search

luntergroup / octopus

Bayesian haplotype-based mutation calling
MIT License
305 stars 38 forks source link

info on UL nanpore reads #267

Open gambalab opened 1 year ago

gambalab commented 1 year ago

Hi, If I use ultra long nanopore reads, are there any specific parameters I should use?

jelber2 commented 1 year ago

I do not think octopus was designed for this (not the Ultra-long part, rather the Nanopore part). Unless these are SUP-duplex reads, it does not have a high error rate long read preset. It does have some presets for low/moderate error rate PacBio Hifi (CCS) reads though.

gambalab commented 1 year ago

I tried for curiosity on several datasets and actually I always get this error at the end of the call

[2023-09-19 22:03:24] GL000214.1:129759 98.0% 1d 3h 32m 1s [2023-09-19 22:07:38] KI270435.1:312 98.1% 1d 3h 30m 23s [2023-09-19 22:13:39] KI270728.1:1872759 98.2% 1d 3h 28m 46s [2023-09-19 22:16:29] KI270736.1:103786 98.3% 1d 3h 27m 6s [2023-09-20 02:30:05] - 100% 1d 7h - [2023-09-20 02:30:13] Starting Call Set Refinement (CSR) filtering [2023-09-20 02:30:15] ------------------------------------------------------------------------ [2023-09-20 02:30:15] current | | time | estimated
[2023-09-20 02:30:15] position | completed | taken | ttc
[2023-09-20 02:30:15] ------------------------------------------------------------------------ terminate called after throwing an instance of 'octopus::hmm::HMMOverflow' terminate called recursively Aborted