public void SanityCheckSequenceOrdering()
{
var smallBam = Path.Combine(TestPaths.LocalTestDataDirectory, "Ins-L3-var12_S12.bam");
var intervals = new Dictionary <string, List <Region> >();
var chrIntervals = new List <Region>
{
new Region(28607838, 28607838),
new Region(28608631, 28608631)
};
var expectedSQorder = new List <string> {
"chr10", "chr11", "chr12", "chr13"
}; //I dont know why it starts with 10, thats just how it is in the bam. thats what makes it a good test case.
intervals.Add("chr13", chrIntervals);
var extractor = new BamFileAlignmentExtractor(smallBam);
List <string> sequencesInTheBamOrder = extractor.SourceReferenceList;
Assert.Equal(expectedSQorder[0], sequencesInTheBamOrder[0]);
Assert.Equal(expectedSQorder[1], sequencesInTheBamOrder[1]);
Assert.Equal(expectedSQorder[3], sequencesInTheBamOrder[3]);
Assert.Equal(25, sequencesInTheBamOrder.Count);
//happyPath
Assert.False(extractor.SequenceOrderingIsNotConsistent(new List <string> {
"chr1", "chr2"
}));
Assert.False(extractor.SequenceOrderingIsNotConsistent(new List <string> {
"chr1", "chr3", "chr4"
}));
Assert.False(extractor.SequenceOrderingIsNotConsistent(new List <string> {
"chr14", "chr9"
})); //only b/c the bam header is silly.
//not OK
Assert.True(extractor.SequenceOrderingIsNotConsistent(new List <string> {
"chr2", "chr1"
}));
Assert.True(extractor.SequenceOrderingIsNotConsistent(new List <string> {
"chr9", "chr14"
}));
Assert.True(extractor.SequenceOrderingIsNotConsistent(new List <string> {
"chr22", "chr21"
}));
//genome has chr not in bam, be ok with it
Assert.False(extractor.SequenceOrderingIsNotConsistent(new List <string> {
"chr1", "chrMotherGoose"
}));
//bam has chr not in genome, be ok with it
Assert.False(extractor.SequenceOrderingIsNotConsistent(new List <string> {
"chr1"
}));
//empty lists
Assert.False(extractor.SequenceOrderingIsNotConsistent(new List <string> {
}));
Assert.False(extractor.SequenceOrderingIsNotConsistent(null));
}
protected virtual IAlignmentSource CreateAlignmentSource(ChrReference chrReference, string bamFilePath, List <string> chrsToProcess = null)
{
AlignmentMateFinder mateFinder = null;
var alignmentExtractor = new BamFileAlignmentExtractor(bamFilePath, chrReference.Name, _bamIntervalLookup.ContainsKey(bamFilePath) && _options.SkipNonIntervalAlignments ? _bamIntervalLookup[bamFilePath] : null);
//Warn if the bam has sequences ordered differently to the reference genome.
//That would confuse us because we will not know how the user wants to order the output gvcf.
if (alignmentExtractor.SequenceOrderingIsNotConsistent(chrsToProcess))
{
Logger.WriteToLog("Warning: Reference sequences in the bam do not match the order of the reference sequences in the genome. Check bam " + bamFilePath);
Logger.WriteToLog("Variants will be ordered according to the reference genome");
}
var config = new AlignmentSourceConfig
{
MinimumMapQuality = _options.MinimumMapQuality,
OnlyUseProperPairs = _options.OnlyUseProperPairs,
};
return(new AlignmentSource(alignmentExtractor, mateFinder, config));
}
protected virtual IAlignmentSource CreateAlignmentSource(ChrReference chrReference, string bamFilePath, bool commandLineSaysStitched, List <string> chrsToProcess = null)
{
AlignmentMateFinder mateFinder = null;
var alignmentExtractor = new BamFileAlignmentExtractor(bamFilePath, commandLineSaysStitched, chrReference.Name);
//Warn if the bam has sequences ordered differently to the reference genome.
//That would confuse us because we will not know how the user wants to order the output gvcf.
if (alignmentExtractor.SequenceOrderingIsNotConsistent(chrsToProcess))
{
Logger.WriteToLog("Warning: Reference sequences in the bam do not match the order of the reference sequences in the genome. Check bam " + bamFilePath);
Logger.WriteToLog("Variants will be ordered according to the reference genome");
}
var config = new AlignmentSourceConfig
{
MinimumMapQuality = _options.BamFilterParameters.MinimumMapQuality,
OnlyUseProperPairs = _options.BamFilterParameters.OnlyUseProperPairs,
SkipDuplicates = _options.BamFilterParameters.RemoveDuplicates
};
return(new AlignmentSource(alignmentExtractor, mateFinder, config));
}
