = referenceSequence.GetSubSequence(
referenceGapStartIndex,
mum2ReferenceStartIndex - referenceGapStartIndex);
sequence2 = querySequence.GetSubSequence(
queryGapStartIndex,
mum2QueryStartIndex - queryGapStartIndex);
sequenceAlignment = this.RunPairWise(sequence1, sequence2);
if (sequenceAlignment != null)
{
foreach (IPairwiseSequenceAlignment pairwiseAlignment in sequenceAlignment)
{
foreach (PairwiseAlignedSequence alignment in pairwiseAlignment.PairwiseAlignedSequences)
{
sequenceResult1.InsertRange(
sequenceResult1.Count,
alignment.FirstSequence);
sequenceResult2.InsertRange(
sequenceResult2.Count,
alignment.SecondSequence);
consensusResult.InsertRange(
consensusResult.Count,
alignment.Consensus);
score += alignment.Score;
if (alignment.Metadata.ContainsKey("Insertions"))
{
List <int> gapinsertions = alignment.Metadata["Insertions"] as List <int>;
if (gapinsertions != null)
{
if (gapinsertions.Count > 0)
{
insertions[0] += gapinsertions[0];
}
if (gapinsertions.Count > 1)
{
insertions[1] += gapinsertions[1];
}
}
}
}
}
}
}
else if (mum2ReferenceStartIndex > referenceGapStartIndex)
{
sequence1 = referenceSequence.GetSubSequence(
referenceGapStartIndex,
mum2ReferenceStartIndex - referenceGapStartIndex);
sequenceResult1.InsertRange(sequenceResult1.Count, sequence1);
sequenceResult2.InsertRange(sequenceResult2.Count, CreateDefaultGap(sequence1.Count));
consensusResult.InsertRange(consensusResult.Count, sequence1);
insertions[1] += sequence1.Count;
if (this.UseGapExtensionCost)
{
score = this.GapOpenCost + ((sequence1.Count - 1) * this.GapExtensionCost);
}
else
{
score = sequence1.Count * this.GapOpenCost;
}
}
else if (mum2QueryStartIndex > queryGapStartIndex)
{
sequence2 = querySequence.GetSubSequence(
queryGapStartIndex,
mum2QueryStartIndex - queryGapStartIndex);
sequenceResult1.InsertRange(sequenceResult1.Count, CreateDefaultGap(sequence2.Count));
sequenceResult2.InsertRange(sequenceResult2.Count, sequence2);
consensusResult.InsertRange(consensusResult.Count, sequence2);
insertions[0] += sequence2.Count;
if (this.UseGapExtensionCost)
{
score = this.GapOpenCost + ((sequence2.Count - 1) * this.GapExtensionCost);
}
else
{
score = sequence2.Count * this.GapOpenCost;
}
}
// Add the MUM to the result
if (0 < mum2Length)
{
mum1String = referenceSequence.GetSubSequence(
mum2ReferenceStartIndex,
mum2Length).ToArray();
sequenceResult1.InsertRange(sequenceResult1.Count, mum1String);
mum2String = querySequence.GetSubSequence(
mum2QueryStartIndex,
mum2Length).ToArray();
sequenceResult2.InsertRange(sequenceResult2.Count, mum2String);
consensusResult.InsertRange(consensusResult.Count, mum1String);
foreach (byte index in mum1String)
{
score += SimilarityMatrix[index, index];
}
}
return(score);
}
/// <summary>
/// General method to validate creation of Segmented sequence.
