hiruna72 / squigualiser

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

squigualiser/docs /RNA_visualisation.md #60

Closed vetmohit89 closed 2 months ago

vetmohit89 commented 3 months ago

Dear @hiruna72 ,

I want to try squigualiser for identifying RNA modifications. I am wondering if you can provide the commands to run. I see you have provided a readme file for RNA (https://github.com/hiruna72/squigualiser/blob/main/docs/RNA_visualisation.md). It would be convenient if you could also provide the parameters used for each command to replicate your example data.

Thank you Mohit

hiruna72 commented 3 months ago

Hello @vetmohit89,

Thanks for trying squigualiser. A simpler pipeline script with an RNA dataset is available here.

A more complex pipeline was used for the above documentation. It is available here https://doi.org/10.5281/zenodo.10896390 You may first read the run.sh scripts and inspect the RUN01 output directories before trying it yourself. You are more likely to face file path could not be found errors as this was tried on my system. However, with some careful edits to the script you should be able to replicate the output.

vetmohit89 commented 3 months ago

Dear @hiruna72,

I am trying to run individual tools one by one.

Here, it states: SIGNAL_FILE="reads.blow5"

buttery-eel -g ${GUPPY} --config ${MODEL_TO_USE} -i ${SIGNAL_FILE} -o ${BASECALL_DIR}/moves.sam --log ${OUTPUT_DIR} --moves_out --port 5558 --use_tcp

Will this also work for Fast5 files? In the example, it appears that a Fast5 file was used (https://github.com/hiruna72/squigualiser/blob/main/docs/RNA_visualisation.md). Does it make any difference if we use Fast5 versus Slow5?

Thank you Mohit

hiruna72 commented 3 months ago

Hi @vetmohit89,

You can use any fast5/pod5/slow5 format for basecalling. All will generate same result. slow5 and pod5 will be faster.

However, I suggest you to use slow5 as you will later need a slow5 for plotting. You can easily create a slow5 as mentioned here. https://github.com/hiruna72/squigualiser/issues/58#issuecomment-2002705907

vetmohit89 commented 3 months ago

Hello @hiruna72,

I have converted fast5 to slow5 using slowtools. Now, trying to run buttery-eel command to convert fast5 to sam file. But earlier it was giving error FileNotFoundError: [Errno 2] No such file or directory: './guppy/ont-guppy/bin/dorado_basecall_server' But then I installed ont-dorado-server, it is giving error: basecall_service::SocketDataHandler::initialize: Failed to load config ./guppy/ont-guppy/data/rna_r9.4.1_70bps_hac_prom.cfg. Selected configuration does not support basecalling in dorado

vetmohit89 commented 3 months ago

Here is my command buttery-eel -g ./guppy/ont-guppy/bin/ --config rna_r9.4.1_70bps_hac_prom.cfg -i ./slow5/ -o ./slow5/moves.sam --log ./slow5/ --moves_out --port 5558 --use_tcp --device cuda:all

Psy-Fer commented 3 months ago

Hey,

Is your data RNA002 or RNA004?

If it's RNA002 you might need to use an older basecaller as the latest dorado server I think deprecated compatibility. Maybe even try the last guppy version, 6.5.7

Let me know if you need help with any of that (I wrote buttery-eel)

James

vetmohit89 commented 3 months ago

It is RNA002. But I am not able to understand why it was asking dorado_basecall_server and giving error No such file. Right now I am using guppy 6.3.7, I doubt newer versions of guppy (anything after 6.4.2) has --move_out option.

Psy-Fer commented 3 months ago

What I mean is you can run any version of guppy or Dorado sever inside buttery-eel with the slow5 file.

You just need to ensure your ont-pyguppy-client-lib library version matches the version of guppy or Dorado you are pointing to in the command.

This is covered in the documentation of buttery-eel

hiruna72 commented 2 months ago

Hey, I will close the issue for now. Feel free to reopen if you face any issues.