public void TestFormatter()
{
string filePath = @"TestUtils\BAM\SeqAlignment.bam".TestDir();
const string outputfilePath = "BAMTests123.bam";
string outputIndexFile = outputfilePath + ".bai";
BAMParser parser = new BAMParser();
SequenceAlignmentMap alignmentMap = parser.ParseOne <SequenceAlignmentMap>(filePath);
Assert.IsNotNull(alignmentMap);
Assert.IsNotNull(alignmentMap.Header);
Assert.AreEqual(alignmentMap.Header.GetReferenceSequencesInfoFromSQHeader().Count, 1);
Assert.AreEqual(alignmentMap.Header.ReferenceSequences.Count, 1);
Assert.IsNotNull(alignmentMap.QuerySequences);
Assert.AreEqual(alignmentMap.QuerySequences.Count, 2);
BAMFormatter formatter = new BAMFormatter();
formatter.Format(alignmentMap, outputfilePath);
formatter.CreateSortedBAMFile = true;
formatter.CreateIndexFile = true;
formatter.Format(alignmentMap, outputfilePath);
Assert.IsTrue(File.Exists("BAMTests123.bam.bai"));
alignmentMap = parser.ParseOne <SequenceAlignmentMap>(outputfilePath);
Assert.IsNotNull(alignmentMap);
Assert.IsNotNull(alignmentMap.Header);
Assert.AreEqual(alignmentMap.Header.GetReferenceSequencesInfoFromSQHeader().Count, 1);
Assert.AreEqual(alignmentMap.Header.ReferenceSequences.Count, 1);
Assert.AreEqual(alignmentMap.QuerySequences.Count, 2);
alignmentMap = parser.ParseRange("BAMTests123.bam", new SequenceRange("chr20", 0, int.MaxValue));
Assert.IsNotNull(alignmentMap);
Assert.IsNotNull(alignmentMap.Header);
Assert.AreEqual(alignmentMap.Header.GetReferenceSequencesInfoFromSQHeader().Count, 1);
Assert.AreEqual(alignmentMap.Header.ReferenceSequences.Count, 1);
Assert.AreEqual(alignmentMap.QuerySequences.Count, 2);
alignmentMap = parser.ParseRange("BAMTests123.bam", new SequenceRange("chr20", 0, 28833));
Assert.IsNotNull(alignmentMap);
Assert.IsNotNull(alignmentMap.Header);
Assert.AreEqual(alignmentMap.Header.GetReferenceSequencesInfoFromSQHeader().Count, 1);
Assert.AreEqual(alignmentMap.Header.ReferenceSequences.Count, 1);
Assert.AreEqual(alignmentMap.QuerySequences.Count, 1);
File.Delete(outputfilePath);
File.Delete(outputIndexFile);
}
public void TestFormatterWithSort()
{
string inputFilePath = @"TestUtils\BAM\SeqAlignment.bam".TestDir();
string outputFilePath1 = "output1.bam";
string outputFilePath2 = "output2.bam";
BAMParser parser = null;
try
{
parser = new BAMParser();
BAMFormatter formatter = new BAMFormatter();
SequenceAlignmentMap alignmentMap = parser.ParseOne <SequenceAlignmentMap>(inputFilePath);
Assert.IsTrue(alignmentMap != null);
Assert.AreEqual(alignmentMap.Header.GetReferenceSequencesInfoFromSQHeader().Count, 1);
Assert.AreEqual(alignmentMap.Header.ReferenceSequences.Count, 1);
Assert.AreEqual(alignmentMap.QuerySequences.Count, 2);
formatter.CreateSortedBAMFile = true;
formatter.SortType = BAMSortByFields.ChromosomeCoordinates;
formatter.Format(alignmentMap, outputFilePath1);
alignmentMap = parser.ParseOne <SequenceAlignmentMap>(inputFilePath);
formatter.Format(alignmentMap, outputFilePath2);
Assert.IsTrue(File.Exists(outputFilePath1));
Assert.IsTrue(File.Exists(outputFilePath2));
Assert.AreEqual(true, FileCompare(outputFilePath1, outputFilePath2));
}
finally
{
if (parser != null)
{
parser.Dispose();
}
File.Delete(outputFilePath1);
File.Delete(outputFilePath2);
}
}
/// <summary>
/// Get Chimera data
/// </summary>
/// <param name="filename">Path of the BAM file</param>
/// <param name="mean">Mean value</param>
/// <param name="deviation">Standard deviation</param>
/// <returns></returns>
private Matrix <string, string, string> GetChimeraData(string filename)
{
SequenceAlignmentMap alignmentMapobj = null;
if (!SAMInput)
{
BAMParser bamParser = new BAMParser();
alignmentMapobj = bamParser.ParseOne <SequenceAlignmentMap>(filename);
}
else
{
SAMParser samParser = new SAMParser();
alignmentMapobj = samParser.ParseOne <SequenceAlignmentMap>(filename);
}
// get reads from sequence alignment map object.