/// <param name="nodeName">xml node name.</param>
/// <param name="methodName">Method Name</param>
/// </summary>
void ValidateSegmentedsequenceCreation(
string nodeName, segmentedSequenceParameters methodName)
{
// Gets the alphabet from the Xml
string alphabet = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.AlphabetNameNode);
string expectedSegmentedSeq = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.ExpectedSequence);
string inputSequence1 = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.Sequence1);
string inputSequence2 = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.Sequence2);
string inputSequence3 = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.Sequence3);
string inputSequence4 = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.Sequence4);
string inputSequence5 = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.Sequence5);
string inputSequence6 = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.Sequence6);
string inputSequence7 = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.Sequence7);
string expectedSegSeqCount = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.SegmetedSeqCount);
string expectedSeguencesCount = _utilityObj._xmlUtil.GetTextValue(
nodeName, Constants.SequencesCount);
ISequence seq = null;
ISequence seq1 = null;
ISequence seq2 = null;
ISequence seq3 = null;
ISequence seq4 = null;
ISequence seq5 = null;
ISequence seq6 = null;
ISequence seq7 = null;
List <ISequence> seqList = new List <ISequence>();
SegmentedSequence segmentedSeq = null;
// Logs information to the log file
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"Segmented Sequence BVT: Sequence {0} is expected.", alphabet));
// create a Isequence.
seq1 = new Sequence(Utility.GetAlphabet(alphabet), inputSequence1);
seq2 = new Sequence(Utility.GetAlphabet(alphabet), inputSequence2);
seq3 = new Sequence(Utility.GetAlphabet(alphabet), inputSequence3);
seq4 = new Sequence(Utility.GetAlphabet(alphabet), inputSequence4);
seq5 = new Sequence(Utility.GetAlphabet(alphabet), inputSequence5);
seq6 = new Sequence(Utility.GetAlphabet(alphabet), inputSequence6);
seq7 = new Sequence(Utility.GetAlphabet(alphabet), inputSequence7);
// Add all sequences to sequence list.
seqList.Add(seq1);
seqList.Add(seq2);
seqList.Add(seq3);
seqList.Add(seq4);
seqList.Add(seq5);
seqList.Add(seq6);
seqList.Add(seq7);
switch (methodName)
{
case segmentedSequenceParameters.SegmentedsequenceList:
// create a Segmented sequence with sequence list.
segmentedSeq = new SegmentedSequence(seqList);
// validate created segmented sequence list.
Assert.AreEqual(expectedSegmentedSeq, segmentedSeq.ToString());
Assert.AreSame(seq1, segmentedSeq.Sequences[0]);
Assert.AreSame(seq2, segmentedSeq.Sequences[1]);
Assert.AreSame(seq3, segmentedSeq.Sequences[2]);
Assert.AreSame(seq4, segmentedSeq.Sequences[3]);
Assert.AreSame(seq5, segmentedSeq.Sequences[4]);
Assert.AreSame(seq6, segmentedSeq.Sequences[5]);
Assert.AreSame(seq7, segmentedSeq.Sequences[6]);
Console.WriteLine(string.Format((IFormatProvider)null,
"Segmented Sequence BVT: Segmented Sequence{0} is as expected.",
segmentedSeq.ToString()));
break;
case segmentedSequenceParameters.SegmentedSequence:
// create a Isequence.
seq = new Sequence(Utility.GetAlphabet(alphabet), inputSequence1);
// create a Segmented sequence
segmentedSeq = new SegmentedSequence(seq);
// validate expected segmented sequence.
Assert.AreEqual(seq.ToString(), segmentedSeq.ToString());
Console.WriteLine(string.Format((IFormatProvider)null,
"Segmented Sequence BVT: Segmented Sequence{0} is as expected.",
segmentedSeq.ToString()));
break;
case segmentedSequenceParameters.Clone:
// create a Segmented sequence with sequence list.
segmentedSeq = new SegmentedSequence(seqList);
// create a segmented sequences clone.
SegmentedSequence segSequenceClone = segmentedSeq.Clone();
// validate Clone sequence.
