private void ValidateSmithWatermanAlignment(string nodeName, bool isTextFile,
SequenceCaseType caseType, AlignParameters additionalParameter,
AlignmentType alignType,
SimilarityMatrixParameters similarityMatrixParam)
{
Sequence aInput, bInput;
IAlphabet alphabet =
Utility.GetAlphabet(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.AlphabetNameNode));
if (isTextFile)
{
// Read the xml file for getting both the files for aligning.
string filePath1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.FilePathNode1);
string filePath2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.FilePathNode2);
// Parse the files and get the sequence.
ISequence originalSequence1 = null;
ISequence originalSequence2 = null;
var parseObjectForFile1 = new FastAParser { Alphabet = alphabet };
originalSequence1 = parseObjectForFile1.Parse(filePath1).ElementAt(0);
originalSequence2 = parseObjectForFile1.Parse(filePath2).ElementAt(0);
// Create input sequence for sequence string in different cases.
GetSequenceWithCaseType(originalSequence1.ConvertToString(), originalSequence2.ConvertToString(),
alphabet, caseType, out aInput, out bInput);
}
else
{
string originalSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.SequenceNode1);
string originalSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.SequenceNode2);
// Create input sequence for sequence string in different cases.
GetSequenceWithCaseType(
originalSequence1,
originalSequence2,
alphabet,
caseType,
out aInput,
out bInput);
}
ApplicationLog.WriteLine(string.Format("SmithWatermanAligner P2 : First sequence used is '{0}'.",
aInput.ConvertToString()));
ApplicationLog.WriteLine(string.Format("SmithWatermanAligner P2 : Second sequence used is '{0}'.",
bInput.ConvertToString()));
// Create similarity matrix object for a given file.
string blosumFilePath = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.BlosumFilePathNode);
SimilarityMatrix sm;
switch (similarityMatrixParam)
{
case SimilarityMatrixParameters.TextReader:
using (TextReader reader = new StreamReader(blosumFilePath))
sm = new SimilarityMatrix(reader);
break;
case SimilarityMatrixParameters.DiagonalMatrix:
string matchValue = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.MatchScoreNode);
string misMatchValue = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.MisMatchScoreNode);
sm = new DiagonalSimilarityMatrix(int.Parse(matchValue, null),
int.Parse(misMatchValue, null));
break;
default:
sm = new SimilarityMatrix(new StreamReader(blosumFilePath));
break;
}
int gapOpenCost = int.Parse(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.GapOpenCostNode), null);
int gapExtensionCost = int.Parse(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.GapExtensionCostNode),
null);
// Create SmithWatermanAligner instance and set its values.
var smithWatermanObj = new SmithWatermanAligner();
if (additionalParameter != AlignParameters.AllParam)
{
smithWatermanObj.SimilarityMatrix = sm;
smithWatermanObj.GapOpenCost = gapOpenCost;
smithWatermanObj.GapExtensionCost = gapExtensionCost;
}
IList<IPairwiseSequenceAlignment> result = null;
// Align the input sequences.
switch (additionalParameter)
{
case AlignParameters.AlignList:
switch (alignType)
{
case AlignmentType.Align:
result = smithWatermanObj.Align(new List<ISequence> {aInput, bInput});
break;
default:
result = smithWatermanObj.AlignSimple(new List<ISequence> {aInput, bInput});
break;
}
break;
case AlignParameters.AlignTwo:
switch (alignType)
{
case AlignmentType.Align:
result = smithWatermanObj.Align(aInput, bInput);
break;
default:
result = smithWatermanObj.AlignSimple(aInput, bInput);
break;
}
break;
case AlignParameters.AllParam:
switch (alignType)
{
case AlignmentType.Align:
result = smithWatermanObj.Align(sm, gapOpenCost,
gapExtensionCost, aInput, bInput);
break;
default:
result = smithWatermanObj.AlignSimple(sm, gapOpenCost, aInput, bInput);
break;
}
break;
default:
break;
}
// Get the expected sequence and scorde from xml config.
