public void TestMuscleMultipleSequenceAlignmentRunningTime()
{
string filepath = @"TestUtils\FASTA\RunningTime\BOX246.xml.afa";
// Test on DNA benchmark dataset
FastAParser parser = new FastAParser(filepath);
IList <ISequence> orgSequences = parser.Parse().ToList();
List <ISequence> sequences = MsaUtils.UnAlign(orgSequences);
//filepath = @"TestUtils\FASTA\RunningTime\12_raw.afa";
//List<ISequence> sequences = parser.Parse(filepath);
int numberOfSequences = orgSequences.Count;
Console.WriteLine("Original sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(new string(sequences[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Benchmark sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(new string(orgSequences[i].Select(a => (char)a).ToArray()));
}
PAMSAMMultipleSequenceAligner.FasterVersion = true;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = false;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
int gapOpenPenalty = -13;
int gapExtendPenalty = -5;
int kmerLength = 2;
int numberOfDegrees = 2; //Environment.ProcessorCount;
int numberOfPartitions = 16; // Environment.ProcessorCount * 2;
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.InnerProductFast;
SimilarityMatrix similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.Blosum62);
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("The number of partitions is: {0}", numberOfPartitions);
Console.WriteLine("The number of degrees is: {0}", numberOfDegrees);
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences));
Console.WriteLine("Benchmark SPS score is: {0}", MsaUtils.MultipleAlignmentScoreFunction(orgSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty));
Console.WriteLine("Aligned sequences in stage 1: {0}", msa.AlignmentScoreA);
for (int i = 0; i < msa.AlignedSequencesA.Count; ++i)
{
Console.WriteLine(new string(msa.AlignedSequencesA[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Aligned sequences in stage 2: {0}", msa.AlignmentScoreB);
for (int i = 0; i < msa.AlignedSequencesB.Count; ++i)
{
Console.WriteLine(new string(msa.AlignedSequencesB[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Aligned sequences in stage 3: {0}", msa.AlignmentScoreC);
for (int i = 0; i < msa.AlignedSequencesC.Count; ++i)
{
Console.WriteLine(new string(msa.AlignedSequencesC[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
for (int i = 0; i < msa.AlignedSequences.Count; ++i)
{
Console.WriteLine(new string(msa.AlignedSequences[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences));
((FastAParser)parser).Dispose();
}
public void TestMsaBenchMarkLargeDataset()
{
// Test on DNA benchmark dataset
ISequenceParser parser = new FastaParser();
string filepath = @"testdata\FASTA\Protein\Balibase\RV913\BOX032.xml.afa";
IList <ISequence> orgSequences = parser.Parse(filepath);
IList <ISequence> sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
String outputFilePath = @"tempBOX032.xml.afa";
StreamWriter writer = new StreamWriter(outputFilePath, true);
foreach (ISequence sequence in sequences)
{
writer.WriteLine(">" + sequence.ID);
// write sequence
BasicDerivedSequence derivedSeq = new BasicDerivedSequence(sequence, false, false, 0, 0);
for (int lineStart = 0; lineStart < sequence.Count; lineStart += 60)
{
derivedSeq.RangeStart = lineStart;
derivedSeq.RangeLength = Math.Min(60, sequence.Count - lineStart);
writer.WriteLine(derivedSeq.ToString());
}
writer.Flush();
}
writer.Close();
sequences.Clear();
sequences = parser.Parse(outputFilePath);
Console.WriteLine("Original sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(sequences[i].ToString());
}
Console.WriteLine("Benchmark sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(orgSequences[i].ToString());
}
PAMSAMMultipleSequenceAligner.FasterVersion = false;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = true;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
int gapOpenPenalty = -13;
int gapExtendPenalty = -5;
int kmerLength = 3;
int numberOfDegrees = 2; //Environment.ProcessorCount;
