public long GetEdgesCount(SequenceModel sequenceData)
{
var sequence = _sequenceProvider.Provide(sequenceData?.FileName, sequenceData?.Content)?.First();
_suffixTree = new MultiWaySuffixTree(sequence);
return(_suffixTree.EdgesCount);
}
public IEnumerable <Match> GetUniqueMatches(SequenceModel sequenceData)
{
var sequence = _sequenceProvider.Provide(sequenceData?.FileName, sequenceData?.Content)?.First();
_suffixTree = new MultiWaySuffixTree(sequence);
return(_suffixTree.SearchMatchesUniqueInReference(sequence));
}
void ValidateFindMatchSuffixGeneralTestCases(string nodeName, bool isFilePath,
AdditionalParameters additionalParam)
{
ISequence referenceSeqs = null;
string[] referenceSequences = null;
string[] searchSequences = null;
List <ISequence> searchSeqList = new List <ISequence>();
if (isFilePath)
{
// Gets the reference sequence from the FastA file
string filePath = utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.FilePathNode);
Assert.IsNotNull(filePath);
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"NUCmer BVT : Successfully validated the File Path '{0}'.", filePath));
using (FastAParser parser = new FastAParser(filePath))
{
IEnumerable <ISequence> referenceSeqList = parser.Parse();
List <Byte> byteList = new List <Byte>();
foreach (ISequence seq in referenceSeqList)
{
byteList.AddRange(seq);
byteList.Add((byte)'+');
}
referenceSeqs = new Sequence(AlphabetExtensions.GetMummerAlphabet(referenceSeqList.ElementAt(0).Alphabet),
byteList.ToArray());
// Gets the query sequence from the FastA file
string queryFilePath = utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.SearchSequenceFilePathNode);
Assert.IsNotNull(queryFilePath);
ApplicationLog.WriteLine(string.Format((IFormatProvider)null,
"NUCmer BVT : Successfully validated the File Path '{0}'.", queryFilePath));
FastAParser queryParser = new FastAParser(queryFilePath);
IEnumerable <ISequence> querySeqList = queryParser.Parse();
foreach (ISequence seq in querySeqList)
{
searchSeqList.Add(seq);
}
}
}
else
{
// Gets the reference & search sequences from the configurtion file
referenceSequences = utilityObj.xmlUtil.GetTextValues(nodeName,
Constants.ReferenceSequencesNode);
searchSequences = utilityObj.xmlUtil.GetTextValues(nodeName,
Constants.SearchSequencesNode);
IAlphabet seqAlphabet = Utility.GetAlphabet(utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.AlphabetNameNode));
List <ISequence> refSeqList = new List <ISequence>();
for (int i = 0; i < referenceSequences.Length; i++)
{
ISequence referSeq = new Sequence(seqAlphabet, encodingObj.GetBytes(referenceSequences[i]));
refSeqList.Add(referSeq);
}
List <Byte> byteListQuery = new List <Byte>();
foreach (ISequence seq in refSeqList)
{
byteListQuery.AddRange(seq);
byteListQuery.Add((byte)'+');
}
referenceSeqs = new Sequence(AlphabetExtensions.GetMummerAlphabet(refSeqList.ElementAt(0).Alphabet),
byteListQuery.ToArray());
for (int i = 0; i < searchSequences.Length; i++)
{
ISequence searchSeq = new Sequence(seqAlphabet, encodingObj.GetBytes(searchSequences[i]));
searchSeqList.Add(searchSeq);
}
}
string mumLength = utilityObj.xmlUtil.GetTextValue(nodeName, Constants.MUMLengthNode);
// Builds the suffix for the reference sequence passed.
MultiWaySuffixTree suffixTreeBuilder = new MultiWaySuffixTree(referenceSeqs as Sequence);
suffixTreeBuilder.MinLengthOfMatch = long.Parse(mumLength, null);
Dictionary <ISequence, IEnumerable <Match> > matches =
new Dictionary <ISequence, IEnumerable <Match> >();
for (int i = 0; i < searchSeqList.Count; i++)
{
matches.Add(searchSeqList[i],
suffixTreeBuilder.SearchMatchesUniqueInReference(searchSeqList[i]));
}
List <Match> mums = new List <Match>();
foreach (var a in matches.Values)
{
mums.AddRange(a);
}
switch (additionalParam)
{
case AdditionalParameters.FindUniqueMatches:
// Validates the Unique Matches.