string expectedSequence1, expectedSequence2, expectedScore;
switch (alignType)
{
case AlignmentType.Align:
expectedScore = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedGapExtensionScoreNode);
switch (caseType)
{
case SequenceCaseType.LowerCase:
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants
.ExpectedGapExtensionSequence1InLower);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants
.ExpectedGapExtensionSequence2InLower);
break;
default:
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants
.ExpectedGapExtensionSequence1Node);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants
.ExpectedGapExtensionSequence2Node);
break;
}
break;
default:
expectedScore = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedScoreNode);
switch (caseType)
{
case SequenceCaseType.LowerCase:
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequence1inLowerNode);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequence2inLowerNode);
break;
case SequenceCaseType.LowerUpperCase:
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequence1inLowerNode);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequenceNode2);
break;
default:
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequenceNode1);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequenceNode2);
break;
}
break;
}
// Match the alignment result with expected result.
IList<IPairwiseSequenceAlignment> expectedOutput = new List<IPairwiseSequenceAlignment>();
IPairwiseSequenceAlignment align = new PairwiseSequenceAlignment();
var alignedSeq = new PairwiseAlignedSequence
{
FirstSequence = new Sequence(alphabet, expectedSequence1),
SecondSequence = new Sequence(alphabet, expectedSequence2),
Score = Convert.ToInt32(expectedScore, null),
FirstOffset = Int32.MinValue,
SecondOffset = Int32.MinValue,
};
align.PairwiseAlignedSequences.Add(alignedSeq);
expectedOutput.Add(align);
ApplicationLog.WriteLine(string.Format(null, "SmithWatermanAligner P2 : Final Score '{0}'.", expectedScore));
ApplicationLog.WriteLine(string.Format(null, "SmithWatermanAligner P2 : Aligned First Sequence is '{0}'.",
expectedSequence1));
ApplicationLog.WriteLine(string.Format(null, "SmithWatermanAligner P2 : Aligned Second Sequence is '{0}'.",
expectedSequence2));
Assert.IsTrue(CompareAlignment(result, expectedOutput));
}
public void PairwiseOverlapMultipleAlignments()
{
Sequence sequence1 = new Sequence(Alphabets.DNA, "CCCAACCC");
Sequence sequence2 = new Sequence(Alphabets.DNA, "CCC");
SimilarityMatrix sm = new DiagonalSimilarityMatrix(5, -20);
int gapPenalty = -10;
PairwiseOverlapAligner overlap = new PairwiseOverlapAligner();
overlap.SimilarityMatrix = sm;
overlap.GapOpenCost = gapPenalty;
IList<IPairwiseSequenceAlignment> result = overlap.AlignSimple(sequence1, sequence2);
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"{0}, Simple; Matrix {1}; GapOpenCost {2}", overlap.Name, overlap.SimilarityMatrix.Name, overlap.GapOpenCost));
foreach (IPairwiseSequenceAlignment sequenceResult in result)
{
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"score {0}", sequenceResult.PairwiseAlignedSequences[0].Score));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"input 0 {0}", sequenceResult.FirstSequence.ToString()));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"input 1 {0}", sequenceResult.SecondSequence.ToString()));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"result 0 {0}", sequenceResult.PairwiseAlignedSequences[0].FirstSequence.ToString()));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"result 1 {0}", sequenceResult.PairwiseAlignedSequences[0].SecondSequence.ToString()));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"consesus {0}", sequenceResult.PairwiseAlignedSequences[0].Consensus.ToString()));
}
IList<IPairwiseSequenceAlignment> expectedOutput = new List<IPairwiseSequenceAlignment>();
IPairwiseSequenceAlignment align = new PairwiseSequenceAlignment();
// First alignment
PairwiseAlignedSequence alignedSeq = new PairwiseAlignedSequence();
alignedSeq.FirstSequence = new Sequence(Alphabets.DNA, "CCC");
alignedSeq.SecondSequence = new Sequence(Alphabets.DNA, "CCC");
alignedSeq.Consensus = new Sequence(Alphabets.DNA, "CCC");
alignedSeq.Score = 15;
alignedSeq.FirstOffset = 0;
alignedSeq.SecondOffset = 0;
align.PairwiseAlignedSequences.Add(alignedSeq);
// Second alignment
alignedSeq = new PairwiseAlignedSequence();
alignedSeq.FirstSequence = new Sequence(Alphabets.DNA, "CCC");
alignedSeq.SecondSequence = new Sequence(Alphabets.DNA, "CCC");
alignedSeq.Consensus = new Sequence(Alphabets.DNA, "CCC");
alignedSeq.Score = 15;
alignedSeq.FirstOffset = 0;
alignedSeq.SecondOffset = 5;
align.PairwiseAlignedSequences.Add(alignedSeq);
expectedOutput.Add(align);
Assert.IsTrue(CompareAlignment(result, expectedOutput));
}
/// <summary>
/// Validates PairwiseOverlapAlignment algorithm for the parameters passed.
