/// <summary>
/// constructor
/// </summary>
private BreakEnd(string referenceName, int position, ChromosomeRenamer renamer)
{
_referenceIndex = renamer.GetReferenceIndex(referenceName);
_referenceName = _referenceIndex >= renamer.NumRefSeqs ? referenceName : renamer.EnsemblReferenceNames[_referenceIndex];
Position = position;
_renamer = renamer;
}
private void AssignVcfFields()
{
ReferenceName = VcfColumns[VcfCommon.ChromIndex];
ReferenceIndex = _renamer.GetReferenceIndex(ReferenceName);
EnsemblReferenceName = ReferenceIndex == ushort.MaxValue? ReferenceName: _renamer.EnsemblReferenceNames[ReferenceIndex];
UcscReferenceName = ReferenceIndex == ushort.MaxValue ? ReferenceName : _renamer.UcscReferenceNames[ReferenceIndex];
var referenceAllele = VcfColumns[VcfCommon.RefIndex];
VcfRefAllele = referenceAllele;
VcfReferenceBegin = int.Parse(VcfColumns[VcfCommon.PosIndex]);
VcfReferenceEnd = VcfReferenceBegin + referenceAllele.Length - 1;
VcfVariantId = VcfColumns[VcfCommon.IdIndex];
OverlapReferenceBegin = VcfReferenceBegin;
OverlapReferenceEnd = VcfReferenceEnd;
IsReference = VcfColumns[VcfCommon.AltIndex] == VcfCommon.NonVariant;
IsRefNoCall = false;
// reset the info field
if (VcfColumns[VcfCommon.InfoIndex] == ".")
{
VcfColumns[VcfCommon.InfoIndex] = "";
}
}
/// <summary>
/// returns a list of reference cytogenetic bands
/// </summary>
private static ICytogeneticBands GetCytogeneticBands(string inputCytobandPath, ChromosomeRenamer renamer)
{
var numRefSeqs = renamer.NumRefSeqs;
var bandLists = new List <Band> [numRefSeqs];
for (int i = 0; i < numRefSeqs; i++)
{
bandLists[i] = new List <Band>();
}
using (var reader = new StreamReader(FileUtilities.GetReadStream(inputCytobandPath)))
{
while (true)
{
// grab the next line
string line = reader.ReadLine();
if (string.IsNullOrEmpty(line))
{
break;
}
// split the line into columns
var cols = line.Split('\t');
// sanity check: make sure we have the right number of columns
const int expectedNumColumns = 5;
if (cols.Length != expectedNumColumns)
{
throw new GeneralException($"Expected {expectedNumColumns} columns, but found {cols.Length} columns: [{line}]");
}
// grab the essential values
var ucscName = cols[0];
var begin = int.Parse(cols[1]) + 1;
var end = int.Parse(cols[2]);
var name = cols[3];
ushort refIndex = renamer.GetReferenceIndex(ucscName);
if (refIndex == ChromosomeRenamer.UnknownReferenceIndex)
{
continue;
}
bandLists[refIndex].Add(new Band(begin, end, name));
}
}
// create the band arrays
var bands = new Band[numRefSeqs][];
for (int i = 0; i < numRefSeqs; i++)
{
bands[i] = bandLists[i].ToArray();
}
return(new CytogeneticBands(bands, renamer));
}
/// <summary>
/// constructor
/// </summary>
public BreakEnd(string referenceName, string referenceName2, int position, int position2, char isSuffix,
char isSuffix2, ChromosomeRenamer renamer) : this(referenceName, position, renamer)
{
ReferenceIndex2 = renamer.GetReferenceIndex(referenceName2);
_referenceName2 = ReferenceIndex2 >= renamer.NumRefSeqs ? referenceName2 : renamer.EnsemblReferenceNames[ReferenceIndex2];
Position2 = position2;
IsSuffix = isSuffix;
IsSuffix2 = isSuffix2;
_orientation = IsSuffix == '+' ? '-' : '+';
_orientation2 = IsSuffix2 == '+' ? '+' : '-';
}
/// <summary>
/// parses the alternate allele
/// </summary>
private void ParseAltAllele(string refAllele, string altAllele)
{
var regexSuccess = false;
// (\w+)([\[\]])([^:]+):(\d+)([\[\]])
// ([\[\]])([^:]+):(\d+)([\[\]])(\w+)
if (altAllele.StartsWith(refAllele))
{
var forwardRegex = new Regex(@"\w+([\[\]])([^:]+):(\d+)([\[\]])", RegexOptions.Compiled);
var match = forwardRegex.Match(altAllele);
if (match.Success)
{
IsSuffix = '-';
_orientation = '+';
_referenceName2 = match.Groups[2].Value;
Position2 = Convert.ToInt32(match.Groups[3].Value);
IsSuffix2 = match.Groups[4].Value == ForwardBreakEnd ? '+' : '-';
_orientation2 = match.Groups[4].Value == ForwardBreakEnd ? '+' : '-';
regexSuccess = true;
ReferenceIndex2 = _renamer.GetReferenceIndex(match.Groups[2].Value);
if (ReferenceIndex2 < _renamer.NumRefSeqs)
{
_referenceName2 = _renamer.EnsemblReferenceNames[ReferenceIndex2];
}
else
{
return;
}
}
}
else
{
var reverseRegex = new Regex(@"([\[\]])([^:]+):(\d+)([\[\]])\w+", RegexOptions.Compiled);
var match = reverseRegex.Match(altAllele);
if (match.Success)
{
IsSuffix = '+';
_orientation = '-';
IsSuffix2 = match.Groups[1].Value == ForwardBreakEnd ? '+' : '-';
_orientation2 = match.Groups[1].Value == ForwardBreakEnd ? '+' : '-';
_referenceName2 = match.Groups[2].Value;
