nanoporetech / modkit

A bioinformatics tool for working with modified bases
https://nanoporetech.com/
Other
116 stars 6 forks source link

Getting error when running modkit dmr pair #208

Closed agoel13 closed 6 days ago

agoel13 commented 2 weeks ago

Hi,

I am trying to run modkit (v0.3.0) dmr pair using replicates and getting following error

Error msg : running with replicates, but not matched samples thread 'main' panicked at src/hmm.rs:330:9: e_s is not finite -inf note: run with RUST_BACKTRACE=1 environment variable to display a backtrace

For your reference here is my script:

_modkit dmr pair \ -a Norm1.bed.gz \ -a Norm2.bed.gz \ -b Tum1.bed.gz \ -b Tum2.bed.gz \ -b Tum3.bed.gz \ -b Tum4.bed.gz \ -o dmr_replicates_pair_segment.bed \ --header \ --segment dmr_segements.bed \ --ref hg38.fa \ --base C \ --threads 16 \ --log-filepath dmr_pairsegment.log

While when I run the same script with only one pair of samples and no replicates, it completes without any errors.

PS: Without turning on the segment argument, the above script runs fine.

ArtRand commented 1 week ago

Hello @agoel13,

Sorry about that. Could you try the attached build and tell me if you continue to encounter the crash?

modkit_dev88c2107_centos7_x86_64.tar.gz

agoel13 commented 1 week ago

Many thanks. Completed successfully without any error msgs. When would this version be released? I would then update my pipeline with the released version.

ArtRand commented 1 week ago

@agoel13 Great. Yes this fix and a few others will be released this week. Thanks for testing it!

agoel13 commented 1 week ago

Thanks. Looking forward to it.

ArtRand commented 6 days ago

This fix has been released in v0.3.1.