cannot install epi2me in the cluster!?

[shankhanath]

Ask away!

I am continuously trying to install epi2me in the server using conda (centos based) but it is not happening. I have downloaded the Ubuntu / Debian version from here https://labs.epi2me.io/downloads/
The installation guideline for linux is also here https://labs.epi2me.io/installation/#installation-on-linux
But this says to use sudo apt. I dont have the administrator permission for sudo. hence i am using conda install. but it is repetitively failing. i have tried creating an environment first which is also not helping. i have tried installing using mamba .. this is also not helping.

My objective:
I want to convert fast5 files obtained from gridion system to fastq. what should I do. I do not have a 64gb ram windows machine.

[shankhanath]

Please help in this issue. I have a big cluster with 128 gb ram. but i do not have a windows system with 64 gb ram. i want to install this in my cluster.

[cjw85]

Hi @shankhanath,

I apologise for the lack of response here. In order to install the deb file you will need administrator privileges. There is unfortunately no way around this. You will need also administrative privileges in order to setup docker if it is not setup already.

You may alternatively install and use the Dorado basecaller directly.

[shankhanath]

HI @cjw85
Thanks for your reply. I am seeing dorado converts pod5 to fastq. But I have Fast5 files. do you have any suggestions for that?

[cjw85]

This is described in the documentation: https://github.com/nanoporetech/dorado?tab=readme-ov-file#performance-tips