public override bool PrepareOptions()
{
if (!PrepareOutputDirectory())
{
return(false);
}
try
{
using (SAMFactory.GetReader(this.NormalFile)) { }
using (SAMFactory.GetReader(this.TumorFile)) { }
}
catch (Exception ex)
{
ParsingErrors.Add(ex.Message);
return(false);
}
if (this.ThreadCount >= 2)
{
Console.WriteLine("Checking chromosome names for thread mode ...");
if (this.ChromosomeNames == null || this.ChromosomeNames.Count == 0)
{
this.ChromosomeNames = SAMUtils.GetChromosomes(this.NormalFile);
}
foreach (var chr in this.ChromosomeNames)
{
Console.WriteLine(chr);
}
}
else
{
if (this.ChromosomeNames != null && this.ChromosomeNames.Count > 0)
{
Console.Out.WriteLine("#mpileup chromosome names: " + this.ChromosomeNames.Merge(","));
}
}
return(true);
}
protected virtual List <T> DoAddCompleted <T>(List <T> samlist) where T : SAMAlignedItem, new()
{
if (_options.EngineType == 4 || _options.GetSAMFormat().HasAlternativeHits || samlist.Count == 0)
{
return(samlist);
}
Progress.SetMessage("Sorting mapped reads by name...");
SAMUtils.SortByName(samlist);
Progress.SetMessage("Merge reads from same query...");
var result = new List <T>();
result.Add(samlist[0]);
T last = samlist[0];
for (int i = 1; i < samlist.Count; i++)
{
var sam = samlist[i];
if (!last.Qname.Equals(sam.Qname))
{
last = sam;
result.Add(last);
}
else
{
last.AddLocations(sam.Locations);
sam.ClearLocations();
}
}
samlist.Clear();
samlist = null;
KeepUniqueLocation <T>(result);
Progress.SetMessage("Total {0} read(s) mapped.", result.Count);
return(result);
}
public int Extract(string sourceFile, string targetFile, IEnumerable <string> exceptQueryNames, string countFile)
{
int result = 0;
var except = new HashSet <string>(exceptQueryNames);
SmallRNACountMap cm = new SmallRNACountMap();
StreamWriter swCount = null;
if (File.Exists(countFile))
{
var oldCm = new SmallRNACountMap(countFile);
foreach (var c in oldCm.Counts)
{
cm.Counts[c.Key.StringBefore(SmallRNAConsts.NTA_TAG)] = c.Value;
}
swCount = new StreamWriter(targetFile + ".dupcount");
}
try
{
using (var sw = StreamUtils.GetWriter(targetFile, targetFile.ToLower().EndsWith(".gz")))
{
using (var sr = SAMFactory.GetReader(sourceFile, true))
{
string line;
var count = 0;
while ((line = sr.ReadLine()) != null)
{
count++;
if (count % 100000 == 0)
{
Progress.SetMessage("{0} reads", count);
if (Progress.IsCancellationPending())
{
throw new UserTerminatedException();
}
}
var ss = SAMUtils.Parse <SAMItemSlim>(line);
ss.Qname = ss.Qname.StringBefore(SmallRNAConsts.NTA_TAG);
if (except.Contains(ss.Qname))
{
continue;
}
if (Filter != null && !Filter.Accept(ss))
{
continue;
}
except.Add(ss.Qname);
ss.WriteFastq(sw);
if (swCount != null)
{
swCount.WriteLine("{0}\t{1}", ss.Qname, cm.Counts[ss.Qname]);
}
result++;
}
}
}
}
finally
{
if (swCount != null)
{
swCount.Close();
}
}
return(result);
}
protected override List <T> DoBuild <T>(string fileName, out List <QueryInfo> totalQueries)
{
var result = new List <T>();
_format = _options.GetSAMFormat();
totalQueries = new List <QueryInfo>();
using (var sr = SAMFactory.GetReader(fileName, true))
{
int count = 0;
int waitingcount = 0;
string line;
while ((line = sr.ReadLine()) != null)
{
count++;
if (count % 1000 == 0)
{
if (Progress.IsCancellationPending())
{
throw new UserTerminatedException();
}
}
var parts = line.Split('\t');
var qname = parts[SAMFormatConst.QNAME_INDEX];
var qi = new QueryInfo(qname);
totalQueries.Add(qi);
SAMFlags flag = (SAMFlags)int.Parse(parts[SAMFormatConst.FLAG_INDEX]);
if (!_filter.AcceptFlags(flag))
{
continue;
}
var mismatchCount = _format.GetNumberOfMismatch(parts);
var seq = parts[SAMFormatConst.SEQ_INDEX];
qi.Mismatch = mismatchCount;
qi.Length = seq.Length;
//too many mismatchs
if (!_filter.AcceptMismatch(mismatchCount))
{
continue;
}
if (!_filter.AcceptQueryName(qname))
{
continue;
}
if (!_filter.AcceptLength(seq.Length))
{
continue;
}
var cigar = parts[SAMFormatConst.CIGAR_INDEX];
if (!_filter.AcceptCigar(cigar))
{
continue;
}
var seqname = parts[SAMFormatConst.RNAME_INDEX].StringAfter("chr");
var start = int.Parse(parts[SAMFormatConst.POS_INDEX]);
var end = SAMUtils.ParseEnd(start, cigar);
bool isReversed = flag.HasFlag(SAMFlags.QueryOnReverseStrand);
char strand;
if (isReversed)
{
strand = '-';
}
else
{
strand = '+';
}
var sam = new T();
var loc = new SAMAlignedLocation(sam)
{
Seqname = seqname,
Start = start,
End = end,
Strand = strand,
};
if (!_filter.AcceptLocus(loc))
{
continue;
}
if (isReversed)
{
seq = SequenceUtils.GetReverseComplementedSequence(seq);
}
sam.Qname = qname;
sam.Sequence = seq;
loc.AlignmentScore = _format.GetAlignmentScore(parts);
loc.Cigar = cigar;
loc.NumberOfMismatch = mismatchCount;
loc.MismatchPositions = _format.GetMismatchPositions(parts);
if (_format.HasAlternativeHits)
{
_format.ParseAlternativeHits(parts, sam);
}
result.Add(sam);
waitingcount++;
if (waitingcount % 100 == 0)
{
Progress.SetMessage("{0} feature reads from {1} reads", waitingcount, count);
}
}
}
return(result);
}
private static SAMItemSlim LineToSamItem(string line)
{
return(SAMUtils.Parse <SAMItemSlim>(line));
}