public void TestNeedlemanWunschProfileAligner()
{
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;
IProfileAligner profileAligner = new NeedlemanWunschProfileAligner();
SimilarityMatrix similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrices.AmbiguousDna);
int gapOpenPenalty = -8;
int gapExtendPenalty = -1;
profileAligner.SimilarityMatrix = similarityMatrix;
profileAligner.GapOpenCost = gapOpenPenalty;
profileAligner.GapExtensionCost = gapExtendPenalty;
ISequence seqA = new Sequence(Alphabets.DNA, "GGGAAAAATCAGATT");
ISequence seqB = new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG");
List <ISequence> sequences = new List <ISequence>();
sequences.Add(seqA);
sequences.Add(seqB);
IProfileAlignment profileAlignmentA = ProfileAlignment.GenerateProfileAlignment(sequences[0]);
IProfileAlignment profileAlignmentB = ProfileAlignment.GenerateProfileAlignment(sequences[1]);
profileAligner.Align(profileAlignmentA, profileAlignmentB);
List <int> eStringSubtree = profileAligner.GenerateEString(profileAligner.AlignedA);
List <int> eStringSubtreeB = profileAligner.GenerateEString(profileAligner.AlignedB);
List <ISequence> alignedSequences = new List <ISequence>();
ISequence seq = profileAligner.GenerateSequenceFromEString(eStringSubtree, sequences[0]);
alignedSequences.Add(seq);
seq = profileAligner.GenerateSequenceFromEString(eStringSubtreeB, sequences[1]);
alignedSequences.Add(seq);
float profileScore = MsaUtils.MultipleAlignmentScoreFunction(alignedSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
ISequence expectedSeqA = new Sequence(Alphabets.DNA, "GGGAA---AAATCAGATT");
ISequence expectedSeqB = new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG---");
Assert.AreEqual(expectedSeqA.ToString(), alignedSequences[0].ToString());
Assert.AreEqual(expectedSeqB.ToString(), alignedSequences[1].ToString());
Assert.AreEqual(40, profileScore);
}
public void TestProfile()
{
ISequence templateSequence = new Sequence(Alphabets.AmbiguousDNA, "ATGCSWRYKMBVHDN-");
Dictionary <byte, int> itemSet = new Dictionary <byte, int>();
for (int i = 0; i < templateSequence.Count; ++i)
{
itemSet.Add(templateSequence[i], i);
if (char.IsLetter((char)templateSequence[i]))
{
itemSet.Add((byte)char.ToLower((char)templateSequence[i]), i);
}
}
Profiles.ItemSet = itemSet;
ISequence seqA = new Sequence(Alphabets.DNA, "GGGAAAAATCAGATT");
ISequence seqB = new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG");
List <ISequence> sequences = new List <ISequence>();
sequences.Add(seqA);
sequences.Add(seqB);
// Test GenerateProfiles
IProfiles profileA = Profiles.GenerateProfiles(sequences[0]);
Assert.AreEqual(16, profileA.ColumnSize);
Assert.AreEqual(sequences[0].Count, profileA.RowSize);
// Test ProfileMatrix
Assert.AreEqual(1, profileA.ProfilesMatrix[0][2]);
Assert.AreEqual(0, profileA.ProfilesMatrix[0][3]);
// Test ProfileAlignment
IProfileAlignment profileAlignmentA = ProfileAlignment.GenerateProfileAlignment(sequences[0]);
Assert.AreEqual(1, profileAlignmentA.ProfilesMatrix[0][2]);
Assert.AreEqual(0, profileAlignmentA.ProfilesMatrix[0][3]);
Assert.AreEqual(1, profileAlignmentA.NumberOfSequences);
IProfileAlignment profileAlignmentB = ProfileAlignment.GenerateProfileAlignment(sequences);
Assert.AreEqual(1, profileAlignmentB.ProfilesMatrix[0][2]);
Assert.AreEqual(0, profileAlignmentB.ProfilesMatrix[0][3]);
Assert.AreEqual(2, profileAlignmentB.NumberOfSequences);
Assert.AreEqual(0.5, profileAlignmentB.ProfilesMatrix[5][0]);
Assert.AreEqual(0.5, profileAlignmentB.ProfilesMatrix[5][1]);
Assert.AreEqual(0, profileAlignmentB.ProfilesMatrix[5][2]);
}
public void TestNeedlemanWunschProfileAligner()
{
Console.WriteLine("Number of logical processors: {0}", Environment.ProcessorCount);
ISequence templateSequence = new Sequence(Alphabets.AmbiguousDNA, "ATGCSWRYKMBVHDN-");
Dictionary <byte, int> itemSet = new Dictionary <byte, int>();
for (int i = 0; i < templateSequence.Count; ++i)
{
itemSet.Add(templateSequence[i], i);