/// </summary>
/// <param name="nodeName">Node Name in the xml.</param>
/// <param name="alignParam">parameter based on which certain validations are done.</param>
/// <param name="similarityMatrixParam">Similarity Matrix Parameter.</param>
/// <param name="alignType">Alignment Type</param>
private void ValidatePairwiseOverlapAlignment(string nodeName, AlignParameters alignParam,
SimilarityMatrixParameters similarityMatrixParam,
AlignmentType alignType)
{
ISequence aInput;
ISequence bInput;
IAlphabet alphabet = Utility.GetAlphabet(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.AlphabetNameNode));
if (alignParam.ToString().Contains("Code"))
{
string sequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.SequenceNode1);
string sequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.SequenceNode2);
aInput = new Sequence(alphabet, sequence1);
bInput = new Sequence(alphabet, sequence2);
}
else
{
// Read the xml file for getting both the files for aligning.
string filePath1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.FilePathNode1);
string filePath2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.FilePathNode2);
var parser1 = new FastAParser { Alphabet = alphabet };
aInput = parser1.Parse(filePath1).ElementAt(0);
bInput = parser1.Parse(filePath2).ElementAt(0);
}
string blosumFilePath = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.BlosumFilePathNode);
SimilarityMatrix sm;
switch (similarityMatrixParam)
{
case SimilarityMatrixParameters.TextReader:
using (TextReader reader = new StreamReader(blosumFilePath))
sm = new SimilarityMatrix(reader);
break;
case SimilarityMatrixParameters.DiagonalMatrix:
string matchValue = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.MatchScoreNode);
string misMatchValue = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.MisMatchScoreNode);
sm = new DiagonalSimilarityMatrix(int.Parse(matchValue, null), int.Parse(misMatchValue, null));
break;
default:
sm = new SimilarityMatrix(new StreamReader(blosumFilePath));
break;
}
int gapOpenCost = int.Parse(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.GapOpenCostNode), null);
int gapExtensionCost = int.Parse(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.GapExtensionCostNode), null);
var pairwiseOverlapObj = new PairwiseOverlapAligner();
if (AlignParameters.AllParam != alignParam)
{
pairwiseOverlapObj.SimilarityMatrix = sm;
pairwiseOverlapObj.GapOpenCost = gapOpenCost;
}
IList<IPairwiseSequenceAlignment> result = null;
switch (alignParam)
{
case AlignParameters.AlignList:
case AlignParameters.AlignListCode:
var sequences = new List<ISequence> {aInput, bInput};
switch (alignType)
{
case AlignmentType.Align:
pairwiseOverlapObj.GapExtensionCost = gapExtensionCost;
result = pairwiseOverlapObj.Align(sequences);
break;
default:
result = pairwiseOverlapObj.AlignSimple(sequences);
break;
}
break;
case AlignParameters.AllParam:
case AlignParameters.AllParamCode:
switch (alignType)
{
case AlignmentType.Align:
pairwiseOverlapObj.GapExtensionCost = gapExtensionCost;
result = pairwiseOverlapObj.Align(sm, gapOpenCost, gapExtensionCost, aInput, bInput);
break;
default:
result = pairwiseOverlapObj.AlignSimple(sm, gapOpenCost, aInput, bInput);
break;
}
break;
case AlignParameters.AlignTwo:
case AlignParameters.AlignTwoCode:
switch (alignType)
{
case AlignmentType.Align:
pairwiseOverlapObj.GapExtensionCost = gapExtensionCost;
result = pairwiseOverlapObj.Align(aInput, bInput);
break;
default:
result = pairwiseOverlapObj.AlignSimple(aInput, bInput);
break;
}
break;
default:
break;
}
// Read the xml file for getting both the files for aligning.