IList <PairedRead> pairedReads = null;
pairedReads = alignmentMapobj.GetPairedReads(200, 50);
// select chimeras from reads.
var chimeras = pairedReads.Where(PE => PE.PairedType == PairedReadType.Chimera);
// Group chimeras based on first reads chromosomes name.
var groupedChimeras =
chimeras.GroupBy(PR => PR.Read1.RName);
IList <string> chrs = alignmentMapobj.GetRefSequences();
// Declare sparse matrix to store statistics.
SparseMatrix <string, string, string> statistics =
SparseMatrix <string, string, string> .CreateEmptyInstance(
chrs, chrs, "0");
// For each group create sub group depending on the second reads chromosomes.
foreach (var group in groupedChimeras)
{
foreach (var subgroup in group.GroupBy(PE => PE.Read2.RName))
{
// store the count to stats
statistics[group.Key, subgroup.Key] = subgroup.Count().ToString();
}
}
return(statistics);
}
/// <summary>
/// Public method to sort BAM file.
/// SAMUtil.exe in.bam out.bam
/// </summary>
public void DoSort()
{
string sortExtension = ".sort";
if (string.IsNullOrEmpty(InputFilename))
{
throw new InvalidOperationException(Resources.SortHelp);
}
BAMParser parse = new BAMParser();
SequenceAlignmentMap map = null;
try
{
map = parse.ParseOne <SequenceAlignmentMap>(InputFilename);
}
catch (Exception ex)
{
throw new InvalidOperationException(Resources.InvalidBAMFile, ex);
}
BAMFormatter format = new BAMFormatter
{
CreateSortedBAMFile = true,
SortType =
this.SortByReadName
? BAMSortByFields.ReadNames
: BAMSortByFields.ChromosomeCoordinates
};
if (string.IsNullOrEmpty(OutputFilename))
{
OutputFilename = InputFilename + sortExtension;
autoGeneratedOutputFilename = true;
}
format.Format(map, OutputFilename);
if (autoGeneratedOutputFilename)
{
Console.WriteLine(Resources.SuccessMessageWithOutputFileName, OutputFilename);
}
}
/// <summary>
/// Display Sequence Item occurences percentage
/// </summary>
/// <param name="inputFile">Path of the input file</param>
/// <param name="possibleOccurence">True to display Nculeaotide distribution</param>
public void DisplaySequenceItemOccurences(string inputFile,
bool possibleOccurence)
{
if (string.IsNullOrEmpty(inputFile))
{
throw new InvalidOperationException("Input File Not specified");
}
SequenceAlignmentMap alignmentMapobj = null;
if (!SAMInput)
{
BAMParser bamParser = new BAMParser();
alignmentMapobj = bamParser.ParseOne <SequenceAlignmentMap>(inputFile);
}
else
{
SAMParser samParser = new SAMParser();
alignmentMapobj = samParser.ParseOne <SequenceAlignmentMap>(inputFile);
}
IList <string> chromosomes = alignmentMapobj.GetRefSequences();
if (possibleOccurence)
{
Console.Write("Nucleotide Distribution:");
Console.Write("\r\nPosition\tA\tT\tG\tC\tPossibility Of Occurences");
foreach (string str in chromosomes)
{
GetCoverage(str, alignmentMapobj, "true");
}
}
else
{
Console.Write("Coverage Profile:");
Console.Write("\r\nPosition\tA\tT\tG\tC");
foreach (string str in chromosomes)
{
GetCoverage(str, alignmentMapobj, "false");
}
}
}
/// <summary>
/// Parses SAM/BAm file based on input file.