Assert.AreEqual(segSequenceClone.ToString(), segmentedSeq.ToString());
Assert.AreEqual(segSequenceClone.Count, segmentedSeq.Count);
Assert.AreEqual(segSequenceClone.Sequences.Count, 7);
Console.WriteLine(string.Format((IFormatProvider)null,
"Segmented Sequence BVT: Segmented Sequence{0} is as expected.",
segmentedSeq.ToString()));
break;
default:
break;
}
// Validate a created segmented Sequence
Assert.AreEqual(segmentedSeq.Count.ToString((IFormatProvider)null),
expectedSegSeqCount);
Assert.AreEqual(segmentedSeq.Sequences.Count.ToString((IFormatProvider)null),
expectedSeguencesCount);
Assert.AreSame(Utility.GetAlphabet(alphabet), segmentedSeq.Alphabet);
// Logs to the NUnit GUI (Console.Out) window
Console.WriteLine(string.Format((IFormatProvider)null,
"Segmented Sequence BVT: Segmented Sequence count{0} is as expected.",
segmentedSeq.Sequences.Count.ToString((IFormatProvider)null)));
Console.WriteLine(string.Format((IFormatProvider)null,
"Segmented Sequence BVT: Segmented Sequences count{0} is as expected.",
segmentedSeq.Count.ToString((IFormatProvider)null)));
// Logs to the NUnit GUI (Console.Out) window
ApplicationLog.WriteLine(
"Segmented Sequence BVT: Segmented Sequence validation is completed successfully.");
}
/// <summary>
/// Remove the specified sequence from this sequencer.
/// </summary>
/// <param name="sequence">to be removed.</param>
/// <returns>true if this sequence was found, false otherwise.</returns>
public bool RemoveGatingSequence(ISequence sequence)
{
return(SequenceGroups.RemoveSequence(ref _gatingSequences, sequence));
}
public void SetSequence(ISequence sequence)
{
_UseSequenceData(sequence);
_UseSequenceFilters(sequence);
}
/// <summary>
/// Get all the gaps in each sequence and call pairwise alignment.
/// </summary>
/// <param name="referenceSequence">Reference sequence.</param>
/// <param name="sequence">Query sequence.</param>
/// <param name="mums">List of MUMs.</param>
/// <returns>Aligned sequences.</returns>
private PairwiseAlignedSequence ProcessGaps(
ISequence referenceSequence,
ISequence sequence,
IList <Match> mums)
{
List <byte> sequenceResult1 = new List <byte>();
List <byte> sequenceResult2 = new List <byte>();
List <byte> consensusResult = new List <byte>();
PairwiseAlignedSequence alignedSequence = new PairwiseAlignedSequence();
Match mum1;
Match mum2;
// Run the alignment for gap before first MUM
List <long> insertions = new List <long>(2);
insertions.Add(0);
insertions.Add(0);
List <long> gapInsertions;
mum1 = mums.First();
alignedSequence.Score += this.AlignGap(
referenceSequence,
sequence,
sequenceResult1,
sequenceResult2,
consensusResult,
new Match()
{
Length = 0
}, // Here the first MUM does not exist
mum1,
out gapInsertions);
insertions[0] += gapInsertions[0];
insertions[1] += gapInsertions[1];
// Run the alignment for all the gaps between MUM
for (int index = 1; index < mums.Count; index++)
{
mum2 = mums[index];
alignedSequence.Score += this.AlignGap(
referenceSequence,
sequence,
sequenceResult1,
sequenceResult2,
consensusResult,
mum1,
mum2,
out gapInsertions);
insertions[0] += gapInsertions[0];
insertions[1] += gapInsertions[1];
mum1 = mum2;
}
// Run the alignment for gap after last MUM
alignedSequence.Score += this.AlignGap(
referenceSequence,
sequence,
sequenceResult1,
sequenceResult2,
consensusResult,
mum1,
new Match()
{
Length = 0
},
out gapInsertions);
insertions[0] += gapInsertions[0];
insertions[1] += gapInsertions[1];
byte[] result1 = sequenceResult1.ToArray();
IAlphabet alphabet = Alphabets.AutoDetectAlphabet(result1, 0, result1.LongLength, referenceSequence.Alphabet);
alignedSequence.FirstSequence = new Sequence(
alphabet,
result1)
{
ID = referenceSequence.ID,
// Do not shallow copy dictionary
//Metadata = referenceSequence.Metadata
};
byte[] result2 = sequenceResult2.ToArray();
alphabet = Alphabets.AutoDetectAlphabet(result2, 0, result2.LongLength, sequence.Alphabet);
alignedSequence.SecondSequence = new Sequence(
alphabet,
result2)
{
ID = sequence.ID,
// Do not shallow copy dictionary
//Metadata = sequence.Metadata
};
byte[] consensus = consensusResult.ToArray();
alphabet = Alphabets.AutoDetectAlphabet(consensus, 0, consensus.LongLength, referenceSequence.Alphabet);
alignedSequence.Consensus = new Sequence(
alphabet,
consensus);
// Offset is not required as Smith Waterman will fragmented alignment.
