public void BuildScaffold()
{
const int kmerLength = 6;
const int dangleThreshold = 3;
const int redundantThreshold = 7;
var sequences = new List <ISequence>(TestInputs.GetReadsForScaffolds());
KmerLength = kmerLength;
SequenceReads.Clear();
this.SetSequenceReads(sequences);
CreateGraph();
DanglingLinksThreshold = dangleThreshold;
DanglingLinksPurger = new DanglingLinksPurger(dangleThreshold);
RedundantPathLengthThreshold = redundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(redundantThreshold);
UnDangleGraph();
RemoveRedundancy();
IList <ISequence> contigs = BuildContigs().ToList();
CloneLibrary.Instance.AddLibrary("abc", 5, 20);
using (GraphScaffoldBuilder scaffold = new GraphScaffoldBuilder())
{
IEnumerable <ISequence> scaffoldSeq = scaffold.BuildScaffold(sequences, contigs, this.KmerLength, 3, 0);
HashSet <string> expected = new HashSet <string>
{
"ATGCCTCCTATCTTAGCGCGC", "CGCGCCGCGC", "TTTTTT", "CGCGCG", "TTTTAGC", "TTTTTA", "TTTAAA", "TTTTAA",
};
AlignmentHelpers.CompareSequenceLists(expected, scaffoldSeq);
}
}
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 BuildScaffold()
{
const int kmerLength = 6;
const int dangleThreshold = 3;
const int redundantThreshold = 7;
List <ISequence> sequences = new List <ISequence>(TestInputs.GetReadsForScaffolds());
KmerLength = kmerLength;
SequenceReads.Clear();
this.SetSequenceReads(sequences);
CreateGraph();
DanglingLinksThreshold = dangleThreshold;
DanglingLinksPurger = new DanglingLinksPurger(dangleThreshold);
RedundantPathLengthThreshold = redundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(redundantThreshold);
UnDangleGraph();
RemoveRedundancy();
IList <ISequence> contigs = BuildContigs().ToList();
Assert.AreEqual(contigs.Count, 10);
CloneLibrary.Instance.AddLibrary("abc", (float)5, (float)20);
GraphScaffoldBuilder scaffold = new GraphScaffoldBuilder();
IEnumerable <ISequence> scaffoldSeq = scaffold.BuildScaffold(
sequences, contigs, this.KmerLength, 3, 0);
Assert.AreEqual(scaffoldSeq.Count(), 8);
Assert.IsTrue(new string(scaffoldSeq.First().Select(a => (char)a).ToArray()).Equals(
"ATGCCTCCTATCTTAGCGCGC"));
scaffold.Dispose();
}
public void TestContigBuilder2()
{
const int KmerLength = 6;
const int RedundantThreshold = 10;
List <ISequence> readSeqs = TestInputs.GetRedundantPathReads();
SequenceReads.Clear();
this.SetSequenceReads(readSeqs);
this.KmerLength = KmerLength;
RedundantPathLengthThreshold = RedundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(RedundantThreshold);
ContigBuilder = new SimplePathContigBuilder();
CreateGraph();
RemoveRedundancy();
long graphCount = Graph.NodeCount;
long graphEdges = Graph.GetNodes().Select(n => n.ExtensionsCount).Sum();
IEnumerable <ISequence> contigs = BuildContigs();
long contigsBuiltGraphCount = Graph.NodeCount;
long contigsBuilt = Graph.GetNodes().Select(n => n.ExtensionsCount).Sum();
// Compare the two graphs
Assert.AreEqual(1, contigs.Count());
string s = new string(contigs.ElementAt(0).Select(a => (char)a).ToArray());
Assert.AreEqual("ATGCCTCCTATCTTAGCGATGCGGTGT", s);
Assert.AreEqual(graphCount, contigsBuiltGraphCount);
Assert.AreEqual(graphEdges, contigsBuilt);
}
private List <DeBruijnPathList> GetIndelPaths(DeBruijnGraph deBruijnGraph)
{
List <DeBruijnPathList> redundantPaths = new List <DeBruijnPathList>();
Parallel.ForEach(
deBruijnGraph.GetNodes(),
node =>
{
if (!node.ContainsSelfReference)
{
// Need to check for both left and right extensions for ambiguity.
