Installation

GPMelt implementation is based on the Python (Van Rossum and Drake 2009) package GPyTorch (Gardner et al. 2018), and the full workflow has been encapsulated in a Nextflow (Di Tommaso et al. 2017) pipeline. This setup ensures simple installation, supports parallel computation, and offers a portable solution for deployment on both local computers and high-performance computing (HPC) clusters.

1 Installing Nextflow and Docker

Before we start, make sure Nextflow and Docker (or any containerization tool compatible with Docker containers) is installed either locally on your machine, or set up on the HPC cluster that you are planning to use.

  • Nextflow: Follow the installation instructions on the Nextflow website.

  • Docker: Download and install Docker from the Docker website.

2 GPMelt repository

GPMelt implementation is found on the following gitlab repository.

  1. Clone the repository:
git clone https://git.embl.de/grp-savitski/gpmelt.git
  1. Navigate to the Nextflow directory:
cd Nextflow

This directory contains the Nextflow pipeline GPMelt_workflow.nf (and the sub-pipelines: .nffiles ), the config files (.config) and the python code (in the bin directory). The dummy_data folder contains example datasets.

References

Di Tommaso, Paolo, Maria Chatzou, Evan W Floden, Pablo Prieto Barja, Emilio Palumbo, and Cedric Notredame. 2017. “Nextflow Enables Reproducible Computational Workflows.” Nature Biotechnology 35 (4): 316–19.
Gardner, Jacob, Geoff Pleiss, Kilian Q Weinberger, David Bindel, and Andrew G Wilson. 2018. “Gpytorch: Blackbox Matrix-Matrix Gaussian Process Inference with Gpu Acceleration.” Advances in Neural Information Processing Systems 31.
Van Rossum, Guido, and Fred L. Drake. 2009. Python 3 Reference Manual. Scotts Valley, CA: CreateSpace.