string expectedSequence1;
string expectedSequence2;
string expectedScore;
switch (alignType)
{
case AlignmentType.Align:
expectedScore = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedGapExtensionScoreNode);
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedGapExtensionSequence1Node);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedGapExtensionSequence2Node);
break;
default:
expectedScore = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedScoreNode);
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedSequenceNode1);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedSequenceNode2);
break;
}
IList<IPairwiseSequenceAlignment> expectedOutput = new List<IPairwiseSequenceAlignment>();
var seperators = new [] {';'};
string[] expectedSequences1 = expectedSequence1.Split(seperators);
string[] expectedSequences2 = expectedSequence2.Split(seperators);
IPairwiseSequenceAlignment align = new PairwiseSequenceAlignment();
for (int i = 0; i < expectedSequences1.Length; i++)
{
PairwiseAlignedSequence alignedSeq = new PairwiseAlignedSequence
{
FirstSequence = new Sequence(alphabet, expectedSequences1[i]),
SecondSequence = new Sequence(alphabet, expectedSequences2[i]),
Score = Convert.ToInt32(expectedScore, null),
FirstOffset = Int32.MinValue,
SecondOffset = Int32.MinValue,
};
align.PairwiseAlignedSequences.Add(alignedSeq);
}
expectedOutput.Add(align);
Assert.IsTrue(AlignmentHelpers.CompareAlignment(result, expectedOutput, true));
ApplicationLog.WriteLine(string.Format(null, "PairwiseOverlapAligner P1 : Final Score '{0}'.", expectedScore));
ApplicationLog.WriteLine(string.Format(null, "PairwiseOverlapAligner P1 : Aligned First Sequence is '{0}'.", expectedSequence1));
ApplicationLog.WriteLine(string.Format(null, "PairwiseOverlapAligner P1 : Aligned Second Sequence is '{0}'.", expectedSequence2));
}
public void PairwiseOverlapProteinSeqWithZeroOverlap()
{
Sequence sequence1 = new Sequence(Alphabets.Protein, "ACDEF");
Sequence sequence2 = new Sequence(Alphabets.Protein, "TUVWY");
SimilarityMatrix sm = new DiagonalSimilarityMatrix(5, -5);
int gapPenalty = -10;
PairwiseOverlapAligner overlap = new PairwiseOverlapAligner();
overlap.SimilarityMatrix = sm;
overlap.GapOpenCost = gapPenalty;
overlap.GapExtensionCost = -1;
IList<IPairwiseSequenceAlignment> result = overlap.Align(sequence1, sequence2);
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"{0}, Simple; Matrix {1}; GapOpenCost {2}", overlap.Name, overlap.SimilarityMatrix.Name, overlap.GapOpenCost));
foreach (IPairwiseSequenceAlignment sequenceResult in result)
{
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"score {0}", sequenceResult.PairwiseAlignedSequences[0].Score));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"input 0 {0}", sequenceResult.FirstSequence.ToString()));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"input 1 {0}", sequenceResult.SecondSequence.ToString()));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"result 0 {0}", sequenceResult.PairwiseAlignedSequences[0].FirstSequence.ToString()));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"result 1 {0}", sequenceResult.PairwiseAlignedSequences[0].SecondSequence.ToString()));
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"consesus {0}", sequenceResult.PairwiseAlignedSequences[0].Consensus.ToString()));
}
IList<IPairwiseSequenceAlignment> expectedOutput = new List<IPairwiseSequenceAlignment>();
Assert.IsTrue(CompareAlignment(result, expectedOutput));
}
/// <summary>
/// Validates PairwiseOverlapAlignment algorithm for the parameters passed.
