public void BuildScaffold()
{
const int kmerLength = 6;
const int dangleThreshold = 3;
const int redundantThreshold = 7;
List <ISequence> sequences = TestInputs.GetReadsForScaffolds();
KmerLength = kmerLength;
SequenceReads.Clear();
this.AddSequenceReads(sequences);
CreateGraph();
DanglingLinksThreshold = dangleThreshold;
DanglingLinksPurger = new DanglingLinksPurger(dangleThreshold);
RedundantPathLengthThreshold = redundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(redundantThreshold);
UnDangleGraph();
RemoveRedundancy();
IList <ISequence> contigs = BuildContigs();
DeBruijnGraph graph = Graph;
CloneLibrary.Instance.AddLibrary("abc", (float)5, (float)20);
GraphScaffoldBuilder scaffold = new GraphScaffoldBuilder();
IList <ISequence> scaffoldSeq = scaffold.BuildScaffold(
sequences, contigs, this.KmerLength, 3, 0);
Assert.AreEqual(scaffoldSeq.Count, 8);
Assert.IsTrue(scaffoldSeq[0].ToString().Equals(
"ATGCCTCCTATCTTAGCGCGC"));
}
public void TestDanglingLinksPurger()
{
const int KmerLength = 11;
const int DangleThreshold = 3;
List <ISequence> readSeqs = TestInputs.GetDanglingReads();
SequenceReads.Clear();
this.AddSequenceReads(readSeqs);
this.KmerLength = KmerLength;
DanglingLinksThreshold = DangleThreshold;
DanglingLinksPurger = new DanglingLinksPurger(DangleThreshold);
CreateGraph();
int graphCount = Graph.Nodes.Count;
int graphEdges = Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
HashSet <string> graphNodes = new HashSet <string>(
Graph.Nodes.Select(n => Graph.GetNodeSequence(n).ToString()));
DanglingLinksThreshold = DangleThreshold;
UnDangleGraph();
int dangleRemovedGraphCount = Graph.Nodes.Count;
int dangleRemovedGraphEdge = Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
HashSet <string> dangleRemovedGraphNodes = new HashSet <string>(
Graph.Nodes.Select(n => Graph.GetNodeSequence(n).ToString()));
// Compare the two graphs
Assert.AreEqual(2, graphCount - dangleRemovedGraphCount);
Assert.AreEqual(4, graphEdges - dangleRemovedGraphEdge);
graphNodes.ExceptWith(dangleRemovedGraphNodes);
Assert.IsTrue(graphNodes.Contains("TCGAACGATGA"));
Assert.IsTrue(graphNodes.Contains("ATCGAACGATG"));
}
public void TestRedundantPathsPurger()
{
const int KmerLength = 5;
const int RedundantThreshold = 10;
List <ISequence> readSeqs = TestInputs.GetRedundantPathReads();
this.SequenceReads.Clear();
this.AddSequenceReads(readSeqs);
this.KmerLength = KmerLength;
this.RedundantPathLengthThreshold = RedundantThreshold;
this.RedundantPathsPurger = new RedundantPathsPurger(RedundantThreshold);
this.CreateGraph();
int graphCount = this.Graph.Nodes.Count;
int graphEdges = this.Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
this.RemoveRedundancy();
int redundancyRemovedGraphCount = this.Graph.Nodes.Count;
int redundancyRemovedGraphEdge = this.Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
// Compare the two graphs
Assert.AreEqual(5, graphCount - redundancyRemovedGraphCount);
Assert.AreEqual(12, graphEdges - redundancyRemovedGraphEdge);
}
public void TestDeBruijnGraphBuilderTiny()
{
const int KmerLength = 3;
List <ISequence> reads = TestInputs.GetTinyReads();
this.KmerLength = KmerLength;
this.SequenceReads.Clear();
this.AddSequenceReads(reads);
this.CreateGraph();
DeBruijnGraph graph = this.Graph;
Assert.AreEqual(9, graph.Nodes.Count);
HashSet <string> nodeStrings = new HashSet <string>(graph.Nodes.Select(n => graph.GetNodeSequence(n).ToString()));
