/// <summary>
/// Sorts the SequenceAlignmentMap based on the sort by fields,
/// either chromosome coordinates or read names and retuns sorted BAM indexer
/// </summary>
/// <example>
/// 1. Sort by chromosome name.
/// BAMSort sorter = new BAMSort([SequenceAlignmentMap], BAMSortByFields.ChromosomeCoordinates);
/// IList<BAMSortedIndex> sortedGroups = sorter.Sort();
/// foreach (BAMSortedIndex sortedGroup in sortedGroups)
/// {
/// sortedGroup.GroupName // Containes the RName
/// foreach (int index in sortedGroup)
/// {
/// index // index of SequenceAlignmentMap.QuerySequences
/// }
/// }
/// 2. Sort by read name.
/// BAMSort sorter = new BAMSort([SequenceAlignmentMap], BAMSortByFields.ChromosomeCoordinates);
/// IList<BAMSortedIndex> sortedGroups = sorter.Sort();
/// foreach (int index in sortedGroups[0]) // There will be only Group in list.
/// {
/// index // index of SequenceAlignmentMap.QuerySequences
/// }
/// </example>
/// <returns>sorted BAM indexer</returns>
public IList <BAMSortedIndex> Sort()
{
IList <BAMSortedIndex> sortedIndices = new List <BAMSortedIndex>();
switch (_sortType)
{
case BAMSortByFields.ChromosomeNameAndCoordinates:
// Sort by Chromosomes (RName) and then by Positions (Pos) and retun "BAMSortedIndex"
// containing the indices of sorted SequenceAlignmentMap.QuerySequences items.
SortedDictionary <string, IList <string> > sortedFiles = SortByChromosomeCoordinates();
foreach (KeyValuePair <string, IList <string> > sortedFile in sortedFiles)
{
BAMSortedIndex sortedIndex = new BAMSortedIndex(sortedFile.Value, _sortType);
sortedIndex.GroupName = sortedFile.Key;
sortedIndices.Add(sortedIndex);
}
break;
case BAMSortByFields.ChromosomeCoordinates:
// Sort by Chromosomes Positions (Pos) and retun "BAMSortedIndex"
// containing the indices of sorted SequenceAlignmentMap.QuerySequences items.
sortedFiles = SortByChromosomeCoordinates();
foreach (string refName in _seqAlignMap.GetRefSequences())
{
IList <string> filenames = null;
if (sortedFiles.TryGetValue(refName, out filenames))
{
BAMSortedIndex sortedIndex = new BAMSortedIndex(filenames, _sortType);
sortedIndex.GroupName = refName;
sortedIndices.Add(sortedIndex);
}
}
break;
case BAMSortByFields.ReadNames:
// Sort by Read name (QName) and retun "BAMSortedIndex" containing the indices of
// sorted SequenceAlignmentMap.QuerySequences items.
sortedIndices.Add(new BAMSortedIndex(SortByReadNames(), _sortType));
break;
}
return(sortedIndices);
}
/// <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.Parse(filename);
}
else
{
SAMParser samParser = new SAMParser();
alignmentMapobj = samParser.Parse(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>
/// 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.Parse(inputFile);
}
else
{
SAMParser samParser = new SAMParser();
alignmentMapobj = samParser.Parse(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>
/// Tests Chimeric stats.
/// </summary>
private static void TestChimeraData(SequenceAlignmentMap alignmentMapobj)
{
string expectedOutput;
string actualOutput;
expectedOutput = "varchr1chr2chr3chr4chr10320chr22040chr33100chr40000";
// 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(CultureInfo.InvariantCulture);
}
}
actualOutput = statistics.ToString2D().Replace(Environment.NewLine, "").Replace("\t", "");
Assert.AreEqual(expectedOutput, actualOutput);
}