/// </summary>
/// <param name="nodeName">Xml node name</param>
/// <param name="isTextFile">Is text file an input.</param>
/// <param name="caseType">Case Type</param>
/// <param name="additionalParameter">parameter based on which certain validations are done.</param>
/// <param name="alignType">Is the Align type Simple or Align with Gap Extension cost?</param>
/// <param name="similarityMatrixParam">Similarity Matrix</param>
private void ValidatePairwiseOverlapAlignment(string nodeName, bool isTextFile,
SequenceCaseType caseType, AlignParameters additionalParameter,
AlignmentType alignType,
SimilarityMatrixParameters similarityMatrixParam)
{
Sequence aInput = null;
Sequence bInput = null;
IAlphabet alphabet = Utility.GetAlphabet(this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.AlphabetNameNode));
if (isTextFile)
{
// Read the xml file for getting both the files for aligning.
string filePath1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.FilePathNode1);
string filePath2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.FilePathNode2);
var parser1 = new FastAParser();
ISequence originalSequence1 = parser1.Parse(filePath1).ElementAt(0);
ISequence originalSequence2 = parser1.Parse(filePath2).ElementAt(0);
// Create input sequence for sequence string in different cases.
GetSequenceWithCaseType(new string(originalSequence1.Select(a => (char) a).ToArray()),
new string(originalSequence2.Select(a => (char) a).ToArray()), alphabet,
caseType, out aInput, out bInput);
}
else
{
string originalSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.SequenceNode1);
string originalSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.SequenceNode2);
// Create input sequence for sequence string in different cases.
GetSequenceWithCaseType(
originalSequence1,
originalSequence2,
alphabet,
caseType,
out aInput,
out bInput);
}
var aInputString = new string(aInput.Select(a => (char) a).ToArray());
var bInputString = new string(bInput.Select(a => (char) a).ToArray());
ApplicationLog.WriteLine(string.Format(null,
"PairwiseOverlapAligner P2 : First sequence used is '{0}'.",
aInputString));
ApplicationLog.WriteLine(string.Format(null,
"PairwiseOverlapAligner P2 : Second sequence used is '{0}'.",
bInputString));
// Create similarity matrix object for a given file.
string blosumFilePath = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.BlosumFilePathNode);
SimilarityMatrix sm = null;
switch (similarityMatrixParam)
{
case SimilarityMatrixParameters.TextReader:
using (TextReader reader = new StreamReader(blosumFilePath))
sm = new SimilarityMatrix(reader);
break;
case SimilarityMatrixParameters.DiagonalMatrix:
string matchValue = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.MatchScoreNode);
string misMatchValue = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.MisMatchScoreNode);
sm = new DiagonalSimilarityMatrix(int.Parse(matchValue, null),
int.Parse(misMatchValue, null));
break;
default:
sm = new SimilarityMatrix(new StreamReader(blosumFilePath));
break;
}
int gapOpenCost = int.Parse(this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.GapOpenCostNode), null);
int gapExtensionCost = int.Parse(this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.GapExtensionCostNode), null);
// Create PairwiseOverlapAligner instance and set its values.
var pairwiseOverlapObj = new PairwiseOverlapAligner();
if (additionalParameter != AlignParameters.AllParam)
{
pairwiseOverlapObj.SimilarityMatrix = sm;
pairwiseOverlapObj.GapOpenCost = gapOpenCost;
pairwiseOverlapObj.GapExtensionCost = gapExtensionCost;
}
IList<IPairwiseSequenceAlignment> result = null;
// Align the input sequences.
