🧬 De Novo Assembly Project Template

This repository provides a standardized and reproducible template for de novo genome assembly projects. It is designed to facilitate consistent project organization, environment management, and seamless synchronization between local machines and High-Performance Computing (HPC) environments.

---

📂 Project Structure

A clean, consistent directory layout is key for reproducible science. This template uses the following structure:

Directory	Description
analyses/	Execution hub. Holds main workflow definitions (e.g., Nextflow, Snakemake) that launch the assembly and analysis pipelines.
code/	Ad-hoc scripts. Contains utility scripts, custom R/Python analysis scripts, and smaller, non-workflow code.
data/	Input & Output. Stores raw data (reads) and all generated results (contigs, alignment files, etc.).
docs/	Documentation. Project-specific notes, reports, lab book entries, diagrams, or detailed protocol write-ups.
envs/	Custom Environments. Definitions for custom software environments (e.g., Dockerfiles, Singularity recipes) if not using pixi.

---

🛠️ Tools & Environments

This template leverages pixi for dependency management and command execution, ensuring your environment is fully reproducible across systems.

• pixi: Used to manage all project dependencies (via pixi.toml) and define convenient, portable tasks.
• Git: Essential for version control and file synchronization between your local machine and HPC.

---

💻 Seamless HPC Integration

This setup allows you to easily sync code and configurations between your local machine (e.g., laptop) and your HPC cluster using standard git and pixi commands.

Initial Setup

1. Clone on both systems: Use git clone to get the repository on both your local machine and your HPC cluster from Github.

2. Link your local repository to your HPC: From your local machine's repository, execute one of the following commands to create a special git remote named hpc:

   • Direct git remote command:

     git remote add hpc user@cluster:/path/to/cloned/repo/on/hpc

   • Using the pixi task:

     pixi run git-link-hpc user@cluster:/path/to/cloned/repo/on/hpc

Tip: If you've configured an alias for your HPC cluster in your ~/.ssh/config file, you can replace user@cluster with your shorter alias (e.g., hpc-alias).

Synchronization

Once linked, you can use standard Git commands and pixi tasks to move data and code.

• To fetch remote changes (e.g., results generated on the HPC):

  git fetch hpc

• For simplified syncing (see pixi.toml for task details):

  pixi run sync-to-hpc  # e.g., to push local data to the cluster
  pixi run fetch-report # e.g., to pull results from the cluster

Check the pixi.toml file to see how the hpc remote is used in these custom tasks.