some paired end BAM files, e.g. from GIAB have a weird thing where there is a duplication of the reads at each position. There is no BAM flag indicating it is a duplicate, so there's no real way to distinguish these. Our Breakpoint split view tries to connect features that have the same QNAME (read name) to match pairs (pairs share the same read name) but this weird duplicate reads also share the same QNAME.
This PR filters makes it so paired end reads specifically don't connect with features at the same position
before
after
focus on the 'problematic reads'
I filtered out just one qname which shows that it matches 4 boxes...typically it should be two (left and right side of pair) and any other matches to the qname should at least be distinguished by something but these aren't
some paired end BAM files, e.g. from GIAB have a weird thing where there is a duplication of the reads at each position. There is no BAM flag indicating it is a duplicate, so there's no real way to distinguish these. Our Breakpoint split view tries to connect features that have the same QNAME (read name) to match pairs (pairs share the same read name) but this weird duplicate reads also share the same QNAME.
This PR filters makes it so paired end reads specifically don't connect with features at the same position
before
after
focus on the 'problematic reads'
I filtered out just one qname which shows that it matches 4 boxes...typically it should be two (left and right side of pair) and any other matches to the qname should at least be distinguished by something but these aren't
https://jbrowse.org/code/jb2/main/?config=test_data%2Fconfig_demo.json&session=share-8ybk_-VNx0&password=5ww3R
all four of those boxes share the same QNAME