switch (additionalParameter)
{
case AlignParameters.AlignList:
var sequences = new List<ISequence>();
sequences.Add(aInput);
sequences.Add(bInput);
switch (alignType)
{
case AlignmentType.Align:
result = pairwiseOverlapObj.Align(sequences);
break;
default:
result = pairwiseOverlapObj.AlignSimple(sequences);
break;
}
break;
case AlignParameters.AlignTwo:
switch (alignType)
{
case AlignmentType.Align:
result = pairwiseOverlapObj.Align(aInput, bInput);
break;
default:
result = pairwiseOverlapObj.AlignSimple(aInput, bInput);
break;
}
break;
case AlignParameters.AllParam:
switch (alignType)
{
case AlignmentType.Align:
result = pairwiseOverlapObj.Align(sm, gapOpenCost,
gapExtensionCost, aInput, bInput);
break;
default:
result = pairwiseOverlapObj.AlignSimple(sm, gapOpenCost, aInput, bInput);
break;
}
break;
default:
break;
}
aInput = null;
bInput = null;
sm = null;
// Get the expected sequence and scorde from xml config.
string expectedSequence1 = string.Empty;
string expectedSequence2 = string.Empty;
string expectedScore = string.Empty;
switch (alignType)
{
case AlignmentType.Align:
expectedScore = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedGapExtensionScoreNode);
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedGapExtensionSequence1Node);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedGapExtensionSequence2Node);
break;
default:
expectedScore = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedScoreNode);
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequenceNode1);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedSequenceNode2);
break;
}
IList<IPairwiseSequenceAlignment> expectedOutput = new List<IPairwiseSequenceAlignment>();
string[] expectedSequences1, expectedSequences2;
var seperators = new char[1] {';'};
expectedSequences1 = expectedSequence1.Split(seperators);
expectedSequences2 = expectedSequence2.Split(seperators);
IPairwiseSequenceAlignment align = new PairwiseSequenceAlignment();
PairwiseAlignedSequence alignedSeq;
for (int i = 0; i < expectedSequences1.Length; i++)
{
alignedSeq = new PairwiseAlignedSequence
{
FirstSequence = new Sequence(alphabet, expectedSequences1[i]),
SecondSequence = new Sequence(alphabet, expectedSequences2[i]),
Score = Convert.ToInt32(expectedScore, null),
FirstOffset = Int32.MinValue,
SecondOffset = Int32.MinValue,
};
align.PairwiseAlignedSequences.Add(alignedSeq);
}
expectedOutput.Add(align);
Assert.IsTrue(AlignmentHelpers.CompareAlignment(result, expectedOutput,true));
ApplicationLog.WriteLine(string.Format(null, "PairwiseOverlapAligner P2 : Final Score '{0}'.", expectedScore));
ApplicationLog.WriteLine(string.Format(null, "PairwiseOverlapAligner P2 : Aligned First Sequence is '{0}'.", expectedSequence1));
ApplicationLog.WriteLine(string.Format(null, "PairwiseOverlapAligner P2 : Aligned Second Sequence is '{0}'.", expectedSequence2));
}
/// <summary>
/// Default constructor
/// </summary>
protected PairwiseSequenceAligner()
{
SimilarityMatrix = new DiagonalSimilarityMatrix(2, -2);
