Pacbio microbial genome pipeline

Overview

This pipeline takes in raw (.bam) PacBio sequel data and does the following:

• Creates .fasta files from raw reads

• Generates assembly using the Flye assembler

• Reorients assembly to start at DNAa using circlator fixstart

• Aligns reads to assembly with PacBio MiniMap2 and polishes with the Arrow algorithm in pbgcpp

• To include:

  • PGAP:

    • Annotation
    • ANI calculation / taxonomy checking

  • GTDB taxonomy check

Details can be found in the Snakefile in project root directory which defines the pipeline.

Installation

• To use this pipeline, clone the git repository and add raw .bam files to data/bam/
• SnakeMake is used to run the pipeline and can be installed with conda install -c bioconda snakemake (if you don't have miniconda, it must be installed first).
• Mamba is a faster version of conda that can speed up the installation of other required software. It can be installed with conda install -c conda-forge mamba and then be used as a replacement for conda.
• To install both snakemake and mamba at the same time (recommended) you can use conda install -c bioconda -c conda-force snamemake mamba.
• If you would like to create a separate conda environment for these packages, you could instead use conda create -n {insert environment name you want here} -c bioconda -c conda-force snamemake mamba, which must be activated with conda activate {env name} before use.

With conda working and snakemake installed, the other needed packages are installed automatically. This is designed to be run on UD's BioMix HPC, and a snakemake profile with instructions specific to it's use is included.

• Simply use this command to run the pipeline efficiently on BioMix and install any necessary packages: snakemake --profile slurm.
• If you installed mamba, you have to tell snakemake to use it with --conda-fontend mamba, the full command is snakemake --profile slurm --conda-frontend mamba.
• You may also have to tell snakemake to be patient and wait a little longer for files to appear/sync across the BioMix filesystem with --latency-wait {20, or some other reasonable number of seconds of your choosing}.

The polished assembly can be found at data/assembly/{strain}_flye/{strain}_polished.fasta

The annotated assembly can be found at data/annotation/{strain}/{strain}.gbk