search

hiruna72 / squigualiser

Visualise and analyse nanopore (ONT) raw signals
https://hiruna72.github.io/squigualiser/
MIT License
100 stars 1 forks source link

Reform parameters (-k & -m) for RNA002 libraries? #48

Closed lkwhite closed 7 months ago

lkwhite commented 9 months ago

Hi again, still familiarizing myself with the tool and have another question. For RNA libraries, you have this nice walkthrough. I see you've got some preset profiles for running reform.py on DNA libraries, but not for RNA. What parameters did you use for this RNA data set? I'm guessing from the pore model documentation that your offset -m is 1, but I'm less clear on where the kmer length -k in these profiles is derived from.

I also see you have a table with -k 5 and -m -3 here (at least for forward mapped reads, which is all I have). Anyway, just hoping you might have some guidance before I start testing a bunch of options.

hiruna72 commented 9 months ago

Hello @lkwhite,

  1. Only reads mapped to a transcriptome are supported for RNA. Hence, no reverse mapped reads https://github.com/hiruna72/squigualiser#plot-conventions.
  2. As listed here if you use rna_r9.4.1_70bps_fast_prom model from guppy you can set -k 1 and -m 0. Otherwise you have to use calculate offsets program to derive the k and m values. What is the basecalling model you are planning to use?
  3. Sorry I will document this better. However the idea is if you get the -k and -m values at reform stage correct, then you don't have to worry about base shift parameter at the plotting stage (for input coming through reform).
  4. You can use this pipeline script and the blow5 file to get started. Please note that this script is a guide only and might not work in the first go. You have to execute the script sitting inside the script's directory. The details of different variables and functions of the script are listed here

Thanks for bringing up these questions. I will improve the documentation. Let me know how it goes.

lkwhite commented 9 months ago

Hi @hiruna72, These are reads base called with rna_r9.4.1_70bps_hac_prom, so sounds like we might need to calculate our own offsets. Thanks for the quick response and pointers!

hiruna72 commented 9 months ago

Hi @lkwhite,

When I tried the script myself I encountered a bug which I fixed in the newest version. There is a new pip version released today https://pypi.org/project/squigualiser/0.5.0/. Or download the linux or macos version from the release

I updated the profiles table to include rna_r9.4.1_70bps_hac_prom and it has -k1 and -m0 values.

hiruna72 commented 7 months ago

Hi @lkwhite,

Thank you for using squigualiser. Hope your issues are resolved. Please feel free to reopen the issue should you need further help.