int numberOfPartitions = 16; // Environment.ProcessorCount * 2;
SimilarityMatrix similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.Blosum62);
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.WeightedInnerProduct;
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, MoleculeType.Protein, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("Benchmark SPS score is: {0}", MsaUtils.MultipleAlignmentScoreFunction(orgSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty));
Console.WriteLine("Aligned sequences in stage 1: {0}", msa.AlignmentScoreA);
for (int i = 0; i < msa.AlignedSequencesA.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesA[i].ToString());
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Aligned sequences in stage 2: {0}", msa.AlignmentScoreB);
for (int i = 0; i < msa.AlignedSequencesB.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesB[i].ToString());
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Aligned sequences in stage 3: {0}", msa.AlignmentScoreC);
for (int i = 0; i < msa.AlignedSequencesC.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesC[i].ToString());
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
for (int i = 0; i < msa.AlignedSequences.Count; ++i)
{
Console.WriteLine(msa.AlignedSequences[i].ToString());
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences));
if (File.Exists(outputFilePath))
{
File.Delete(outputFilePath);
}
}
public void TestMsaBenchMarkOnBralibase()
{
List <float> allQ = new List <float>();
List <float> allTC = new List <float>();
string fileDirectory = @"testData\FASTA\RNA\k10";
DirectoryInfo iD = new DirectoryInfo(fileDirectory);
PAMSAMMultipleSequenceAligner.FasterVersion = false;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = false;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
MoleculeType mt = MoleculeType.RNA;
SimilarityMatrix similarityMatrix;
int gapOpenPenalty = -20;
int gapExtendPenalty = -5;
int kmerLength = 4;
int numberOfDegrees = 2; //Environment.ProcessorCount;
int numberOfPartitions = 16; // Environment.ProcessorCount * 2;
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.WeightedInnerProductCached;
switch (mt)
{
case (MoleculeType.DNA):
similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.AmbiguousDna);
break;
case (MoleculeType.RNA):
similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.AmbiguousRna);
break;
case (MoleculeType.Protein):
similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.Blosum62);
break;
default:
throw new Exception("Invalid molecular type");
}
foreach (DirectoryInfo fi in iD.GetDirectories())
{
foreach (FileInfo fiii in fi.GetFiles())
{
String filePath = fiii.FullName;
Console.WriteLine(filePath);
ISequenceParser parser = new FastaParser();
IList <ISequence> orgSequences = parser.Parse(filePath);
List <ISequence> sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
Console.WriteLine("The number of sequences is: {0}", numberOfSequences);
Console.WriteLine("Original unaligned sequences are:");
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, mt, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
for (int i = 0; i < msa.AlignedSequences.Count; ++i)
{
//Console.WriteLine(msa.AlignedSequences[i].ToString());
}
float scoreQ = MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences);
float scoreTC = MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences);
allQ.Add(scoreQ);
allTC.Add(scoreTC);
Console.WriteLine("Alignment score Q is: {0}", scoreQ);
Console.WriteLine("Alignment score TC is: {0}", scoreTC);
if (allQ.Count % 1000 == 0)
{
Console.WriteLine(">>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>");
Console.WriteLine("average Q score is: {0}", MsaUtils.Mean(allQ.ToArray()));
Console.WriteLine("average TC score is: {0}", MsaUtils.Mean(allTC.ToArray()));
}
}
}
Console.WriteLine("number of datasets is: {0}", allQ.Count);
Console.WriteLine("average Q score is: {0}", MsaUtils.Mean(allQ.ToArray()));
