/// <summary>
/// Step 4: Remove redundant paths from graph.
/// </summary>
protected void RemoveRedundancy()
{
if (RedundantPathsPurger != null)
{
DeBruijnPathList redundantNodes;
do
{
redundantNodes = RedundantPathsPurger.DetectErroneousNodes(Graph);
RedundantPathsPurger.RemoveErroneousNodes(Graph, redundantNodes);
}while (redundantNodes.Paths.Count > 0);
}
}
/// <summary>
/// Initializes the above defined fields. For each step in assembly
/// we use a separate class for implementation. This method assigns
/// these variables to classes with desired implementation.
/// </summary>
protected void InitializeDefaultGraphModifiers()
{
// Assign uninitialized fields to default values
if (DanglingLinksPurger == null)
{
DanglingLinksPurger = new DanglingLinksPurger();
}
if (RedundantPathsPurger == null)
{
RedundantPathsPurger = new RedundantPathsPurger(RedundantPathLengthThreshold);
}
if (LowCoverageContigPurger == null)
{
LowCoverageContigPurger = new SimplePathContigBuilder();
}
}
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());
}
/// <summary>
/// Creates RedundantPathPurger instance by passing pathlength and count. Detect
/// redundant error nodes and remove these nodes from the graph. Validate the graph.
/// </summary>
/// <param name="nodeName">xml node name used for different testcases</param>
/// <param name="isMicroOrganism">Is micro organism</param>
internal void ValidateRedundantPathPurgerCtor(string nodeName, bool isMicroOrganism)
{
string filePath = utilityObj.xmlUtil.GetTextValue(nodeName, Constants.FilePathNode);
string kmerLength = utilityObj.xmlUtil.GetTextValue(nodeName, Constants.KmerLengthNode);
string expectedNodesCount = utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.ExpectedNodesCountAfterDangling);
// Get the input reads and build kmers
IEnumerable<ISequence> sequenceReads = null;
using (FastAParser parser = new FastAParser(filePath))
{
sequenceReads = parser.Parse();
// Build kmers from step1,graph in step2
// Remove the dangling links from graph in step3
// Validate the graph
this.KmerLength = int.Parse(kmerLength, (IFormatProvider)null);
this.SequenceReads.Clear();
this.SetSequenceReads(sequenceReads.ToList());
this.CreateGraph();
DeBruijnGraph graph = this.Graph;
this.DanglingLinksPurger = new DanglingLinksPurger(this.KmerLength);
this.UnDangleGraph();
// Create RedundantPathPurger instance, detect redundant nodes and remove error nodes
RedundantPathsPurger redundantPathPurger =
new RedundantPathsPurger(int.Parse(kmerLength, (IFormatProvider)null) + 1);
DeBruijnPathList redundantnodelist = redundantPathPurger.DetectErroneousNodes(graph);
redundantPathPurger.RemoveErroneousNodes(graph, redundantnodelist);
if (isMicroOrganism)
Assert.AreEqual(expectedNodesCount, graph.GetNodes().Count());
else
ValidateGraph(graph, nodeName);
}
ApplicationLog.WriteLine(@"Padena P1 :RedundantPathsPurger ctor and methods validation for Padena step4 completed successfully");
}
/// <summary>
/// Initializes the above defined fields. For each step in assembly
/// we use a separate class for implementation. This method assigns
/// these variables to classes with desired implementation.
/// </summary>
protected void InitializeDefaultGraphModifiers()
{
// Assign uninitialized fields to default values
if (DanglingLinksPurger == null)
{
DanglingLinksPurger = new DanglingLinksPurger();
}
if (RedundantPathsPurger == null)
{
RedundantPathsPurger = new RedundantPathsPurger(RedundantPathLengthThreshold);
}
if (LowCoverageContigPurger == null)
{
LowCoverageContigPurger = new SimplePathContigBuilder();
}
}
/// <summary>
/// Creates RedundantPathPurger instance by passing pathlength and count. Detect
/// redundant error nodes and remove these nodes from the graph. Validate the graph.
/// </summary>
/// <param name="nodeName">xml node name used for different testcases</param>
internal void ValidateRedundantPathPurgerCtor(string nodeName)
{
string filePath = utilityObj.xmlUtil.GetTextValue(nodeName, Constants.FilePathNode);
string kmerLength = utilityObj.xmlUtil.GetTextValue(nodeName, Constants.KmerLengthNode);
// Get the input reads and build kmers
IEnumerable<ISequence> sequenceReads = null;
FastAParser parser = new FastAParser();
parser.Open( filePath.Replace("\\", System.IO.Path.DirectorySeparatorChar.ToString()));
sequenceReads = parser.Parse().ToList();
parser.Close();
// Build kmers from step1,graph in step2
// Remove the dangling links from graph in step3
// Validate the graph
this.KmerLength = int.Parse(kmerLength, (IFormatProvider)null);
this.SequenceReads.Clear();
this.SetSequenceReads(sequenceReads.ToList());
this.CreateGraph();
this.UnDangleGraph();
// Create RedundantPathPurger instance, detect redundant nodes and remove error nodes
RedundantPathsPurger redundantPathPurger =
new RedundantPathsPurger(int.Parse(kmerLength, (IFormatProvider)null) + 1);
DeBruijnPathList redundantnodelist =
redundantPathPurger.DetectErroneousNodes(this.Graph);
redundantPathPurger.RemoveErroneousNodes(this.Graph, redundantnodelist);
ValidateGraph(this.Graph, nodeName);
ApplicationLog.WriteLine(
@"Padena BVT :RedundantPathsPurger ctor and methods validation for
Padena step4 completed successfully");
}