if (node.RightExtensionNodesCount > 1)
{
TraceDivergingExtensionPaths(node, node.GetRightExtensionNodesWithOrientation(), true, redundantPaths);
}
if (node.LeftExtensionNodesCount > 1)
{
TraceDivergingExtensionPaths(node, node.GetLeftExtensionNodesWithOrientation(), false, redundantPaths);
}
}
}
);
RedundantPathsPurger.ValidatePathsAreFromSameDirection(redundantPaths, deBruijnGraph.KmerLength);
var indelPaths = redundantPaths.Where(x => x.Paths.Select(z => z.PathNodes.Count).Distinct().Count() != 1).ToList();
//TODO: Could merge the two filters here
indelPaths = RemoveDuplicates(indelPaths);
indelPaths = RemoveEmbeddedPaths(indelPaths);
return(indelPaths);
}
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 TestContigBuilder1()
{
const int KmerLength = 11;
const int DangleThreshold = 3;
const int RedundantThreshold = 10;
List <ISequence> readSeqs = TestInputs.GetDanglingReads();
this.SequenceReads.Clear();
this.SetSequenceReads(readSeqs);
this.KmerLength = KmerLength;
DanglingLinksThreshold = DangleThreshold;
DanglingLinksPurger = new DanglingLinksPurger(DangleThreshold);
RedundantPathLengthThreshold = RedundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(RedundantThreshold);
ContigBuilder = new SimplePathContigBuilder();
CreateGraph();
UnDangleGraph();
RemoveRedundancy();
long graphCount = Graph.NodeCount;
long graphEdges = Graph.GetNodes().Select(n => n.ExtensionsCount).Sum();
IEnumerable <ISequence> contigs = BuildContigs();
long contigsBuiltGraphCount = this.Graph.NodeCount;
long contigsBuilt = Graph.GetNodes().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)
{
string s = new string(contig.Select(a => (char)a).ToArray());
Assert.IsTrue(expectedContigs.Contains(s));
}
Assert.AreEqual(graphCount, contigsBuiltGraphCount);
Assert.AreEqual(graphEdges, contigsBuilt);
}
public void TracePathTestWithPalindromicContig()
{
const int kmerLength = 6;
const int dangleThreshold = 3;
const int redundantThreshold = 7;
List <ISequence> sequences = new List <ISequence>();
Sequence seq = new Sequence(Alphabets.DNA, "ATGCCTC");
seq.DisplayID = ">10.x1:abc";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "CCTCCTAT");
seq.DisplayID = "1";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "TCCTATC");
seq.DisplayID = "2";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "TGCCTCCT");
seq.DisplayID = "3";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "ATCTTAGC");
seq.DisplayID = "4";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "CTATCTTAG");
seq.DisplayID = "5";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "CTTAGCG");
seq.DisplayID = "6";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "GCCTCCTAT");
seq.DisplayID = ">8.x1:abc";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "TAGCGCGCTA");
seq.DisplayID = ">8.y1:abc";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "AGCGCGC");
seq.DisplayID = ">9.x1:abc";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "TTTTTT");
seq.DisplayID = "7";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "TTTTTAAA");
seq.DisplayID = "8";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "TAAAAA");
seq.DisplayID = "9";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "TTTTAG");
seq.DisplayID = "10";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "TTTAGC");
seq.DisplayID = "11";
sequences.Add(seq);
seq = new Sequence(Alphabets.DNA, "GCGCGCCGCGCG");
seq.DisplayID = "12";
sequences.Add(seq);
KmerLength = kmerLength;
SequenceReads.Clear();
AddSequenceReads(sequences);
CreateGraph();
DanglingLinksThreshold = dangleThreshold;
DanglingLinksPurger = new DanglingLinksPurger(dangleThreshold);
RedundantPathLengthThreshold = redundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(redundantThreshold);
UnDangleGraph();