GapOpenCost = -8;
GapExtensionCost = -1;
IncludeScoreTable = false;
}
public void TestMUMmer3MultipleMumWithCustomMatrix()
{
string reference = "ATGCGCATCCCCTT";
string search = "GCGCCCCCTA";
Sequence referenceSeq = null;
Sequence searchSeq = null;
referenceSeq = new Sequence(Alphabets.DNA, reference);
searchSeq = new Sequence(Alphabets.DNA, search);
List<ISequence> searchSeqs = new List<ISequence>();
searchSeqs.Add(searchSeq);
int[,] customMatrix = new int[256, 256];
customMatrix[(byte)'A', (byte)'A'] = 3;
customMatrix[(byte)'A', (byte)'T'] = -2;
customMatrix[(byte)'A', (byte)'G'] = -2;
customMatrix[(byte)'A', (byte)'c'] = -2;
customMatrix[(byte)'G', (byte)'G'] = 3;
customMatrix[(byte)'G', (byte)'A'] = -2;
customMatrix[(byte)'G', (byte)'T'] = -2;
customMatrix[(byte)'G', (byte)'C'] = -2;
customMatrix[(byte)'T', (byte)'T'] = 3;
customMatrix[(byte)'T', (byte)'A'] = -2;
customMatrix[(byte)'T', (byte)'G'] = -2;
customMatrix[(byte)'T', (byte)'C'] = -2;
customMatrix[(byte)'C', (byte)'C'] = 3;
customMatrix[(byte)'C', (byte)'T'] = -2;
customMatrix[(byte)'C', (byte)'A'] = -2;
customMatrix[(byte)'C', (byte)'G'] = -2;
DiagonalSimilarityMatrix matrix = new DiagonalSimilarityMatrix(3, -2);
int gapOpenCost = -6;
MUMmerAligner mummer = new MUMmerAligner();
mummer.LengthOfMUM = 4;
mummer.PairWiseAlgorithm = new NeedlemanWunschAligner();
mummer.SimilarityMatrix = matrix;
mummer.GapOpenCost = gapOpenCost;
mummer.GapExtensionCost = -2;
IList<IPairwiseSequenceAlignment> result = mummer.AlignSimple(referenceSeq, searchSeqs);
// Check if output is not null
Assert.AreNotEqual(null, result);
IList<IPairwiseSequenceAlignment> expectedOutput = new List<IPairwiseSequenceAlignment>();
IPairwiseSequenceAlignment align = new PairwiseSequenceAlignment();
PairwiseAlignedSequence alignedSeq = new PairwiseAlignedSequence();
alignedSeq.FirstSequence = new Sequence(Alphabets.DNA, "ATGCGCATCCCCTT");
alignedSeq.SecondSequence = new Sequence(Alphabets.DNA, "--GCGC--CCCCTA");
alignedSeq.Consensus = new Sequence(AmbiguousDnaAlphabet.Instance, "ATGCGCATCCCCTW");
alignedSeq.Score = 1;
alignedSeq.FirstOffset = 0;
alignedSeq.SecondOffset = 2;
align.PairwiseAlignedSequences.Add(alignedSeq);
expectedOutput.Add(align);
Assert.IsTrue(CompareAlignment(result, expectedOutput));
}
private void ValidateNeedlemanWunschAlignment(string nodeName, AlignParameters alignParam,
SimilarityMatrixParameters similarityMatrixParam,
AlignmentType alignType)
{
ISequence aInput, bInput;
IAlphabet alphabet =
Utility.GetAlphabet(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.AlphabetNameNode));
// Parse the files and get the sequence.
if (alignParam.ToString().Contains("Code"))
{
string sequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.SequenceNode1);
string sequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.SequenceNode2);
aInput = new Sequence(alphabet, sequence1);
bInput = new Sequence(alphabet, sequence2);
}
else
{
// Read the xml file for getting both the files for aligning.