Console.WriteLine("average TC score is: {0}", MsaUtils.Mean(allTC.ToArray()));
}
public void TestMsaBenchMark()
{
string fileDirectory = @"TestUtils\FASTA\Protein\Balibase\RV911\";
DirectoryInfo iD = new DirectoryInfo(fileDirectory);
PAMSAMMultipleSequenceAligner.FasterVersion = false;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = true;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
SimilarityMatrix similarityMatrix;
int gapOpenPenalty = -20;
int gapExtendPenalty = -5;
int kmerLength = 4;
int numberOfDegrees = 2; //Environment.ProcessorCount;
int numberOfPartitions = 16; // Environment.ProcessorCount * 2;
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.WeightedInnerProductCached;
similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.Blosum62);
List <float> allQ = new List <float>();
List <float> allTC = new List <float>();
foreach (FileInfo fi in iD.GetFiles())
{
String filePath = fi.FullName;
Console.WriteLine(filePath);
FastAParser parser = new FastAParser(filePath);
parser.Alphabet = AmbiguousProteinAlphabet.Instance;
IList <ISequence> orgSequences = parser.Parse().ToList();
List <ISequence> sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
Console.WriteLine("The number of sequences is: {0}", numberOfSequences);
Console.WriteLine("Original unaligned sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
//Console.WriteLine(sequences[i].ToString());
}
Console.WriteLine("Original aligned sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
//Console.WriteLine(orgSequences[i].ToString());
}
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("Aligned sequences in stage 1: {0}", msa.AlignmentScoreA);
for (int i = 0; i < msa.AlignedSequencesA.Count; ++i)
{
//Console.WriteLine(msa.AlignedSequencesA[i].ToString());
}
Console.WriteLine("Aligned sequences in stage 2: {0}", msa.AlignmentScoreB);
for (int i = 0; i < msa.AlignedSequencesB.Count; ++i)
{
//Console.WriteLine(msa.AlignedSequencesB[i].ToString());
}
Console.WriteLine("Aligned sequences in stage 3: {0}", msa.AlignmentScoreC);
for (int i = 0; i < msa.AlignedSequencesC.Count; ++i)
{
//Console.WriteLine(msa.AlignedSequencesC[i].ToString());
}
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
for (int i = 0; i < msa.AlignedSequences.Count; ++i)
{
//Console.WriteLine(msa.AlignedSequences[i].ToString());
}
float scoreQ = MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences);
float scoreTC = MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences);
allQ.Add(scoreQ);
allTC.Add(scoreTC);
Console.WriteLine("Alignment score Q is: {0}", scoreQ);
Console.WriteLine("Alignment score TC is: {0}", scoreTC);
((FastAParser)parser).Dispose();
}
Console.WriteLine("Number of datasets is: {0}", allQ.Count);
Console.WriteLine("average Q score is: {0}", MsaUtils.Mean(allQ.ToArray()));
Console.WriteLine("average TC score is: {0}", MsaUtils.Mean(allTC.ToArray()));
}
public void TestAlignmentScore()
{
ISequence templateSequence = new Sequence(Alphabets.DNA, "ATGCSWRYKMBVHDN-");
Dictionary <ISequenceItem, int> itemSet = new Dictionary <ISequenceItem, int>();
for (int i = 0; i < templateSequence.Count; ++i)
{
itemSet.Add(templateSequence[i], i);
}
Profiles.ItemSet = itemSet;
SimilarityMatrix similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.AmbiguousDna);
int gapOpenPenalty = -8;
int gapExtendPenalty = -1;
// Test PairWiseScoreFunction
ISequence seqA = new Sequence(Alphabets.DNA, "ACG");
ISequence seqB = new Sequence(Alphabets.DNA, "ACG");
float score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(15, score);
seqA = new Sequence(Alphabets.DNA, "ACG");
seqB = new Sequence(Alphabets.DNA, "ACC");
score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(6, score);
seqA = new Sequence(Alphabets.DNA, "AC-");
seqB = new Sequence(Alphabets.DNA, "ACC");
score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(2, score);
seqA = new Sequence(Alphabets.DNA, "AC--");
seqB = new Sequence(Alphabets.DNA, "ACCG");
