This pipeline automates via snakemake the genome assembly of PacBio HiFi reads using Hifiasm. It takes BAM files from sequencing as input and will output a final primary assembly plus two haplotype assemblies (by default) and basic assembly QC metrics from QUAST.
Simply clone this repository from Github, after making sure git is installed on your machine! (Note: if you are using the Harvard Canon computing cluster, git is installed by default):
git clone https://github.com/harvardinformatics/pacbio_hifi_assembly.git
Then enter the directory: cd pacbio_hifi_assembly/
You will also need snakemake installed to run the pipeline. The easiest way to install snakemake is via Conda/Mamba (preferably Mamba), instructions for which can be found here.
In the repo directory, there is a file in the config/ subdirectory called config.yaml that you will need to modify to point towards your data. For a basic assembly, just change the following lines:
sample: "sample_name" #Name of the sample to act as base name
reads: " " #List of BAM files output from sequencer (include full path). If multiple files, separate by SPACES
You do not need to change any other options in config/config.yaml unless you are incorporating HiC data (note this is NOT the same as scaffolding with HiC!) or only want a primary assembly.
From the main directory, navigate into the workflow/ subdirectory, which contains the Snakefile that determines the order in which the pipeline runs. For running the assembly on the cluster, here is an example SLURM file to run from within the workflow directory:
#!/bin/bash
#SBATCH -N 1
#SBATCH -t 7-00:00:00
#SBATCH --mem 120G
#SBATCH -n 20
#SBATCH --partition shared
#SBATCH -J snakemake
source ~/anaconda3/bin/activate snakemake
snakemake -r --cores 20 --use-conda --rerun-incomplete
Resources needed will depend on the size and complexity of the genome, as will the time to complete. If successful, you should see a subdirectory workflow/results/ that contains the finished assembly and QC stats!