Variant reported out of region of interest

[PabloCabaleiro]

Hi,

I've detected variants reported out of the region of interest defined either by the .bed or the -R option.

With the files I've attached you, I've detected the variant 15:74499956-74499956G>C with the execution:

> VarDict-1.8.2 -G hs37d5.fa -b example.bam -R 15:74500095-74500098
example 15 15 74499956 74499956 G C 1988 986 0 0 684 302 C/C 0.4960 0;2 39.2 1 33.2 1 60.0 12.324 0.9989 0 0 2.000 1 1.5 912 913 TTTACAATACTCAAACCTCA GTACACCCCCAACTTCGATA 15:74500095-74500098 SNV 0 61-0-0

But with the execution > VarDict-1.8.2 -G hs37d5.fa -b example.bam -R 15:74500096-74500098 is no longer reported.

The variant is 141pb before the roi, but I've detected other cases with a different distance. Is there any explanation of this behavior?

example.zip

[clintval]

@PolinaBevad I've also noticed this issue. Do you think a fix could be simple? I would be willing to give it a shot if you could provide some direction. Thank you for your time!