score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(1, score);
seqA = new Sequence(Alphabets.DNA, "A---");
seqB = new Sequence(Alphabets.DNA, "A--C");
score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(-3, score);
seqA = new Sequence(Alphabets.DNA, "GGGA---AAAATCAGATT");
seqB = new Sequence(Alphabets.DNA, "GGGA--CAAAATCAG---");
score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(42, score);
seqA = new Sequence(Alphabets.DNA, "GGG---AAAAATCAGATT");
seqB = new Sequence(Alphabets.DNA, "GGGA--CAAAATCAG---");
score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(33, score);
seqA = new Sequence(Alphabets.DNA, "GGGA---AAAATCAGATT");
seqB = new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG---");
score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(40, score);
seqA = new Sequence(Alphabets.DNA, "GGGA--CAAAATCAG---");
seqB = new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG---");
score = MsaUtils.PairWiseScoreFunction(seqA, seqB, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
//Assert.AreEqual(56, score);
// Test MultipleAlignmentScoreFunction
List <ISequence> sequences = new List <ISequence>();
seqA = new Sequence(Alphabets.DNA, "GGGA---AAAATCAGATT");
seqB = new Sequence(Alphabets.DNA, "GGGA--CAAAATCAG---");
sequences.Add(seqA);
sequences.Add(seqB);
score = MsaUtils.MultipleAlignmentScoreFunction(sequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
Console.WriteLine("alignment score is: {0}", score);
for (int i = 0; i < sequences.Count; ++i)
{
Console.WriteLine(sequences[i].ToString());
}
//Assert.AreEqual(42, score);
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG---"));
score = MsaUtils.MultipleAlignmentScoreFunction(sequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
Console.WriteLine("alignment score is: {0}", score);
for (int i = 0; i < sequences.Count; ++i)
{
Console.WriteLine(sequences[i].ToString());
}
//Assert.AreEqual(46, score);
sequences[0] = new Sequence(Alphabets.DNA, "GGG---AAAAATCAGATT");
score = MsaUtils.MultipleAlignmentScoreFunction(sequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
for (int i = 0; i < sequences.Count; ++i)
{
Console.WriteLine(sequences[i].ToString());
}
Console.WriteLine("alignment score is: {0}", score);
for (int i = 0; i < sequences.Count; ++i)
{
Console.WriteLine(sequences[i].ToString());
}
//Assert.AreEqual(40, score);
// Test CalculateOffset
seqA = new Sequence(Alphabets.DNA, "ABCD");
seqB = new Sequence(Alphabets.DNA, "ABCD");
List <int> offset = MsaUtils.CalculateOffset(seqA, seqB);
Console.WriteLine("offsets are:");
for (int i = 0; i < offset.Count; ++i)
{
Console.Write("{0}\t", offset[i]);
}
seqA = new Sequence(Alphabets.DNA, "A-BCD");
seqB = new Sequence(Alphabets.DNA, "AB-CD");
offset = MsaUtils.CalculateOffset(seqA, seqB);
Console.WriteLine("\noffsets are:");
for (int i = 0; i < offset.Count; ++i)
{
Console.Write("{0}\t", offset[i]);
}
seqA = new Sequence(Alphabets.DNA, "A-BCD");
seqB = new Sequence(Alphabets.DNA, "----AB-CD");
offset = MsaUtils.CalculateOffset(seqA, seqB);
Console.WriteLine("\noffsets are:");
for (int i = 0; i < offset.Count; ++i)
{
Console.Write("{0}\t", offset[i]);
}
sequences.Clear();
sequences.Add(seqA);
sequences.Add(new Sequence(Alphabets.DNA, "ABBCG"));
List <ISequence> sequencesRef = new List <ISequence>();
sequencesRef.Add(seqA);
sequencesRef.Add(new Sequence(Alphabets.DNA, "ABBCG"));
for (int i = 0; i < sequences.Count; ++i)
{
offset = MsaUtils.CalculateOffset(sequences[i], sequencesRef[i]);
Console.WriteLine("\noffsets are:");
for (int j = 0; j < offset.Count; ++j)
{
Console.Write("{0}\t", offset[j]);
}
}
Console.WriteLine("Q score is: {0}", MsaUtils.CalculateAlignmentScoreQ(sequences, sequencesRef));
Console.WriteLine("TC score is: {0}", MsaUtils.CalculateAlignmentScoreTC(sequences, sequencesRef));
// Test on one example
sequences.Clear();