Assert.IsTrue(nodeStrings.Contains("ATG") || nodeStrings.Contains("CAT"));
Assert.IsTrue(nodeStrings.Contains("TGC") || nodeStrings.Contains("GCA"));
Assert.IsTrue(nodeStrings.Contains("GCC") || nodeStrings.Contains("GGC"));
Assert.IsTrue(nodeStrings.Contains("TCC") || nodeStrings.Contains("GGA"));
Assert.IsTrue(nodeStrings.Contains("CCT") || nodeStrings.Contains("AGG"));
Assert.IsTrue(nodeStrings.Contains("CTA") || nodeStrings.Contains("TAG"));
Assert.IsTrue(nodeStrings.Contains("TAT") || nodeStrings.Contains("ATA"));
Assert.IsTrue(nodeStrings.Contains("ATC") || nodeStrings.Contains("GAT"));
Assert.IsTrue(nodeStrings.Contains("CTC") || nodeStrings.Contains("GAG"));
int totalEdges = graph.Nodes.Select(n => n.ExtensionsCount).Sum();
Assert.AreEqual(31, totalEdges);
}
public void TestDeBruijnGraphBuilderSmall()
{
const int KmerLength = 6;
List <ISequence> reads = TestInputs.GetSmallReads();
this.KmerLength = KmerLength;
this.SequenceReads.Clear();
this.AddSequenceReads(reads);
this.CreateGraph();
DeBruijnGraph graph = this.Graph;
Assert.AreEqual(20, graph.Nodes.Count);
HashSet <string> nodeStrings = GetGraphNodesForSmallReads();
string nodeStr, nodeStrRC;
foreach (DeBruijnNode node in graph.Nodes)
{
nodeStr = graph.GetNodeSequence(node).ToString();
nodeStrRC = graph.GetNodeSequence(node).ReverseComplement.ToString();
Assert.IsTrue(nodeStrings.Contains(nodeStr) || nodeStrings.Contains(nodeStrRC));
}
int totalEdges = graph.Nodes.Select(n => n.ExtensionsCount).Sum();
Assert.AreEqual(51, totalEdges);
}
public void TestContigBuilder2()
{
const int KmerLength = 6;
const int RedundantThreshold = 10;
List <ISequence> readSeqs = TestInputs.GetRedundantPathReads();
SequenceReads.Clear();
this.AddSequenceReads(readSeqs);
this.KmerLength = KmerLength;
RedundantPathLengthThreshold = RedundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(RedundantThreshold);
ContigBuilder = new SimplePathContigBuilder();
CreateGraph();
RemoveRedundancy();
int graphCount = Graph.Nodes.Count;
int graphEdges = Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
IList <ISequence> contigs = BuildContigs();
int contigsBuiltGraphCount = Graph.Nodes.Count;
int contigsBuilt = Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
// Compare the two graphs
Assert.AreEqual(1, contigs.Count);
Assert.AreEqual("ATGCCTCCTATCTTAGCGATGCGGTGT", contigs[0].ToString());
Assert.AreEqual(graphCount, contigsBuiltGraphCount);
Assert.AreEqual(graphEdges, contigsBuilt);
}
public void AssemblerTest()
{
const int KmerLength = 11;
const int DangleThreshold = 3;
const int RedundantThreshold = 10;
List <ISequence> readSeqs = TestInputs.GetDanglingReads();
ParallelDeNovoAssembler assembler = new ParallelDeNovoAssembler();
assembler.KmerLength = KmerLength;
assembler.DanglingLinksThreshold = DangleThreshold;
assembler.RedundantPathLengthThreshold = RedundantThreshold;
IDeNovoAssembly result = assembler.Assemble(readSeqs);
// Compare the two graphs
Assert.AreEqual(1, result.AssembledSequences.Count);
HashSet <string> expectedContigs = new HashSet <string>()
{
"ATCGCTAGCATCGAACGATCATT"
};
foreach (ISequence contig in result.AssembledSequences)
{
Assert.IsTrue(expectedContigs.Contains(contig.ToString()));
}
}
public void TestContigGraphBuilder2()
{
const int KmerLength = 6;
const int RedundantThreshold = 10;
List <ISequence> readSeqs = TestInputs.GetRedundantPathReads();
this.SequenceReads.Clear();
this.AddSequenceReads(readSeqs);
this.KmerLength = KmerLength;