if (char.IsLetter((char)templateSequence[i]))
{
itemSet.Add((byte)char.ToLower((char)templateSequence[i]), i);
}
}
Profiles.ItemSet = itemSet;
SimilarityMatrix similarityMatrix = new SimilarityMatrix(SimilarityMatrix.StandardSimilarityMatrix.AmbiguousDna);
int gapOpenPenalty = -3;
int gapExtendPenalty = -1;
IProfileAligner profileAligner = new NeedlemanWunschProfileAlignerSerial(similarityMatrix, ProfileScoreFunctionNames.WeightedInnerProduct,
gapOpenPenalty, gapExtendPenalty, Environment.ProcessorCount);
ISequence seqA = new Sequence(Alphabets.DNA, "GGGAAAAATCAGATT");
ISequence seqB = new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG");
List <ISequence> sequences = new List <ISequence>();
sequences.Add(seqA);
sequences.Add(seqB);
IProfileAlignment profileAlignmentA = ProfileAlignment.GenerateProfileAlignment(sequences[0]);
IProfileAlignment profileAlignmentB = ProfileAlignment.GenerateProfileAlignment(sequences[1]);
profileAligner.Align(profileAlignmentA, profileAlignmentB);
List <int> eStringSubtree = profileAligner.GenerateEString(profileAligner.AlignedA);
List <int> eStringSubtreeB = profileAligner.GenerateEString(profileAligner.AlignedB);
List <ISequence> alignedSequences = new List <ISequence>();
ISequence seq = profileAligner.GenerateSequenceFromEString(eStringSubtree, sequences[0]);
alignedSequences.Add(seq);
seq = profileAligner.GenerateSequenceFromEString(eStringSubtreeB, sequences[1]);
alignedSequences.Add(seq);
float profileScore = MsaUtils.MultipleAlignmentScoreFunction(alignedSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
Console.WriteLine("alignment score is: {0}", profileScore);
Console.WriteLine("the aligned sequences are:");
for (int i = 0; i < alignedSequences.Count; ++i)
{
Console.WriteLine(new string(alignedSequences[i].Select(a => (char)a).ToArray()));
}
// Test on case 3: 36 sequences
string filepath = @"\TestUtils\RV11_BBS_allSmall.afa";
string filePathObj = Directory.GetCurrentDirectory() + filepath;
FastAParser parser = new FastAParser(filePathObj);
IList <ISequence> orgSequences = parser.Parse().ToList();
sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
Console.WriteLine("Original unaligned sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(">");
Console.WriteLine(new string(sequences[i].Select(a => (char)a).ToArray()));
}
for (int i = 1; i < numberOfSequences - 1; ++i)
{
for (int j = i + 1; j < numberOfSequences; ++j)
{
profileAlignmentA = ProfileAlignment.GenerateProfileAlignment(sequences[i]);
profileAlignmentB = ProfileAlignment.GenerateProfileAlignment(sequences[j]);
profileAligner = new NeedlemanWunschProfileAlignerSerial(similarityMatrix, ProfileScoreFunctionNames.WeightedInnerProduct,
gapOpenPenalty, gapExtendPenalty, Environment.ProcessorCount);
profileAligner.Align(profileAlignmentA, profileAlignmentB);
eStringSubtree = profileAligner.GenerateEString(profileAligner.AlignedA);
eStringSubtreeB = profileAligner.GenerateEString(profileAligner.AlignedB);
Console.WriteLine("Sequences lengths are: {0}-{1}", sequences[i].Count, sequences[j].Count);
Console.WriteLine("estring 1:");
for (int k = 0; k < eStringSubtree.Count; ++k)
{
Console.Write("{0}\t", eStringSubtree[k]);
}
Console.WriteLine("\nestring 2:");
for (int k = 0; k < eStringSubtreeB.Count; ++k)
{
Console.Write("{0}\t", eStringSubtreeB[k]);
}
alignedSequences = new List <ISequence>();
seq = profileAligner.GenerateSequenceFromEString(eStringSubtree, sequences[i]);
alignedSequences.Add(seq);
seq = profileAligner.GenerateSequenceFromEString(eStringSubtreeB, sequences[j]);
alignedSequences.Add(seq);
profileScore = MsaUtils.MultipleAlignmentScoreFunction(alignedSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
Console.WriteLine("\nalignment score is: {0}", profileScore);
Console.WriteLine("the aligned sequences are:");
for (int k = 0; k < alignedSequences.Count; ++k)
{
Console.WriteLine(new string(alignedSequences[k].Select(a => (char)a).ToArray()));