// Offset is the starting position of alignment of sequence1 with respect to sequence2.
if (this.PairWiseAlgorithm is NeedlemanWunschAligner)
{
alignedSequence.FirstOffset = alignedSequence.FirstSequence.IndexOfNonGap() -
referenceSequence.IndexOfNonGap();
alignedSequence.SecondOffset = alignedSequence.SecondSequence.IndexOfNonGap() -
sequence.IndexOfNonGap();
}
List <long> startOffsets = new List <long>(2);
List <long> endOffsets = new List <long>(2);
startOffsets.Add(0);
startOffsets.Add(0);
endOffsets.Add(referenceSequence.Count - 1);
endOffsets.Add(sequence.Count - 1);
alignedSequence.Metadata["StartOffsets"] = startOffsets;
alignedSequence.Metadata["EndOffsets"] = endOffsets;
alignedSequence.Metadata["Insertions"] = insertions;
// return the aligned sequence
return(alignedSequence);
}
private void GenerateSequence(ISequence sequence)
{
var methodName = nameof(RelationalModelBuilderExtensions.HasSequence);
if (sequence.ClrType != Sequence.DefaultClrType)
{
methodName += $"<{_code.Reference(sequence.ClrType)}>";
}
var parameters = _code.Literal(sequence.Name);
if (!string.IsNullOrEmpty(sequence.Schema) &&
sequence.Model.GetDefaultSchema() != sequence.Schema)
{
parameters += $", {_code.Literal(sequence.Schema)}";
}
var lines = new List <string> {
$"modelBuilder.{methodName}({parameters})"
};
if (sequence.StartValue != Sequence.DefaultStartValue)
{
lines.Add($".{nameof(SequenceBuilder.StartsAt)}({sequence.StartValue})");
}
if (sequence.IncrementBy != Sequence.DefaultIncrementBy)
{
lines.Add($".{nameof(SequenceBuilder.IncrementsBy)}({sequence.IncrementBy})");
}
if (sequence.MinValue != Sequence.DefaultMinValue)
{
lines.Add($".{nameof(SequenceBuilder.HasMin)}({sequence.MinValue})");
}
if (sequence.MaxValue != Sequence.DefaultMaxValue)
{
lines.Add($".{nameof(SequenceBuilder.HasMax)}({sequence.MaxValue})");
}
if (sequence.IsCyclic != Sequence.DefaultIsCyclic)
{
lines.Add($".{nameof(SequenceBuilder.IsCyclic)}()");
}
if (lines.Count == 2)
{
lines = new List <string> {
lines[0] + lines[1]
};
}
_sb.AppendLine();
_sb.Append(lines[0]);
using (_sb.Indent())
{
foreach (var line in lines.Skip(1))
{
_sb.AppendLine();
_sb.Append(line);
}
}
_sb.AppendLine(";");
}
protected override string GetNextValueRawSql(ISequence sequence)
{
var sqlServerSequence = (Sequence)sequence;
return($"SELECT NEXT VALUE FOR [{sqlServerSequence.Schema}].[{sequence.Name}];");
}
/// <summary>
/// Find the matches of sequence in suffix tree
/// </summary>
/// <param name="suffixTree">Suffix Tree</param>
/// <param name="searchSequence">Query searchSequence</param>
/// <param name="lengthOfMUM">Mininum length of MUM</param>
/// <returns>Matches found</returns>
public IList <MaxUniqueMatch> FindMatches(
SequenceSuffixTree suffixTree,
ISequence searchSequence,
long lengthOfMUM)
{
if (suffixTree == null)
{
throw new ArgumentNullException("suffixTree");
}
if (searchSequence == null)
{
throw new ArgumentNullException("searchSequence");
}
// Initialize
_referenceString = string.Empty;
_minimumLengthOfMUM = lengthOfMUM;
_suffixTree = suffixTree;