Position2 = Convert.ToInt32(match.Groups[3].Value);
regexSuccess = true;
ReferenceIndex2 = _renamer.GetReferenceIndex(match.Groups[2].Value);
if (ReferenceIndex2 < _renamer.NumRefSeqs)
{
_referenceName2 = _renamer.EnsemblReferenceNames[ReferenceIndex2];
}
else
{
return;
}
}
}
if (!regexSuccess)
{
throw new GeneralException(
"Unable to successfully parse the complex rearrangements for the following allele: " + altAllele);
}
}
private static CacheFile TryMatchFilename(string ndbPath, Func <string, Match> matcher, MiniCacheType type,
ChromosomeRenamer renamer)
{
string filename = Path.GetFileName(ndbPath);
if (filename == null)
{
return(null);
}
var match = matcher(filename);
if (!match.Success)
{
return(null);
}
IUpdater updater;
string id, transcriptDataSource;
int position;
ushort refIndex;
switch (type)
{
case MiniCacheType.Transcript:
var tuple = FormatUtilities.SplitVersion(match.Groups[1].Value);
id = tuple.Item1;
refIndex = renamer.GetReferenceIndex(match.Groups[2].Value);
transcriptDataSource = match.Groups[3].Value;
updater = new TranscriptUpdater(id, refIndex, transcriptDataSource);
break;
case MiniCacheType.Regulatory:
id = match.Groups[1].Value;
refIndex = renamer.GetReferenceIndex(match.Groups[2].Value);
transcriptDataSource = match.Groups[3].Value;
updater = new RegulatoryUpdater(id, refIndex, transcriptDataSource);
break;
case MiniCacheType.Position:
refIndex = renamer.GetReferenceIndex(match.Groups[1].Value);
position = int.Parse(match.Groups[2].Value);
string refAllele = match.Groups[3].Value;
string altAllele = match.Groups[4].Value;
transcriptDataSource = match.Groups[5].Value;
updater = new PositionUpdater(refIndex, position, refAllele, altAllele, transcriptDataSource);
break;
case MiniCacheType.PositionRange:
refIndex = renamer.GetReferenceIndex(match.Groups[1].Value);
position = int.Parse(match.Groups[2].Value);
int endPosition = int.Parse(match.Groups[3].Value);
transcriptDataSource = match.Groups[4].Value;
updater = new PositionRangeUpdater(refIndex, position, endPosition, transcriptDataSource);
break;
default:
throw new GeneralException($"Unexpected mini-cache type encountered: {type}");
}
return(new CacheFile(ndbPath, updater.RefIndex, ConvertTranscriptDataSource(updater.TranscriptDataSource),
type, updater));
}
private SupplementaryPositionCreator GetNextSupplementaryAnnotation()
{
// no more active iterators left
if (_iSupplementaryDataItemList.Count == 0 && _additionalItemsList.Count == 0)
{
return(null);
}
var minSupplementaryDataItem = CurrentMinSupplementaryDataItem();
if (minSupplementaryDataItem == null)
{
return(null); //nothing more to retun. All enumerators are empty.
}
var sa = new SupplementaryAnnotationPosition(minSupplementaryDataItem.Start);
var saCreator = new SupplementaryPositionCreator(sa)
{
RefSeqName = minSupplementaryDataItem.Chromosome
};
string refSequence = null;
if (_currentRefName == null || !_currentRefName.Equals(saCreator.RefSeqName))
{
CloseCurrentSaWriter();
_currentRefName = saCreator.RefSeqName;
var refIndex = _renamer.GetReferenceIndex(_currentRefName);
if (refIndex == ChromosomeRenamer.UnknownReferenceIndex)
{
throw new GeneralException($"Could not find the reference index for: {_currentRefName}");
}
_dataFileManager.LoadReference(refIndex, () => {});
OpenNewSaWriter();
}
if (_compressedSequence != null)
{
refSequence = _compressedSequence.Substring(sa.ReferencePosition - 1, ReferenceWindowSize);
}
// list of data items to be removed and added
var deleteList = new List <IEnumerator <SupplementaryDataItem> >();
foreach (var iDataEnumerator in _iSupplementaryDataItemList)
{
// only using items at the same location as minSuppDataItem
if (!iDataEnumerator.Current.Equals(minSupplementaryDataItem))
{
continue;
}
if (iDataEnumerator.Current.IsInterval)
{
var suppInterval = iDataEnumerator.Current.GetSupplementaryInterval(_renamer);
_supplementaryIntervalList.Add(suppInterval);
}
else
{
var additionalSuppData = iDataEnumerator.Current.SetSupplementaryAnnotations(saCreator, refSequence);
if (additionalSuppData != null)
{
_additionalItemsList.Add(additionalSuppData);
}
}
// adding empty enumerators to deleteList
if (!iDataEnumerator.MoveNext())
{
deleteList.Add(iDataEnumerator);
}
}
// add annotations from additional items if applicable.
AddAdditionalItems(minSupplementaryDataItem, saCreator);
// removing lists that are empty and therfore should be removed from the list of enumerators
_iSupplementaryDataItemList.RemoveAll(x => deleteList.Contains(x));
return(saCreator);
}