sequences.Add(new Sequence(Alphabets.DNA, "GGGA---A-AAATCAGATT"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCA-AAATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCA-AAATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGA---A-A--TC-G---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCA-A--TCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCTTA--TCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGA--CA-AAATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGA---A-AAATCAGATT"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCA-AAATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGA--CA-AAATCAG---"));
gapOpenPenalty = -4;
gapExtendPenalty = -1;
Console.WriteLine("score is: {0}", MsaUtils.MultipleAlignmentScoreFunction(sequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty));
sequences.Clear();
sequences.Add(new Sequence(Alphabets.DNA, "GGGA---AAAATCAGATT"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGA-----AATC-G---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATC--AATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCTTA-TCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGA--CAAAATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGA---AAAATCAGATT"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG---"));
sequences.Add(new Sequence(Alphabets.DNA, "GGGA--CAAAATCAG---"));
Console.WriteLine("score is: {0}", MsaUtils.MultipleAlignmentScoreFunction(sequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty));
// Test Quick Sort
float[] a = new float[5] {
0, 2, 1, 5, 4
};
int[] aIndex = new int[5] {
0, 1, 2, 3, 4
};
MsaUtils.QuickSort(a, aIndex, 0, a.Length - 1);
Console.WriteLine("quicksort");
for (int i = 0; i < a.Length; ++i)
{
Console.WriteLine(a[i]);
}
for (int i = 0; i < a.Length; ++i)
{
Console.WriteLine(aIndex[i]);
}
Console.WriteLine("quicksortM");
a = new float[5] {
0, 2, 1, 5, 4
};
int[] aIndexB = null;
MsaUtils.QuickSortM(a, out aIndexB, 0, 4);
for (int i = 0; i < a.Length; ++i)
{
Console.WriteLine(a[i]);
}
for (int i = 0; i < a.Length; ++i)
{
Console.WriteLine(aIndexB[i]);
}
Console.WriteLine("quicksort");
a = new float[5] {
0, 2, 1, 5, 4
};
int[] aIndexC = MsaUtils.CreateIndexArray(a.Length);
MsaUtils.QuickSort(a, aIndexC, 0, a.Length - 1);
for (int i = 0; i < a.Length; ++i)
{
Console.WriteLine(aIndexC[i]);
}
a = new float[5] {
1, 0, 0, 0, 0
};
aIndex = new int[5] {
0, 1, 2, 3, 4
};
MsaUtils.QuickSort(a, aIndex, 0, a.Length - 1);
for (int i = 0; i < a.Length; ++i)
{
Console.WriteLine(a[i]);
}
for (int i = 0; i < a.Length; ++i)
{
Console.WriteLine(aIndex[i]);
}
}
public void TestMsaBenchMarkLargeDataset()
{
string filepath = @"\TestUtils\BOX032Small.xml.afa";
string filePathObj = Directory.GetCurrentDirectory() + filepath;
// Test on DNA benchmark dataset
FastAParser parser = new FastAParser(filePathObj);
IList <ISequence> orgSequences = parser.Parse().ToList();
IList <ISequence> sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
String outputFilePath = @"tempBOX032.xml.afa";
using (StreamWriter writer = new StreamWriter(outputFilePath, true))
{
foreach (ISequence sequence in sequences)
{
writer.WriteLine(">" + sequence.ID);
// write sequence
for (int lineStart = 0; lineStart < sequence.Count; lineStart += 60)
{
writer.WriteLine(new String(sequence.Skip(lineStart).Take((int)Math.Min(60, sequence.Count - lineStart)).Select(a => (char)a).ToArray()));
}
writer.Flush();
}
}
sequences.Clear();
parser = new FastAParser(outputFilePath);
sequences = parser.Parse().ToList();
Console.WriteLine("Original sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(new string(sequences[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Benchmark sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(new string(orgSequences[i].Select(a => (char)a).ToArray()));
}
PAMSAMMultipleSequenceAligner.FasterVersion = false;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = true;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
int gapOpenPenalty = -13;
int gapExtendPenalty = -5;
int kmerLength = 3;
int numberOfDegrees = 2; //Environment.ProcessorCount;
int numberOfPartitions = 16; // Environment.ProcessorCount * 2;