_searchSequence = searchSequence;
_queryString = _searchSequence.ToString();
SegmentedSequence referenceSequence = _suffixTree.Sequence as SegmentedSequence;
if (null != referenceSequence)
{
foreach (Sequence sequence in referenceSequence.Sequences)
{
_referenceString += sequence.ToString() + CONCATENATING_SYMBOL;
}
// remove the concatenating symbol form end and add terminating symbol
_referenceString = _referenceString.TrimEnd(CONCATENATING_SYMBOL);
_referenceString += TERMINATING_SYMBOL;
}
else
{
_referenceString = _suffixTree.Sequence.ToString() + TERMINATING_SYMBOL;
}
int interval = (int)(_queryString.Length - (_minimumLengthOfMUM - 1)) / Environment.ProcessorCount;
IList <Task <List <MaxUniqueMatch> > > result = new List <Task <List <MaxUniqueMatch> > >();
for (int index = 0; index < _queryString.Length - (_minimumLengthOfMUM - 1); index += interval)
{
int taskIndex = index;
result.Add(
Task.Factory.StartNew <List <MaxUniqueMatch> >(
o => FindMUMs(taskIndex, interval),
TaskCreationOptions.None));
}
List <MaxUniqueMatch> mergedList = new List <MaxUniqueMatch>();
foreach (List <MaxUniqueMatch> local in result.Select(l => l.Result))
{
// Check if there is overlap, last MUM of mergedList overlaps with first MUM of local
if (0 == mergedList.Count)
{
mergedList.AddRange(local.Select(m => m));
}
else
{
if (0 < local.Count)
{
MaxUniqueMatch previous = mergedList.Last();
MaxUniqueMatch current = local.First();
if ((current.SecondSequenceStart >= previous.SecondSequenceStart &&
current.SecondSequenceStart <= previous.SecondSequenceStart + previous.Length) &&
(current.SecondSequenceStart + current.Length >= previous.SecondSequenceStart &&
current.SecondSequenceStart + current.Length <= previous.SecondSequenceStart + previous.Length))
{
local.RemoveAt(0);
}
if (0 < local.Count)
{
mergedList.AddRange(local.Select(m => m));
}
}
}
}
// Order the mum list with query sequence order
for (int index = 0; index < mergedList.Count; index++)
{
mergedList[index].FirstSequenceMumOrder = index + 1;
mergedList[index].SecondSequenceMumOrder = index + 1;
}
return(mergedList);
}
public InvocableSequenceExceptionsOrReturns(ISetup <TMock, TReturn> wrapped, Mock mock, ISequence sequence,
IInvocationResponder <ISetup <TMock, TReturn>, ExceptionOrReturn <TReturn> > invocationResponder) : base(wrapped, mock, sequence, invocationResponder)
{
}
/// <summary> /// Abstract method for discarding reads /// </summary> abstract public bool CanDiscard(ISequence seqObj);
/// <summary>
/// Builds the Suffix Tree using Kurtz Algorithm(using Hash Table)
/// </summary>
/// <example>
/// --------------------------------------------------
/// Create the Sequence from string (let say DNA sequence "CACCAS")
/// --------------------------------------------------
/// string aOriginalStr = "CACCAS";
/// Sequence aInput = new Sequence(Alphabets.DNA, aOriginalStr);
/// --------------------------------------------------
/// Instantiate and run the suffix tree builder
/// --------------------------------------------------
/// ISuffixTreeBuilder suffixTreeBuilder = new KurtzSuffixTreeBuilder();