ApplicationLog.WriteLine("NUCmer BVT : Validating the Unique Matches");
Assert.IsTrue(ValidateUniqueMatches(mums, nodeName, additionalParam, isFilePath));
Console.WriteLine(
"NUCmer BVT : Successfully validated the all the unique matches for the sequences.");
break;
case AdditionalParameters.PerformClusterBuilder:
// Validates the Unique Matches.
ApplicationLog.WriteLine(
"NUCmer BVT : Validating the Unique Matches using Cluster Builder");
Assert.IsTrue(ValidateUniqueMatches(mums, nodeName, additionalParam, isFilePath));
Console.WriteLine(
"NUCmer BVT : Successfully validated the all the cluster builder matches for the sequences.");
break;
default:
break;
}
ApplicationLog.WriteLine(
"NUCmer BVT : Successfully validated the all the unique matches for the sequences.");
}
static void Main(string[] args)
{
try
{
// DateTime dStart = DateTime.Now;
Stopwatch swMumUtil = Stopwatch.StartNew();
Stopwatch swInterval = new Stopwatch();
Console.Error.WriteLine(SplashString());
if (args.Length > 0)
{
CommandLineOptions myArgs = ProcessCommandLine(args);
if (myArgs.help)
{
Console.WriteLine(Resources.MumUtilHelp);
}
else
{
FileInfo refFileinfo = new FileInfo(myArgs.fileList[0]);
long refFileLength = refFileinfo.Length;
refFileinfo = null;
swInterval.Restart();
IEnumerable <ISequence> referenceSequences = ParseFastA(myArgs.fileList[0]);
Sequence referenceSequence = referenceSequences.First() as Sequence;
swInterval.Stop();
if (myArgs.verbose)
{
Console.Error.WriteLine();
Console.Error.WriteLine(" Processed Reference FastA file: {0}", Path.GetFullPath(myArgs.fileList[0]));
Console.Error.WriteLine(" Length of first Sequence: {0:#,000}", referenceSequence.Count);
Console.Error.WriteLine(" Read/Processing time: {0}", swInterval.Elapsed);
Console.Error.WriteLine(" File size : {0:#,000} bytes", refFileLength);
}
FileInfo queryFileinfo = new FileInfo(myArgs.fileList[1]);
long queryFileLength = queryFileinfo.Length;
refFileinfo = null;
IEnumerable <ISequence> parsedQuerySequences = ParseFastA(myArgs.fileList[1]);
IEnumerable <ISequence> querySequences = parsedQuerySequences;
if (myArgs.reverseOnly)
{
// convert to reverse complement sequences
querySequences = ReverseComplementSequenceList(parsedQuerySequences);
}
else if (myArgs.both)
{
// add the reverse complement sequences along with query sequences.
querySequences = AddReverseComplementsToSequenceList(parsedQuerySequences);
}
// DISCUSSION:
// Three possible outputs desired. Globally unique 'mum' (v1), unique in reference sequence (v2),
// or get the maximum matches of length or greater.
//
mummerTime = new TimeSpan();
writetime = new TimeSpan();
IEnumerable <Match> mums;
long memoryAtStart = 0;
long memoryAtEnd = 0;
if (myArgs.verbose)
{
swMumUtil.Stop();
memoryAtStart = GC.GetTotalMemory(true);
swMumUtil.Start();
}
swInterval.Restart();
MultiWaySuffixTree suffixTreee = new MultiWaySuffixTree(referenceSequence);
swInterval.Stop();
if (myArgs.verbose)
{
swMumUtil.Stop();
memoryAtEnd = GC.GetTotalMemory(true);
swMumUtil.Start();
}
MUMmer mummer = new MUMmer(suffixTreee);
if (myArgs.verbose)
{
Console.Error.WriteLine();
Console.Error.WriteLine("Suffix tree construction time : {0}", swInterval.Elapsed);
Console.Error.WriteLine("Memory consumed by Suffix tree : {0:#,000}", memoryAtEnd - memoryAtStart);
Console.Error.WriteLine("Total edges created : {0:#,000}", suffixTreee.EdgesCount);
Console.Error.WriteLine("Memory per edge : {0:#,000.00} bytes", (((double)(memoryAtEnd - memoryAtStart)) / suffixTreee.EdgesCount));
Console.Error.WriteLine();
Console.Error.WriteLine(" Processed Query FastA file: {0}", Path.GetFullPath(myArgs.fileList[1]));
Console.Error.WriteLine(" File Size : {0:#,000} bytes", queryFileLength);
}
mummer.LengthOfMUM = myArgs.length;
mummer.NoAmbiguity = myArgs.noAmbiguity;
long querySeqCount = 0;
double sumofSeqLength = 0;
TimeSpan totalTimetakenToProcessQuerySequences = new TimeSpan();
string outputOption = string.Empty;
if (myArgs.maxmatch)
{
outputOption = "GetMumsMaxMatch()";
swInterval.Restart();
foreach (Sequence qSeq in querySequences)
{
// Stop the wath after each query sequence parsed.
swInterval.Stop();
// Add total time to process query sequence.