SimilarityMatrix similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.Blosum62);
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.WeightedInnerProduct;
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("Benchmark SPS score is: {0}", MsaUtils.MultipleAlignmentScoreFunction(orgSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty));
Console.WriteLine("Aligned sequences in stage 1: {0}", msa.AlignmentScoreA);
for (int i = 0; i < msa.AlignedSequencesA.Count; ++i)
{
Console.WriteLine(new string(msa.AlignedSequencesA[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Aligned sequences in stage 2: {0}", msa.AlignmentScoreB);
for (int i = 0; i < msa.AlignedSequencesB.Count; ++i)
{
Console.WriteLine(new string(msa.AlignedSequencesB[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Aligned sequences in stage 3: {0}", msa.AlignmentScoreC);
for (int i = 0; i < msa.AlignedSequencesC.Count; ++i)
{
Console.WriteLine(new string(msa.AlignedSequencesC[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
for (int i = 0; i < msa.AlignedSequences.Count; ++i)
{
Console.WriteLine(new string(msa.AlignedSequences[i].Select(a => (char)a).ToArray()));
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences));
((FastAParser)parser).Dispose();
if (File.Exists(outputFilePath))
{
File.Delete(outputFilePath);
}
}
public void TestMuscleMultipleSequenceAlignmentRunningTime()
{
// Test on DNA benchmark dataset
ISequenceParser parser = new FastaParser();
//string filepath = @"testdata\FASTA\RunningTime\122.afa";
string filepath = @"testdata\FASTA\RunningTime\BOX246.xml.afa";
MoleculeType mt = MoleculeType.Protein;
IList <ISequence> orgSequences = parser.Parse(filepath);
List <ISequence> sequences = MsaUtils.UnAlign(orgSequences);
//filepath = @"testdata\FASTA\RunningTime\12_raw.afa";
//List<ISequence> sequences = parser.Parse(filepath);
int numberOfSequences = orgSequences.Count;
Console.WriteLine("Original sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(sequences[i].ToString());
}
Console.WriteLine("Benchmark sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(orgSequences[i].ToString());
}
PAMSAMMultipleSequenceAligner.FasterVersion = true;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = false;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
int gapOpenPenalty = -13;
int gapExtendPenalty = -5;
int kmerLength = 2;
int numberOfDegrees = 2; //Environment.ProcessorCount;
int numberOfPartitions = 16; // Environment.ProcessorCount * 2;
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.InnerProductFast;
SimilarityMatrix similarityMatrix = null;
switch (mt)
{
case (MoleculeType.DNA):
similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.AmbiguousDna);
break;
case (MoleculeType.RNA):
similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.AmbiguousRna);
break;
case (MoleculeType.Protein):
similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.Blosum62);
break;
default:
throw new Exception("Invalid molecular type");
}
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, mt, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("The number of partitions is: {0}", numberOfPartitions);
Console.WriteLine("The number of degrees is: {0}", numberOfDegrees);
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences));
Console.WriteLine("Benchmark SPS score is: {0}", MsaUtils.MultipleAlignmentScoreFunction(orgSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty));
Console.WriteLine("Aligned sequences in stage 1: {0}", msa.AlignmentScoreA);
for (int i = 0; i < msa.AlignedSequencesA.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesA[i].ToString());
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Aligned sequences in stage 2: {0}", msa.AlignmentScoreB);
for (int i = 0; i < msa.AlignedSequencesB.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesB[i].ToString());
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Aligned sequences in stage 3: {0}", msa.AlignmentScoreC);