/// </summary>
private void PerformParse()
{
string samExtension = ".sam";
string bamExtension = ".bam";
if (Helper.IsBAM(InputFilename))
{
BAMParser parser = new BAMParser();
try
{
_sequenceAlignmentMap = parser.ParseOne <SequenceAlignmentMap>(InputFilename);
}
catch (Exception ex)
{
throw new InvalidOperationException(Resources.InvalidBAMFile, ex);
}
if (string.IsNullOrEmpty(OutputFilename))
{
OutputFilename = InputFilename + samExtension;
}
}
else
{
SAMParser parser = new SAMParser();
try
{
_sequenceAlignmentMap = parser.ParseOne <SequenceAlignmentMap>(InputFilename);
}
catch (Exception ex)
{
throw new InvalidOperationException(Resources.InvalidSAMFile, ex);
}
_isSAM = true;
if (string.IsNullOrEmpty(OutputFilename))
{
OutputFilename = InputFilename + bamExtension;
}
}
}
public void TestParser()
{
string FilePath = @"TestUtils\BAM\SeqAlignment.bam".TestDir();
BAMParser parser = null;
try
{
parser = new BAMParser();
SequenceAlignmentMap alignmentMap = parser.ParseOne <SequenceAlignmentMap>(FilePath);
Assert.IsTrue(alignmentMap != null);
Assert.AreEqual(alignmentMap.Header.GetReferenceSequencesInfoFromSQHeader().Count, 1);
Assert.AreEqual(alignmentMap.QuerySequences.Count, 2);
}
finally
{
if (parser != null)
{
parser.Dispose();
}
}
}
/// <summary>
/// Get chromoses with orphan regions
/// </summary>
/// <param name="filename">Path of the BAM file</param>
/// <param name="mean">Mean value</param>
/// <param name="deviation">Standard deviation</param>
/// <returns></returns>
private void DisplayOrphans(string filename)
{
SequenceAlignmentMap alignmentMapobj = null;
if (!SAMInput)
{
BAMParser bamParser = new BAMParser();
alignmentMapobj = bamParser.ParseOne <SequenceAlignmentMap>(filename);
}
else
{
SAMParser samParser = new SAMParser();
alignmentMapobj = samParser.ParseOne <SequenceAlignmentMap>(filename);
}
// get reads from sequence alignment map object.
IList <PairedRead> pairedReads = null;
// Get Aligned sequences
IList <SAMAlignedSequence> alignedSeqs = alignmentMapobj.QuerySequences;
pairedReads = alignmentMapobj.GetPairedReads(0, 0);
// Get the orphan regions.
var orphans = pairedReads.Where(PR => PR.PairedType == PairedReadType.Orphan);
int count = orphans.Count();
if (count == 0)
{
Console.WriteLine("No Orphans to display");
}
var orphanRegions = new List <ISequenceRange>(count);
orphanRegions.AddRange(orphans.Select(orphanRead => GetRegion(orphanRead.Read1)));
// Get sequence range grouping object.