// if reverse complement option is set, includes reverse complement time also.
totalTimetakenToProcessQuerySequences = totalTimetakenToProcessQuerySequences.Add(swInterval.Elapsed);
mums = mummer.GetMatches(qSeq);
WriteMums(mums, referenceSequence, qSeq, myArgs);
querySeqCount++;
sumofSeqLength += qSeq.Count;
// Start the watch for next query sequence parse.
swInterval.Restart();
}
swInterval.Stop();
}
else if (myArgs.mum)
{
// mums = mum3.GetMumsMum( referenceSequences[0], querySequences);
outputOption = "GetMumsMum()";
swInterval.Restart();
foreach (Sequence qSeq in querySequences)
{
// Stop the wath after each query sequence parsed.
swInterval.Stop();
// Add total time to process query sequence.
// if reverse complement option is set, includes reverse complement time also.
totalTimetakenToProcessQuerySequences = totalTimetakenToProcessQuerySequences.Add(swInterval.Elapsed);
swInterval.Restart();
// TODO: After implementing GetMatchesUniqueInBothReferenceAndQuery() in MUMmer
//// GetMatchesUniqueInReference() with GetMatchesUniqueInBothReferenceAndQuery() in the line below.
mums = mummer.GetMatchesUniqueInReference(qSeq);
swInterval.Stop();
// Add time taken by GetMatchesUniqueInBothReferenceAndQuery().
mummerTime = mummerTime.Add(swInterval.Elapsed);
swInterval.Restart();
WriteMums(mums, referenceSequence, qSeq, myArgs);
swInterval.Stop();
// Add time taken by write matches.
writetime = writetime.Add(swInterval.Elapsed);
querySeqCount++;
sumofSeqLength += qSeq.Count;
// Start the watch for next query sequence parse.
swInterval.Restart();
}
swInterval.Stop();
}
else if (myArgs.mumreference)
{
// NOTE:
// mum3.GetMUMs() this really implements the GetMumReference() functionality
// mums = mum3.GetMumsReference( referenceSequences[0], querySequences); // should be
//swInterval.Restart();
outputOption = "GetMumsReference()";
swInterval.Restart();
foreach (Sequence qSeq in querySequences)
{
// Stop the wath after each query sequence parsed.
swInterval.Stop();
// Add total time to process query sequence.
// if reverse complement option is set, includes reverse complement time also.
totalTimetakenToProcessQuerySequences = totalTimetakenToProcessQuerySequences.Add(swInterval.Elapsed);
swInterval.Restart();
mums = mummer.GetMatchesUniqueInReference(qSeq);
swInterval.Stop();
// Add time taken by GetMatchesUniqueInReference().
mummerTime = mummerTime.Add(swInterval.Elapsed);
swInterval.Restart();
WriteMums(mums, referenceSequence, qSeq, myArgs);
swInterval.Stop();
// Add time taken by write matches.
writetime = writetime.Add(swInterval.Elapsed);
querySeqCount++;
sumofSeqLength += qSeq.Count;
// Start the watch for next query sequence parse.
swInterval.Restart();
}
swInterval.Stop();
}
else
{
// cannot happen as argument processing already asserted one of the three options must be specified
Console.Error.WriteLine("\nError: one of /maxmatch, /mum, /mumreference options must be specified.");
Environment.Exit(-1);
// kill the error about unitialized use of 'mums' in the next block...the compiler does not recognize
// Environment.Exit() as a no-return function
throw new Exception("Never hit this");
}
if (myArgs.verbose)
{
if (myArgs.reverseOnly || myArgs.both)
{
Console.Error.WriteLine(" Read/Processing time : {0}", timeTakenToParseQuerySequences);
Console.Error.WriteLine(" Reverse Complement time : {0}", timeTakenToGetReverseComplement);
Console.Error.WriteLine(" Total time taken to Process reads: {0}", totalTimetakenToProcessQuerySequences);
}
else
{
Console.Error.WriteLine(" Read/Processing time : {0}", totalTimetakenToProcessQuerySequences);
}
Console.Error.WriteLine();
Console.Error.WriteLine(" Number of query Sequences : {0:#,000}", querySeqCount);
Console.Error.WriteLine(" Average length of query Sequences: {0:#,000}", sumofSeqLength / querySeqCount);
Console.Error.WriteLine();
Console.Error.WriteLine("Compute {0,20} time : {1}", outputOption, mummerTime);
Console.Error.WriteLine(" WriteMums() time : {0}", writetime);
}
swMumUtil.Stop();
if (myArgs.verbose)
{
Console.Error.WriteLine(" Total MumUtil Runtime : {0}", swMumUtil.Elapsed);
}
}
}
else
{
Console.WriteLine(Resources.MumUtilHelp);
}
}
catch (Exception ex)
{
DisplayException(ex);
}
}
/// <summary>
/// Validates most of the find matches suffix tree test cases with varying parameters.
