public ChromosomeIndelSource(List <HashableIndel> indels, IGenomeSnippetSource snippetSource, int bucketSize = 1000)
{
_bucketSize = bucketSize;
Indels = indels;
var partnerIndelsLookup = new Dictionary <HashableIndel, IEnumerable <HashableIndel> >();
var positionalBuckets = new Dictionary <int, List <HashableIndel> >();
var snippetsLookup = new Dictionary <int, GenomeSnippet>();
if (!indels.Any())
{
_partnerIndels = new ReadOnlyDictionary <HashableIndel, IEnumerable <HashableIndel> >(partnerIndelsLookup);
_genomeSnippetsLookup = new ReadOnlyDictionary <int, GenomeSnippet>(snippetsLookup);
_positionalBucketsOfIndels = new ReadOnlyDictionary <int, List <HashableIndel> >(positionalBuckets);
return;
}
LowestPosition = indels.Min(x => x.ReferencePosition);
HighestPosition = indels.Max(x => x.ReferencePosition);
foreach (var indel in indels)
{
var bucketNum = ((indel.ReferencePosition - LowestPosition) / bucketSize);
if (!positionalBuckets.TryGetValue(bucketNum, out var indelsForbucket))
{
indelsForbucket = new List <HashableIndel>();
positionalBuckets.Add(bucketNum, indelsForbucket);
}
// TODO come back to this if needed - was thinking I could link indels to each other so that if multi-indels contain indels in multiple buckets we would make sure to grab all
// Maybe add start and end pos of hashables, and this could take into account the main and the other
//var indelString = Helper.HashableToString(indel);
var indelString = indel.StringRepresentation;
var partnerIndels = indels.Where(x =>
x.StringRepresentation == indelString || x.OtherIndel == indelString);
partnerIndelsLookup.Add(indel, partnerIndels);
indelsForbucket.Add(indel);
_numIndels++;
}
_partnerIndels = new ReadOnlyDictionary <HashableIndel, IEnumerable <HashableIndel> >(partnerIndelsLookup);
_positionalBucketsOfIndels = new ReadOnlyDictionary <int, List <HashableIndel> >(positionalBuckets);
foreach (var kvp in _positionalBucketsOfIndels)
{
var bucket = kvp.Value;
if (bucket.Any())
{
var firstIndel = bucket.First();
var snippet = snippetSource.GetGenomeSnippet(firstIndel.ReferencePosition);
snippetsLookup[kvp.Key] = snippet;
}
}
_genomeSnippetsLookup = new ReadOnlyDictionary <int, GenomeSnippet>(snippetsLookup);
}
public GenomeSnippet GetGenomeSnippet(int position)
{
if (position < 0)
{
throw new ArgumentException(
$"Invalid snippet reference position ({position}): must be non-negative.");
}
if (Math.Abs(position - _lastPosition) < _snippetBuffer && _currentEndPos - position > _snippetBuffer)
{
return(_snippet);
}
else
{
_snippet = _snippetSource.GetGenomeSnippet(position);
_lastPosition = position;
_currentEndPos = _snippet.StartPosition + _snippet.Sequence.Length;
return(_snippet);
}
}
public int GetNm(BamAlignment alignment)
{
var positionMap = new PositionMap(alignment.Bases.Length);
Read.UpdatePositionMap(alignment.Position + 1, alignment.CigarData, positionMap);
var snippet = _genomeSnippetSource.GetGenomeSnippet(alignment.Position);
var numMismatches =
Helper.GetNumMismatches(alignment.Bases, positionMap, snippet.Sequence, snippet.StartPosition);
if (numMismatches == null)
{
throw new Exception("Num mismatches is null");
}
var numIndelBases = alignment.CigarData.NumIndelBases();
return(numMismatches.Value + numIndelBases);
}