this.RedundantPathLengthThreshold = RedundantThreshold;
this.RedundantPathsPurger = new RedundantPathsPurger(RedundantThreshold);
this.ContigBuilder = new SimplePathContigBuilder();
this.CreateGraph();
this.RemoveRedundancy();
IList <ISequence> contigs = this.BuildContigs();
this.Graph.BuildContigGraph(contigs, KmerLength);
int contigGraphCount = this.Graph.Nodes.Count;
int contigGraphEdges = this.Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
Assert.AreEqual(contigs.Count, contigGraphCount);
Assert.AreEqual(0, contigGraphEdges);
HashSet <string> contigSeqs = new HashSet <string>(contigs.Select(c => c.ToString()));
foreach (DeBruijnNode node in this.Graph.Nodes)
{
Assert.IsTrue(contigSeqs.Contains(this.Graph.GetNodeSequence(node).ToString()));
}
}
public void TestContigBuilder1()
{
const int KmerLength = 11;
const int DangleThreshold = 3;
const int RedundantThreshold = 10;
List <ISequence> readSeqs = TestInputs.GetDanglingReads();
this.SequenceReads.Clear();
this.AddSequenceReads(readSeqs);
this.KmerLength = KmerLength;
DanglingLinksThreshold = DangleThreshold;
DanglingLinksPurger = new DanglingLinksPurger(DangleThreshold);
RedundantPathLengthThreshold = RedundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(RedundantThreshold);
ContigBuilder = new SimplePathContigBuilder();
CreateGraph();
UnDangleGraph();
RemoveRedundancy();
int graphCount = Graph.Nodes.Count;
int graphEdges = Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
IList <ISequence> contigs = BuildContigs();
int contigsBuiltGraphCount = this.Graph.Nodes.Count;
int contigsBuilt = Graph.Nodes.Select(n => n.ExtensionsCount).Sum();
// Compare the two graphs
Assert.AreEqual(1, contigs.Count);
HashSet <string> expectedContigs = new HashSet <string>()
{
"ATCGCTAGCATCGAACGATCATT"
};
foreach (ISequence contig in contigs)
{
Assert.IsTrue(expectedContigs.Contains(contig.ToString()));
}
Assert.AreEqual(graphCount, contigsBuiltGraphCount);
Assert.AreEqual(graphEdges, contigsBuilt);
}
public void AssemblerTestWithScaffoldBuilder()
{
const int kmerLength = 6;
const int dangleThreshold = 3;
const int redundantThreshold = 7;
ParallelDeNovoAssembler assembler = new ParallelDeNovoAssembler();
assembler.KmerLength = kmerLength;
assembler.DanglingLinksThreshold = dangleThreshold;
assembler.RedundantPathLengthThreshold = redundantThreshold;
assembler.ScaffoldRedundancy = 0;
assembler.Depth = 3;
CloneLibrary.Instance.AddLibrary("abc", (float)5, (float)20);
PaDeNAAssembly result = (PaDeNAAssembly)assembler.Assemble(TestInputs.GetReadsForScaffolds(), true);
Assert.AreEqual(10, result.ContigSequences.Count);
HashSet <string> expectedContigs = new HashSet <string>
{
"GCGCGC",
"TTTTTT",
"TTTTTA",
"TTTTAA",
"TTTAAA",
"ATGCCTCCTATCTTAGC",
"TTTTAGC",
"TTAGCGCG",
"CGCGCCGCGC",
"CGCGCG"
};
foreach (ISequence contig in result.ContigSequences)
{
string contigSeq = contig.ToString();
Assert.IsTrue(
expectedContigs.Contains(contigSeq) ||
expectedContigs.Contains(contigSeq.GetReverseComplement(new char[contigSeq.Length])));
}
Assert.AreEqual(8, result.Scaffolds.Count);
HashSet <string> expectedScaffolds = new HashSet <string>
{
"ATGCCTCCTATCTTAGCGCGC",
"TTTTTT",
"TTTTTA",
"TTTTAA",
"TTTAAA",
"CGCGCCGCGC",
"TTTTAGC",
"CGCGCG"
};
foreach (ISequence scaffold in result.Scaffolds)
{
string scaffoldSeq = scaffold.ToString();
Assert.IsTrue(
expectedScaffolds.Contains(scaffoldSeq) ||
expectedScaffolds.Contains(scaffoldSeq.GetReverseComplement(new char[scaffoldSeq.Length])));
}
}