/// </summary>
/// <param name="nodeName">Node name which needs to be read for execution.</param>
/// <param name="isFilePath">Is File Path?</param>
/// <param name="additionalParam">LIS action type enum</param>
void ValidateFindMatchSuffixGeneralTestCases(string nodeName, bool isFilePath,
AdditionalParameters additionalParam)
{
ISequence referenceSeqs;
var searchSeqList = new List <ISequence>();
if (isFilePath)
{
// Gets the reference sequence from the FastA file
string filePath = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.FilePathNode);
Assert.IsNotNull(filePath);
FastAParser parser = new FastAParser();
IEnumerable <ISequence> referenceSeqList = parser.Parse(filePath);
List <Byte> byteList = new List <Byte>();
foreach (ISequence seq in referenceSeqList)
{
byteList.AddRange(seq);
byteList.Add((byte)'+');
}
referenceSeqs = new Sequence(referenceSeqList.First().Alphabet.GetMummerAlphabet(), byteList.ToArray());
// Gets the query sequence from the FastA file
string queryFilePath = this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.SearchSequenceFilePathNode);
Assert.IsNotNull(queryFilePath);
IEnumerable <ISequence> querySeqList = parser.Parse(queryFilePath);
searchSeqList.AddRange(querySeqList);
}
else
{
// Gets the reference & search sequences from the configuration file
string[] referenceSequences = this.utilityObj.xmlUtil.GetTextValues(nodeName,
Constants.ReferenceSequencesNode);
string[] searchSequences = this.utilityObj.xmlUtil.GetTextValues(nodeName,
Constants.SearchSequencesNode);
IAlphabet seqAlphabet = Utility.GetAlphabet(this.utilityObj.xmlUtil.GetTextValue(nodeName,
Constants.AlphabetNameNode));
List <ISequence> refSeqList = referenceSequences.Select(t => new Sequence(seqAlphabet, this.encodingObj.GetBytes(t))).Cast <ISequence>().ToList();
List <Byte> byteListQuery = new List <Byte>();
foreach (ISequence seq in refSeqList)
{
byteListQuery.AddRange(seq);
byteListQuery.Add((byte)'+');
}
referenceSeqs = new Sequence(refSeqList.First().Alphabet.GetMummerAlphabet(),
byteListQuery.ToArray());
searchSeqList.AddRange(searchSequences.Select(t => new Sequence(seqAlphabet, this.encodingObj.GetBytes(t))).Cast <ISequence>());
}
string mumLength = this.utilityObj.xmlUtil.GetTextValue(nodeName, Constants.MUMLengthNode);
// Builds the suffix for the reference sequence passed.
MultiWaySuffixTree suffixTreeBuilder = new MultiWaySuffixTree(referenceSeqs as Sequence)
{
MinLengthOfMatch = long.Parse(mumLength, null)
};
var matches = new Dictionary <ISequence, IEnumerable <Match> >();
foreach (ISequence sequence in searchSeqList)
{
matches.Add(sequence,
suffixTreeBuilder.SearchMatchesUniqueInReference(sequence));
}
List <Match> mums = new List <Match>();
foreach (var a in matches.Values)
{
mums.AddRange(a);
}
switch (additionalParam)
{
case AdditionalParameters.FindUniqueMatches:
// Validates the Unique Matches.
Assert.IsTrue(this.ValidateUniqueMatches(mums, nodeName, additionalParam, isFilePath));
break;
case AdditionalParameters.PerformClusterBuilder:
// Validates the Unique Matches.
Assert.IsTrue(this.ValidateUniqueMatches(mums, nodeName, additionalParam, isFilePath));
break;
default:
break;
}
}
