Hi guys, I'm analysing some samples from a resequencing project.
After mapping (BWA-mem) and removing duplicates I've found a background of anomalous paired-end reads mapping to almost every exon:
1. read1 mapped +read2 unmapped (or viceversa).
2. reads with same orientation (FF or RR) and abnormal insert size.
3. reads with abnormal insert size.
4. reads whose mate is mapping to another chromosome.
I assume this kind of reads have to be filtered when performing a variant call (by using flags 0x0002 and 0x100), so none of this reads are passing the filter.
Does anyone know why are they appearing?
Would be useful to keep them to detect SVs after establishing a background noise?
Thanks
After mapping (BWA-mem) and removing duplicates I've found a background of anomalous paired-end reads mapping to almost every exon:
1. read1 mapped +read2 unmapped (or viceversa).
2. reads with same orientation (FF or RR) and abnormal insert size.
3. reads with abnormal insert size.
4. reads whose mate is mapping to another chromosome.
I assume this kind of reads have to be filtered when performing a variant call (by using flags 0x0002 and 0x100), so none of this reads are passing the filter.
Does anyone know why are they appearing?
Would be useful to keep them to detect SVs after establishing a background noise?
Thanks