}
}
((FastAParser)parser).Dispose();
}
}
public void TestNeedlemanWunschProfileAligner()
{
Console.WriteLine("Number of logical processors: {0}", Environment.ProcessorCount);
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 = -3;
int gapExtendPenalty = -1;
IProfileAligner profileAligner = new NeedlemanWunschProfileAlignerSerial(similarityMatrix, ProfileScoreFunctionNames.WeightedInnerProduct,
gapOpenPenalty, gapExtendPenalty, Environment.ProcessorCount);
ISequence seqA = new Sequence(Alphabets.DNA, "GGGAAAAATCAGATT");
ISequence seqB = new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG");
List <ISequence> sequences = new List <ISequence>();
sequences.Add(seqA);
sequences.Add(seqB);
IProfileAlignment profileAlignmentA = ProfileAlignment.GenerateProfileAlignment(sequences[0]);
IProfileAlignment profileAlignmentB = ProfileAlignment.GenerateProfileAlignment(sequences[1]);
profileAligner.Align(profileAlignmentA, profileAlignmentB);
List <int> eStringSubtree = profileAligner.GenerateEString(profileAligner.AlignedA);
List <int> eStringSubtreeB = profileAligner.GenerateEString(profileAligner.AlignedB);
List <ISequence> alignedSequences = new List <ISequence>();
ISequence seq = profileAligner.GenerateSequenceFromEString(eStringSubtree, sequences[0]);
alignedSequences.Add(seq);
seq = profileAligner.GenerateSequenceFromEString(eStringSubtreeB, sequences[1]);
alignedSequences.Add(seq);
float profileScore = MsaUtils.MultipleAlignmentScoreFunction(alignedSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
Console.WriteLine("alignment score is: {0}", profileScore);
Console.WriteLine("the aligned sequences are:");
for (int i = 0; i < alignedSequences.Count; ++i)
{
Console.WriteLine(alignedSequences[i].ToString());
}
ISequence expectedSeqA = new Sequence(Alphabets.DNA, "GGGAA---AAATCAGATT");
ISequence expectedSeqB = new Sequence(Alphabets.DNA, "GGGAATCAAAATCAG---");
// Test on case 3: 36 sequences
ISequenceParser parser = new FastaParser();
string filepath = @"testdata\FASTA\RV11_BBS_all.afa";
IList <ISequence> orgSequences = parser.Parse(filepath);
sequences = MsaUtils.UnAlign(orgSequences);
int numberOfSequences = orgSequences.Count;
Console.WriteLine("Original unaligned sequences are:");
for (int i = 0; i < numberOfSequences; ++i)
{
Console.WriteLine(">");
Console.WriteLine(sequences[i].ToString());
}
for (int i = 1; i < numberOfSequences - 1; ++i)
{
for (int j = i + 1; j < numberOfSequences; ++j)
{
profileAlignmentA = ProfileAlignment.GenerateProfileAlignment(sequences[i]);
profileAlignmentB = ProfileAlignment.GenerateProfileAlignment(sequences[j]);
profileAligner = new NeedlemanWunschProfileAlignerSerial(similarityMatrix, ProfileScoreFunctionNames.WeightedInnerProduct,
gapOpenPenalty, gapExtendPenalty, Environment.ProcessorCount);
profileAligner.Align(profileAlignmentA, profileAlignmentB);
eStringSubtree = profileAligner.GenerateEString(profileAligner.AlignedA);
eStringSubtreeB = profileAligner.GenerateEString(profileAligner.AlignedB);
Console.WriteLine("Sequences lengths are: {0}-{1}", sequences[i].Count, sequences[j].Count);
Console.WriteLine("estring 1:");
for (int k = 0; k < eStringSubtree.Count; ++k)
{
Console.Write("{0}\t", eStringSubtree[k]);
}
Console.WriteLine("\nestring 2:");
for (int k = 0; k < eStringSubtreeB.Count; ++k)
{
Console.Write("{0}\t", eStringSubtreeB[k]);
}
alignedSequences = new List <ISequence>();
seq = profileAligner.GenerateSequenceFromEString(eStringSubtree, sequences[i]);
alignedSequences.Add(seq);
seq = profileAligner.GenerateSequenceFromEString(eStringSubtreeB, sequences[j]);
alignedSequences.Add(seq);
profileScore = MsaUtils.MultipleAlignmentScoreFunction(alignedSequences, similarityMatrix, gapOpenPenalty, gapExtendPenalty);
Console.WriteLine("\nalignment score is: {0}", profileScore);
Console.WriteLine("the aligned sequences are:");
for (int k = 0; k < alignedSequences.Count; ++k)
{
Console.WriteLine(alignedSequences[k].ToString());
}
}
}
}