string filePath1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.FilePathNode1);
string filePath2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.FilePathNode2);
var parseObjectForFile1 = new FastAParser { Alphabet = alphabet };
ISequence originalSequence1 = parseObjectForFile1.Parse(filePath1).FirstOrDefault();
Assert.IsNotNull(originalSequence1);
aInput = new Sequence(alphabet, originalSequence1.ConvertToString());
var parseObjectForFile2 = new FastAParser { Alphabet = alphabet };
ISequence originalSequence2 = parseObjectForFile2.Parse(filePath2).FirstOrDefault();
Assert.IsNotNull(originalSequence2);
bInput = new Sequence(alphabet, originalSequence2.ConvertToString());
}
string blosumFilePath = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.BlosumFilePathNode);
SimilarityMatrix sm;
switch (similarityMatrixParam)
{
case SimilarityMatrixParameters.TextReader:
using (TextReader reader = new StreamReader(blosumFilePath))
sm = new SimilarityMatrix(reader);
break;
case SimilarityMatrixParameters.DiagonalMatrix:
string matchValue = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.MatchScoreNode);
string misMatchValue = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.MisMatchScoreNode);
sm = new DiagonalSimilarityMatrix(int.Parse(matchValue, null),
int.Parse(misMatchValue, null));
break;
default:
sm = new SimilarityMatrix(new StreamReader(blosumFilePath));
break;
}
int gapOpenCost = int.Parse(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.GapOpenCostNode), null);
int gapExtensionCost = int.Parse(this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.GapExtensionCostNode),
null);
var needlemanWunschObj = new NeedlemanWunschAligner();
if (AlignParameters.AllParam != alignParam)
{
needlemanWunschObj.SimilarityMatrix = sm;
needlemanWunschObj.GapOpenCost = gapOpenCost;
}
IList<IPairwiseSequenceAlignment> result = null;
switch (alignParam)
{
case AlignParameters.AlignList:
case AlignParameters.AlignListCode:
var sequences = new List<ISequence> {aInput, bInput};
switch (alignType)
{
case AlignmentType.Align:
needlemanWunschObj.GapExtensionCost = gapExtensionCost;
result = needlemanWunschObj.Align(sequences);
break;
default:
result = needlemanWunschObj.AlignSimple(sequences);
break;
}
break;
case AlignParameters.AllParam:
case AlignParameters.AllParamCode:
switch (alignType)
{
case AlignmentType.Align:
needlemanWunschObj.GapExtensionCost = gapExtensionCost;
result = needlemanWunschObj.Align(sm,
gapOpenCost, gapExtensionCost, aInput, bInput);
break;
default:
result = needlemanWunschObj.AlignSimple(sm, gapOpenCost, aInput, bInput);
break;
}
break;
case AlignParameters.AlignTwo:
case AlignParameters.AlignTwoCode:
switch (alignType)
{
case AlignmentType.Align:
needlemanWunschObj.GapExtensionCost = gapExtensionCost;
result = needlemanWunschObj.Align(aInput, bInput);
break;
default:
result = needlemanWunschObj.AlignSimple(aInput, bInput);
break;
}
break;
default:
break;
}
// Read the xml file for getting both the files for aligning.
string expectedSequence1, expectedSequence2, expectedScore;
switch (alignType)
{
case AlignmentType.Align:
expectedScore = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedGapExtensionScoreNode);
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedGapExtensionSequence1Node);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedGapExtensionSequence2Node);
break;
default:
expectedScore = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedScoreNode);
expectedSequence1 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedSequenceNode1);
expectedSequence2 = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.ExpectedSequenceNode2);
break;
}
IList<IPairwiseSequenceAlignment> expectedOutput = new List<IPairwiseSequenceAlignment>();
IPairwiseSequenceAlignment align = new PairwiseSequenceAlignment(aInput, bInput);
var alignedSeq = new PairwiseAlignedSequence
{
FirstSequence = new Sequence(alphabet, expectedSequence1),
SecondSequence = new Sequence(alphabet, expectedSequence2),
Score = Convert.ToInt32(expectedScore, null)
};
align.PairwiseAlignedSequences.Add(alignedSeq);
expectedOutput.Add(align);
ApplicationLog.WriteLine(string.Format("NeedlemanWunschAligner P1 : Final Score '{0}'.", expectedScore));
ApplicationLog.WriteLine(string.Format("NeedlemanWunschAligner P1 : Aligned First Sequence is '{0}'.", expectedSequence1));
ApplicationLog.WriteLine(string.Format("NeedlemanWunschAligner P1 : Aligned Second Sequence is '{0}'.", expectedSequence2));
Assert.IsTrue(CompareAlignment(result, expectedOutput));
}
