Wrong wt_coverage with multiple duplications at the same location
When a same read has several duplications almost at the same loci, the wt_coverage is wrong as the coverage of all the duplications are removed while only one read contributed to them.
See for instance this example where we have a duplication of ATATGAATATGA (which, itself is an ITD…) and ATATGAATATGAATATGAATA (which also includes the first):seq.fa
{"start_pos" : 296,"stop_pos" : 298,"size" : 6,"occurrence" : 7,"raw_occurrence" : 6,"region_coverage" : 14,"wt_coverage" : 1,"MT/WT" : 7.000000,"VAF": 0.500000,"break_size" : 3,"is_duplication" : true,"is_wt_duplication" : true, "sequence": "GAATAT", "approximate_full_sequence": "GAATAT", "reference_pos": "13:28034128:-1"},
{"start_pos" : 286,"stop_pos" : 299,"size" : 21,"occurrence" : 3,"raw_occurrence" : 3,"region_coverage" : 13,"wt_coverage" : 8,"MT/WT" : 0.375000,"VAF": 0.230769,"break_size" : 7,"is_duplication" : true,"is_wt_duplication" : false, "sequence": "ATATGAATATGAATATGAATA", "approximate_full_sequence": "ATATGAATATGAATATGAATA", "reference_pos": "13:28034123:-1"}
While we have 23 reads in total, including 20 reads of the reference, the wt_coverage
is way too low.