for (int i = 0; i < msa.AlignedSequencesC.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesC[i].ToString());
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
for (int i = 0; i < msa.AlignedSequences.Count; ++i)
{
Console.WriteLine(msa.AlignedSequences[i].ToString());
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences));
}
public void TestMsaBenchMarkOnBralibase()
{
var allQ = new List <float>();
var allTC = new List <float>();
string fileDirectory = @"TestUtils\Fasta\RNA\k10".TestDir();
DirectoryInfo iD = new DirectoryInfo(fileDirectory);
PAMSAMMultipleSequenceAligner.FasterVersion = false;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = false;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
var similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.AmbiguousRna);;
int gapOpenPenalty = -20;
int gapExtendPenalty = -5;
int kmerLength = 4;
int numberOfDegrees = 2;
int numberOfPartitions = 16;
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.WeightedInnerProductCached;
foreach (DirectoryInfo fi in iD.GetDirectories())
{
foreach (FileInfo fiii in fi.GetFiles())
{
String filePath = fiii.FullName;
Console.WriteLine($"Loading: {filePath}");
var orgSequences = new FastAParser()
{
Alphabet = AmbiguousRnaAlphabet.Instance
}.Parse(filePath).ToList();
var sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
Console.WriteLine("The number of sequences is: {0}", numberOfSequences);
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
float scoreQ = MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences);
float scoreTC = MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences);
Console.WriteLine("Alignment score Q is: {0}", scoreQ);
Console.WriteLine("Alignment score TC is: {0}", scoreTC);
allQ.Add(scoreQ);
allTC.Add(scoreTC);
if (allQ.Count % 1000 == 0)
{
Console.WriteLine(">>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>");
Console.WriteLine("average Q score is: {0}", MsaUtils.Mean(allQ.ToArray()));
Console.WriteLine("average TC score is: {0}", MsaUtils.Mean(allTC.ToArray()));
}
}
}
Console.WriteLine("number of datasets is: {0}", allQ.Count);
Console.WriteLine("average Q score is: {0}", MsaUtils.Mean(allQ.ToArray()));
Console.WriteLine("average TC score is: {0}", MsaUtils.Mean(allTC.ToArray()));
}
public void TestMsaBenchMarkOnSABmark()
{
List <float> allQ = new List <float>();
List <float> allTC = new List <float>();
string fileDirectory = @"TestUtils\Fasta\Protein\SABmark".TestDir();
DirectoryInfo iD = new DirectoryInfo(fileDirectory);
PAMSAMMultipleSequenceAligner.FasterVersion = false;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = true;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
SimilarityMatrix similarityMatrix;
int gapOpenPenalty = -13;
int gapExtendPenalty = -5;
int kmerLength = 3;
int numberOfDegrees = 2; //Environment.ProcessorCount;
int numberOfPartitions = 16; // Environment.ProcessorCount * 2;
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.WeightedInnerProduct;
similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.Blosum62);
foreach (DirectoryInfo fi in iD.GetDirectories())
{
foreach (DirectoryInfo fii in fi.GetDirectories())
{
foreach (FileInfo fiii in fii.GetFiles())
{
String filePath = fiii.FullName;
Console.WriteLine(filePath);
FastAParser parser = new FastAParser();
IList <ISequence> orgSequences = parser.Parse(filePath).ToList();
List <ISequence> sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
Console.WriteLine("The number of sequences is: {0}", numberOfSequences);
Console.WriteLine("Original unaligned sequences are:");
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
for (int i = 0; i < msa.AlignedSequences.Count; ++i)
{
//Console.WriteLine(msa.AlignedSequences[i].ToString());
}
float scoreQ = MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences);
float scoreTC = MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences);
allQ.Add(scoreQ);
allTC.Add(scoreTC);
Console.WriteLine("Alignment score Q is: {0}", scoreQ);