/// SuffixTree suffixTree = suffixTreeBuilder.BuildSuffixTree(aInput);
/// </example>
/// <param name="sequence">Input Sequence</param>
/// <returns>Suffix Tree</returns>
public SequenceSuffixTree BuildSuffixTree(ISequence sequence)
{
// Initialize
Edge.NodeCount = 1;
SegmentedSequence referenceSequence = sequence as SegmentedSequence;
if (null != referenceSequence)
{
foreach (Sequence subSequence in referenceSequence.Sequences)
{
_referenceString += subSequence.ToString() + CONCATENATING_SYMBOL;
}
// remove the concatenating symbol form end and add terminating symbol
_referenceString = _referenceString.TrimEnd(CONCATENATING_SYMBOL);
_referenceString += TERMINATING_SYMBOL;
}
else
{
_referenceString = sequence.ToString() + TERMINATING_SYMBOL;
}
_suffixTree = new SequenceSuffixTree(sequence);
// Create Tasks
Dictionary <char, IList <int> > treeTasks = new Dictionary <char, IList <int> >();
// Loop through subset of sequence string and build the suffix tree
for (int index = 0; index < _referenceString.Length; index++)
{
IList <int> startIndices = null;
if (!treeTasks.TryGetValue(_referenceString[index], out startIndices))
{
startIndices = new List <int>();
treeTasks.Add(_referenceString[index], startIndices);
}
startIndices.Add(index);
}
IList <Task <SequenceSuffixTree> > tasks = treeTasks.Values.Select(
indices => Task <SequenceSuffixTree> .Factory.StartNew(
t => AppendSuffix(indices, sequence), TaskCreationOptions.None)).ToList();
// Wait for all the task
Task.WaitAll(tasks.ToArray());
// Merge the branches of tree
Edge edgeFound = null;
foreach (Task <SequenceSuffixTree> task in tasks)
{
foreach (KeyValuePair <int, Edge> edge in task.Result.Edges)
{
if (_suffixTree.Edges.TryGetValue(edge.Key, out edgeFound))
{
Insert(edgeFound, _suffixTree);
}
else
{
_suffixTree.Edges.Add(edge.Key, edge.Value);
}
}
}
// return the suffix tree
return(_suffixTree);
}
public StringPathPartsSequence(IProperty <string> path, ISequence <ITextRange> sequence = null)
{
_path = path;
_sequence = sequence ?? new RunOnceSequence <ITextRange>(new TextRangeSequence(new CachedValue <IText>(() => new StringText(path.Value)), new CachedValue <int>(() => path.Value.Length), Path.DirectorySeparatorChar));
}
/// <summary>
/// Constructor
/// </summary>
/// <param name="sequence">Sequence to wrap</param>
public SequenceViewModel(ISequence sequence)
{
_sequence = sequence;
}
public static void TryAddPosition(Dictionary <IChromosome, List <int> > chromPositions, IChromosome chromosome,
int position, string refAllele, string altAllele, ISequence refSequence)
{
if (!chromPositions.ContainsKey(chromosome))
{
chromPositions.Add(chromosome, new List <int>(16 * 1024));
}
foreach (string allele in altAllele.OptimizedSplit(','))
{
if (allele.OptimizedStartsWith('<') && allele != "<NON_REF>" || allele.Contains('[') || altAllele.Contains(']'))
{
continue;
}
(int shiftedPos, string _, string _) =
VariantUtils.TrimAndLeftAlign(position, refAllele, allele, refSequence);
chromPositions[chromosome].Add(shiftedPos);
}
}