SequenceRangeGrouping rangeGroup = new SequenceRangeGrouping(orphanRegions);
if (!rangeGroup.GroupIDs.Any())
{
Console.Write("\r\nNo Orphan reads to display");
}
else
{
Console.Write("Region of Orphan reads:");
DisplaySequenceRange(rangeGroup);
}
SequenceRangeGrouping mergedRegions = rangeGroup.MergeOverlaps();
if (!mergedRegions.GroupIDs.Any())
{
Console.Write("\r\nNo hot spots to display");
}
else
{
Console.Write("\r\nChromosomal hot spot:");
DisplaySequenceRange(mergedRegions);
}
}
/// <summary>
/// Indentify hot spot chromosomes for length anamoly regions.
/// </summary>
/// <param name="inputFile"> Input file</param>
/// <param name="mean">Mean value</param>
/// <param name="standardDeviation">Standard deviation</param>
private void IdentifyLentghAnamolies(string filename,
float mean = -1, float deviation = -1)
{
bool calculateMeanNdeviation = false;
if (mean == -1 || deviation == -1)
{
calculateMeanNdeviation = true;
}
SequenceAlignmentMap alignmentMapobj = null;
if (!SAMInput)
{
BAMParser bamParser = new BAMParser();
alignmentMapobj = bamParser.ParseOne <SequenceAlignmentMap>(filename);
}
else
{
SAMParser samParser = new SAMParser();
alignmentMapobj = samParser.ParseOne <SequenceAlignmentMap>(filename);
}
// get reads from sequence alignment map object.
IList <PairedRead> pairedReads = null;
if (calculateMeanNdeviation)
{
pairedReads = alignmentMapobj.GetPairedReads();
}
else
{
pairedReads = alignmentMapobj.GetPairedReads(mean, deviation);
}
// Get the orphan regions.
var orphans = pairedReads.Where(PR => PR.PairedType == PairedReadType.Orphan);
if (orphans.Count() == 0)
{
Console.WriteLine("No Orphans to display");
}
List <ISequenceRange> orphanRegions = new List <ISequenceRange>(orphans.Count());
foreach (PairedRead orphanRead in orphans)
{
orphanRegions.Add(GetRegion(orphanRead.Read1));
}
// Get sequence range grouping for Orphan regions.
SequenceRangeGrouping orphanRangegroup = new SequenceRangeGrouping(orphanRegions);
// Get the Length anomalies regions.
var lengthAnomalies = pairedReads.Where(PE => PE.PairedType == PairedReadType.LengthAnomaly);
if (lengthAnomalies.Count() == 0)
{
Console.WriteLine("No Anomalies to display");
}
List <ISequenceRange> lengthAnomalyRegions = new List <ISequenceRange>(lengthAnomalies.Count());
foreach (PairedRead laRead in lengthAnomalies)
{
SequenceRange range = new SequenceRange();
range.ID = laRead.Read1.RName;
range.Start = laRead.Read1.Pos;
range.End = laRead.Read1.Pos + laRead.InsertLength;
lengthAnomalyRegions.Add(range);
}
// Get sequence range grouping for length anomaly regions.
SequenceRangeGrouping lengthAnomalyRangegroup =
new SequenceRangeGrouping(lengthAnomalyRegions);
if (lengthAnomalyRangegroup.GroupIDs.Count() == 0)
{
Console.Write("\r\nNo Length anomalies reads to display");
}
else
{
Console.Write("Region of length anomaly:");
DisplaySequenceRange(lengthAnomalyRangegroup);
}
if (orphanRangegroup.GroupIDs.Count() == 0)
{
Console.Write("\r\nNo Orphan reads to display");
}
else
{
Console.Write("\r\nRegion of Orphan reads:");
DisplaySequenceRange(orphanRangegroup);
}
SequenceRangeGrouping intersectedRegions =
lengthAnomalyRangegroup.Intersect(orphanRangegroup);
if (intersectedRegions.GroupIDs.Count() == 0)
{
Console.Write("\r\nNo Hot spots found");
}
else
{
Console.Write("\r\nChromosomal Hot spot of length anomaly and Orphan region:");
DisplaySequenceRange(intersectedRegions);
}
}
/// <summary>
/// Merge multiple sorted alignments.