RemoveRedundancy();
IList <ISequence> contigs = BuildContigs();
ReadContigMapper mapper = new ReadContigMapper();
ReadContigMap maps = mapper.Map(contigs, sequences, kmerLength);
MatePairMapper builder = new MatePairMapper();
CloneLibrary.Instance.AddLibrary("abc", (float)5, (float)15);
ContigMatePairs pairedReads = builder.MapContigToMatePairs(sequences, maps);
ContigMatePairs overlap;
OrientationBasedMatePairFilter filter = new OrientationBasedMatePairFilter();
overlap = filter.FilterPairedReads(pairedReads, 0);
DistanceCalculator dist = new DistanceCalculator();
dist.CalculateDistance(overlap);
Graph.BuildContigGraph(contigs, this.KmerLength);
TracePath path = new TracePath();
IList <ScaffoldPath> paths = path.FindPaths(Graph, overlap, kmerLength, 3);
Assert.AreEqual(paths.Count, 3);
Assert.AreEqual(paths.First().Count, 3);
ScaffoldPath scaffold = paths.First();
DeBruijnGraph graph = Graph;
Assert.IsTrue(graph.GetNodeSequence(scaffold[0].Key).ToString().Equals("ATGCCTCCTATCTTAGC"));
Assert.IsTrue(graph.GetNodeSequence(scaffold[1].Key).ToString().Equals("TTAGCGCG"));
Assert.IsTrue(graph.GetNodeSequence(scaffold[2].Key).ToString().Equals("GCGCGC"));
}
public void TracePathTestWithPalindromicContig()
{
const int kmerLengthConst = 5;
const int dangleThreshold = 3;
const int redundantThreshold = 6;
var sequences = new List <ISequence>()
{
new Sequence(Alphabets.DNA, "ATGCCTC")
{
ID = "0"
},
new Sequence(Alphabets.DNA, "CCTCCTAT")
{
ID = "1"
},
new Sequence(Alphabets.DNA, "TCCTATC")
{
ID = "2"
},
new Sequence(Alphabets.DNA, "TGCCTCCT")
{
ID = "3"
},
new Sequence(Alphabets.DNA, "ATCTTAGC")
{
ID = "4"
},
new Sequence(Alphabets.DNA, "CTATCTTAG")
{
ID = "5"
},
new Sequence(Alphabets.DNA, "CTTAGCG")
{
ID = "6"
},
new Sequence(Alphabets.DNA, "GCCTCCTAT")
{
ID = "7"
},
new Sequence(Alphabets.DNA, "TAGCGCGCTA")
{
ID = "8"
},
new Sequence(Alphabets.DNA, "AGCGCGC")
{
ID = "9"
},
new Sequence(Alphabets.DNA, "TTTTTT")
{
ID = "10"
},
new Sequence(Alphabets.DNA, "TTTTTAAA")
{
ID = "11"
},
new Sequence(Alphabets.DNA, "TAAAAA")
{
ID = "12"
},
new Sequence(Alphabets.DNA, "TTTTAG")
{
ID = "13"
},
new Sequence(Alphabets.DNA, "TTTAGC")
{
ID = "14"
},
new Sequence(Alphabets.DNA, "GCGCGCCGCGCG")
{
ID = "15"
},
};
KmerLength = kmerLengthConst;
SequenceReads.Clear();
SetSequenceReads(sequences);
CreateGraph();
DanglingLinksThreshold = dangleThreshold;
DanglingLinksPurger = new DanglingLinksPurger(dangleThreshold);
RedundantPathLengthThreshold = redundantThreshold;
RedundantPathsPurger = new RedundantPathsPurger(redundantThreshold);
UnDangleGraph();
RemoveRedundancy();
IList <ISequence> contigs = BuildContigs().ToList();
ReadContigMapper mapper = new ReadContigMapper();
ReadContigMap maps = mapper.Map(contigs, sequences, kmerLengthConst);
MatePairMapper builder = new MatePairMapper();
CloneLibrary.Instance.AddLibrary("abc", 5, 15);
ContigMatePairs pairedReads = builder.MapContigToMatePairs(sequences, maps);
OrientationBasedMatePairFilter filter = new OrientationBasedMatePairFilter();
ContigMatePairs overlap = filter.FilterPairedReads(pairedReads, 0);
DistanceCalculator dist = new DistanceCalculator(overlap);
overlap = dist.CalculateDistance();
ContigGraph graph = new ContigGraph();
graph.BuildContigGraph(contigs, this.KmerLength);
TracePath path = new TracePath();
IList <ScaffoldPath> paths = path.FindPaths(graph, overlap, kmerLengthConst, 3);
Assert.AreEqual(paths.Count, 3);
Assert.AreEqual(paths.First().Count, 3);
ScaffoldPath scaffold = paths.First();
Assert.AreEqual("ATGCCTCCTATCTTAGC", graph.GetNodeSequence(scaffold[0].Key).ConvertToString());
Assert.AreEqual("TTAGCGCG", graph.GetNodeSequence(scaffold[1].Key).ConvertToString());
Assert.AreEqual("GCGCGC", graph.GetNodeSequence(scaffold[2].Key).ConvertToString());
}