cjprybol / Mycelia

MIT License
4 stars 3 forks source link

Mycelia

Multiomic analysis and data integration for biological characterization, prediction, and design.

Mycelia wraps existing best-in-class bioinformatics tools via Conda where existing solutions are available, and extends and integrates those tools with code written in Julia.

Designed for linux-based HPC and cloud systems

Install

Install Julia (if not already installed)

I have had trouble getting the visualization libraries Plots.jl and Makie.jl (and associated packages) to load correctly on HPC due to the complexities of the default LD_LIBRARY_PATH

I imagine other research supercomputer users may have similar issues, although I don't have these issues on cloud vendors like GCP or AWS

To enable Julia to install all of it's own necessary dependencies independent of the system, I reset the LD_LIBRARY_PATH variable prior to launching Julia !!

This can be done easily when launching Julia from the command line by

export LD_LIBRARY_PATH="" && julia

And can be done for Julia jupyter kernels by setting the env key => value pair in the appropriate kernel.json file

Clone the repo directly

cd /path/where/you/want/the/repo
# for production usage
git clone https://github.com/cjprybol/Mycelia.git
# for development
git clone git@github.com:cjprybol/Mycelia.git

Or as Julia package

import Pkg
# for production usage
Pkg.add(url="https://github.com/cjprybol/Mycelia.git")
# for development
Pkg.develop(url="git@github.com:cjprybol/Mycelia.git")

documentation in prep

https://doi.org/10.1101/2024.05.29.596541