Skip to content

Installation

This guide covers installing the aa-tRNA-seq pipeline and its dependencies.

Prerequisites

  • Operating System: Linux (tested on CentOS/RHEL 9, Ubuntu 20.04+)
  • Python: 3.10+
  • GPU: NVIDIA GPU with CUDA support (required for basecalling and charging classification)
  • Storage: ~50GB for tools, models, and test data

Install Pixi

The pipeline uses Pixi for environment management.

Bash
1
curl -fsSL https://pixi.sh/install.sh | sh
Bash
1
brew install pixi
PowerShell
1
powershell -ExecutionPolicy Bypass -c "irm -useb https://pixi.sh/install.ps1 | iex"

After installation, restart your shell or run:

Bash
1
source ~/.bashrc  # or ~/.zshrc for Zsh

For additional installation options, see the official Pixi installation guide.

Clone the Repository

Bash
1
2
git clone https://github.com/rnabioco/aa-tRNA-seq-pipeline.git
cd aa-tRNA-seq-pipeline

Install Dependencies

Install all Python dependencies via Pixi:

Bash
1
pixi install

This creates a .pixi directory with all required packages including:

  • Snakemake 8.0+
  • pysam
  • pandas
  • pod5
  • remora
  • samtools
  • bwa
  • deeptools

Install External Tools

The pipeline requires several external tools. Install them with a single command:

Bash
1
pixi run setup

This downloads and installs:

  • Dorado - Oxford Nanopore basecaller (version set in config/config-base.yml)
  • Dorado model - rna004_130bps_sup@v5.3.0 basecalling model
  • Remora - ONT signal analysis for charging classification
  • WarpDemuX - Barcode demultiplexing (optional, for multiplexed samples)

Dorado and models are installed to resources/tools/ and resources/models/. Modkit is managed by pixi (installed via conda).

Download Test Data (Optional)

To run the test pipeline, download the test dataset:

Bash
1
pixi run dl-test-data

This downloads ~1GB of test POD5 files to .tests/.

Verify Installation

Verify everything is installed correctly:

Bash
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
# Check Snakemake version
pixi run snakemake --version

# Check Dorado installation
resources/tools/dorado/*/bin/dorado --version

# Check Modkit installation (managed by pixi)
pixi run modkit --version

# Dry run with test config
pixi run dry-run

Directory Structure After Installation

Text Only
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
aa-tRNA-seq-pipeline/
├── .pixi/                    # Pixi environment (includes modkit, remora)
├── resources/
│   ├── tools/
│   │   ├── dorado/<version>/  # Dorado binaries
│   │   └── WarpDemuX/       # WarpDemuX (if demux enabled)
│   ├── models/
│   │   ├── rna004_130bps_sup@v5.3.0/  # Basecalling model
│   │   └── cca_classifier.pt          # Remora charging model
│   ├── ref/                  # Reference sequences
│   └── kmers/               # Kmer level tables
├── .tests/                  # Test data (if downloaded)
├── workflow/                # Snakemake workflow
├── config/                  # Configuration files
└── cluster/                 # Cluster profiles

Updating

To update the pipeline:

Bash
1
2
git pull
pixi install  # Update dependencies if pixi.lock changed

To update external tools, modify the version in config/config-base.yml and rerun:

Bash
1
pixi run setup

Troubleshooting

Pixi Installation Issues

If Pixi fails to install, ensure you have:

  • curl installed
  • Write permissions to ~/.pixi
  • Internet access to download packages

GPU Not Detected

If Dorado fails to detect GPU:

  1. Check CUDA is installed: nvidia-smi
  2. Verify CUDA_VISIBLE_DEVICES is set correctly
  3. Ensure GPU drivers are up to date

Remora Installation Issues

If Remora fails to install with CUDA/PyTorch errors:

Bash
1
2
# Manually specify CUDA version
CUDA_VERSION=cu121 pixi run setup

Next Steps