SearchReference results in java.lang.NullPointerException

[birnbera]

Hello, I have been excited to try this tool for off-target prediction and finally got to play with it today. Unfortunately, I can't seem to get it to work. The issue is a NullPointerException that appears to be caught here, although I couldn't track down where it originates.

I tried running the JAR directly as well as via conda, but both methods result in the same error. Interestingly, the application continues, which makes me wonder why the error is ignored/ignorable. Also, sometimes other messages are visible if I manage to interrupt before too many error messages pile up:

[2021/05/04 17:36:35 | SearchReference | Info] Seen many non-increasing record positions. Printing Read-names as well.

If I let the application continue, it never finishes and eventually seems to freeze. Any idea what may be causing this?

[tfenne]

@birnbera Apologies for the super slow response. It's rather hard to say what's going on without more context. Is there any chance you could share the command line you're running and the stack trace (if one is produced) from the output?

Failing that if you could share how you're constructing your inputs (e.g. reference genome, any VCFs) etc. that would also be very helpful in tracking this down.

[ct-dwhitney]

I have a similar problem. On the first run using the example for input provided in your README.md, I got thousands of errors as the post as the top.

Next, I lowered the mismatch criteria from 5 to 2 and the run still threw intermittent errors as above, but it appeared to scan the genome appropriately. Unfortunately, the output text file had only a header and no other entries. Perhaps this is correct based on the guide provided?

I will note that I used hg19 for the genome reference rather than hg38. Do you think this made a difference?