Genome assembly and variant calling from PacBio HiFi reads

This folder contains two Snakemake workflows:

• assembly: from PacBio HiFi consensus reads to annotated genome assemblies
• variantcalling: combine assemblies into a pangenome graph and call variants from the graph

This is ongoing work, some things will change.

Requirements

On the sciCORE cluster, the pipeline is installed in the GROUP folder (/scicore/home/gagneux/GROUP/PacbioSnake) and ready to run.

In other contexts, four things need to be set up before the pipeline can be run:

1. Install Snakemake and Singularity
2. Build the singularity container for the assembly pipeline
3. Download the bakta database for genome annotation
4. Pull the singularity container for the variant calling pipeline

These steps are detailed below.

1. Install Snakemake and Singularity

As described on the Snakemake and the Singularity sites.

2. Build the singularity container for the assembly pipeline

cd assembly/container
sudo singularity build assemblySC.sif assemblySC.def

3. Download the bakta database for genome annotation

Light-weight (1.3 Gb) and full (33.1 Gb) databases for the bakta annotation tool can be downloaded from https://zenodo.org/records/7669534. The extracted folder should be located at assembly/resources/bakta_db/. Otherwise the path to the database can be modified in the assembly config file.

4. Pull the singularity container for the variant calling pipeline

singularity pull docker://ghcr.io/pangenome/pggb:latest