Console.WriteLine("Alignment score TC is: {0}", scoreTC);
if (allQ.Count % 1000 == 0)
{
Console.WriteLine(">>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>");
Console.WriteLine("average Q score is: {0}", MsaUtils.Mean(allQ.ToArray()));
Console.WriteLine("average TC score is: {0}", MsaUtils.Mean(allTC.ToArray()));
}
}
}
}
Console.WriteLine("average Q score is: {0}", MsaUtils.Mean(allQ.ToArray()));
Console.WriteLine("average TC score is: {0}", MsaUtils.Mean(allTC.ToArray()));
}
public void TestMsaBenchMarkLargeDataset()
{
// Test on DNA benchmark dataset
string filePathObj = @"TestUtils\BOX032Small.xml.afa".TestDir();
var orgSequences = new FastAParser().Parse(filePathObj).ToList();
var sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
Assert.AreEqual(numberOfSequences, sequences.Count);
string outputFilePath = Path.GetTempFileName();
try
{
using (StreamWriter writer = new StreamWriter(outputFilePath, true))
{
foreach (ISequence sequence in sequences)
{
// write sequence
writer.WriteLine(">" + sequence.ID);
for (int lineStart = 0; lineStart < sequence.Count; lineStart += 60)
{
writer.WriteLine(new String(sequence.Skip(lineStart).Take((int)Math.Min(60, sequence.Count - lineStart)).Select(a => (char)a).ToArray()));
}
writer.Flush();
}
}
sequences = new FastAParser().Parse(outputFilePath).ToList();
}
finally
{
File.Delete(outputFilePath);
}
Console.WriteLine("Original sequences are:");
sequences.ForEach(Console.WriteLine);
Console.WriteLine("Benchmark sequences are:");
orgSequences.ForEach(Console.WriteLine);
// Begin alignment
PAMSAMMultipleSequenceAligner.FasterVersion = false;
PAMSAMMultipleSequenceAligner.UseWeights = false;
PAMSAMMultipleSequenceAligner.UseStageB = true;
PAMSAMMultipleSequenceAligner.NumberOfCores = 2;
int gapOpenPenalty = -13;
int gapExtendPenalty = -5;
int kmerLength = 3;
int numberOfDegrees = 2;
int numberOfPartitions = 16;
SimilarityMatrix similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.Blosum62);
DistanceFunctionTypes distanceFunctionName = DistanceFunctionTypes.EuclideanDistance;
UpdateDistanceMethodsTypes hierarchicalClusteringMethodName = UpdateDistanceMethodsTypes.Average;
ProfileAlignerNames profileAlignerName = ProfileAlignerNames.NeedlemanWunschProfileAligner;
ProfileScoreFunctionNames profileProfileFunctionName = ProfileScoreFunctionNames.WeightedInnerProduct;
PAMSAMMultipleSequenceAligner msa = new PAMSAMMultipleSequenceAligner
(sequences, kmerLength, distanceFunctionName, hierarchicalClusteringMethodName,
profileAlignerName, profileProfileFunctionName, similarityMatrix, gapOpenPenalty, gapExtendPenalty,
numberOfPartitions, numberOfDegrees);
Console.WriteLine("Benchmark SPS score is: {0}", MsaUtils.MultipleAlignmentScoreFunction(orgSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty));
Console.WriteLine("Aligned sequences in stage 1: {0}", msa.AlignmentScoreA);
for (int i = 0; i < msa.AlignedSequencesA.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesA[i]);
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesA, orgSequences));
Console.WriteLine("Aligned sequences in stage 2: {0}", msa.AlignmentScoreB);
for (int i = 0; i < msa.AlignedSequencesB.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesB[i]);
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesB, orgSequences));
Console.WriteLine("Aligned sequences in stage 3: {0}", msa.AlignmentScoreC);
for (int i = 0; i < msa.AlignedSequencesC.Count; ++i)
{
Console.WriteLine(msa.AlignedSequencesC[i]);
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequencesC, orgSequences));
Console.WriteLine("Aligned sequences final: {0}", msa.AlignmentScore);
for (int i = 0; i < msa.AlignedSequences.Count; ++i)
{
Console.WriteLine(msa.AlignedSequences[i]);
}
Console.WriteLine("Alignment score Q is: {0}", MsaUtils.CalculateAlignmentScoreQ(msa.AlignedSequences, orgSequences));
Console.WriteLine("Alignment score TC is: {0}", MsaUtils.CalculateAlignmentScoreTC(msa.AlignedSequences, orgSequences));
}