/// SAMUtil.exe out.bam in1.bam in2.bam
/// </summary>
public void DoMerge()
{
if (FilePaths == null)
{
throw new InvalidOperationException("FilePath");
}
if (FilePaths.Length < 2)
{
throw new InvalidOperationException(Resources.MergeHelp);
}
IList <IList <BAMSortedIndex> > sortedIndexes = new List <IList <BAMSortedIndex> >();
IList <SequenceAlignmentMap> sequenceAlignmentMaps = new List <SequenceAlignmentMap>();
IList <int> help = new List <int>();
Parallel.For(0, FilePaths.Length, (int index) =>
{
IList <BAMSortedIndex> sortedIndex;
BAMParser parser = new BAMParser();;
SequenceAlignmentMap map;
if (index == 0)
{
try
{
map = parser.ParseOne <SequenceAlignmentMap>(FilePaths[0]);
}
catch
{
throw new InvalidOperationException(Resources.InvalidBAMFile);
}
if (map == null)
{
throw new InvalidOperationException(Resources.EmptyFile);
}
if (string.IsNullOrEmpty(HeaderFile) && map.Header.RecordFields.Count == 0)
{
throw new InvalidOperationException(Resources.HeaderMissing);
}
if (!string.IsNullOrEmpty(HeaderFile))
{
SAMParser parse = new SAMParser();
SequenceAlignmentMap head;
try
{
head = parse.ParseOne <SequenceAlignmentMap>(HeaderFile);
}
catch
{
throw new InvalidOperationException(Resources.IncorrectHeaderFile);
}
if (head == null)
{
throw new InvalidOperationException(Resources.EmptyFile);
}
header = head.Header;
}
else
{
header = map.Header;
}
sortedIndex = Sort(map, SortByReadName ? BAMSortByFields.ReadNames : BAMSortByFields.ChromosomeCoordinates);
}
else
{
try
{
map = parser.ParseOne <SequenceAlignmentMap>(FilePaths[index]);
}
catch
{
throw new InvalidOperationException(Resources.InvalidBAMFile);
}
if (map == null)
{
throw new InvalidOperationException(Resources.EmptyFile);
}
sortedIndex = Sort(map, SortByReadName ? BAMSortByFields.ReadNames : BAMSortByFields.ChromosomeCoordinates);
}
lock (sortedIndexes)
{
sortedIndexes.Add(sortedIndex);
sequenceAlignmentMaps.Add(map);
}
});
if (string.IsNullOrEmpty(OutputFilename))
{
OutputFilename = "out.bam";
autoGeneratedOutputFilename = true;
}
string filePath = Path.GetTempFileName();
using (FileStream fstemp = new FileStream(filePath, FileMode.Create, FileAccess.ReadWrite))
{
BAMFormatter formatter = new BAMFormatter();
formatter.WriteHeader(header, fstemp);
int[] indexes = new int[sortedIndexes.Count];
if (SortByReadName)
{
IList <BAMSortedIndex> sortedIndex = sortedIndexes.Select(a => a.First()).ToList();
WriteMergeFileSortedByReadName(sortedIndex, fstemp, formatter, sequenceAlignmentMaps);
}
else
{
WriteMergeFile(sortedIndexes, fstemp, formatter, sequenceAlignmentMaps);
}
using (FileStream fsoutput = new FileStream(OutputFilename, FileMode.Create, FileAccess.Write))
{
fstemp.Seek(0, SeekOrigin.Begin);
formatter.CompressBAMFile(fstemp, fsoutput);
}
}
File.Delete(filePath);
if (autoGeneratedOutputFilename)
{
Console.WriteLine(Properties.Resources.SuccessMessageWithOutputFileName, OutputFilename);
}
}