Metabee

This repository contains all the scripts and data used to perform a meta-analysis on the bacterial species diversity of the gut microbiome of Apis mellifera. The input data corresponds to publicly available metagenomes and metagenomes recovered by fellow members of the Engel lab at UNIL.

Installation

Requirements:

• conda v23.10.0
• mamba v1.5.6
  The following tools installed in their own conda environment:
• snakemake v7.32.4
• R v.4.3.2
  • r-ape v5.7-1
  • r-dendextend v1.17.1
  • r-ecodist v2.1.3
  • r-ggplot2 v3.5.0
  • r-gplots v3.1.3.1
  • r-tidyverse v2.0.0
  • r-rmarkdown v2.26
• sra-tools v3.0.10
• ncbi-datasets v16.4.4
• kraken2 v2.1.3 with the PlusPFP database downloaded and unpacked + custom database for the bee gut
• fastqc v0.11.8
• multiqc v1.6
• bbmap v39.01
• seqkit v.2.6.1
• motus v3.1
• spades v3.15.5
• megahit v1.2.9
• simka v1.5.3
• bowtie2 v2.5.2
• quast v5.2.0
• samtools v1.10 + msamtools v1.1.3 + bedtools v2.31.1
• metabat2 v2.15 with merge_depths.pl added to bin (see /envs)
• checkm v1.2.2 with the database downloaded, decompressed and dearchived. The path to this database must be put in the environment yaml file at the end, so that it is stored in an environment variable.
• gtdb-tk v2.3.2 with the database downloaded and decompressed. The path to this database must be put in the environment yaml file at the end, so that it is stored in an environment variable.
• dRep v3.4.5
• skani v0.2.1
• inStrain v.1.8.0 with parse_stb.py (run pip install drep --upgrade)

All environment YAML files can be found in /envs.

Usage

Note: the resources directive in each rule of the snakefile is written for execution on the slurm cluster of UNIL.

1. Create a directory with the following sub-directories:

.
│
└─── benchmarks
│   
└─── data
│
└─── logs
│   
└─── results
│
└─── workflow

2. Clone this repository in /workflow.
3. Create all conda environments using the YAML files in /workflow/envs.
4. Download and uncompress all required databases (CheckM, GTDB-Tk).
5. Download the kraken custom database and add it to /data.
6. Download raw sequencing data from NCBI using download_reads.sh.
7. Sort and list files with mv_files.sh.
8. Zip all downloaded files with zip_all.sh.
9. Download reference genomes with download_refgenomes.sh.
10. Move reference genomes for host filtering with mv_refgenomes.sh.
11. Download and add adapter sequences to /data/adapters.
12. Activate your snakemake conda environment and run each part of the pipeline written in different snakefiles. If snakemake raises issues with conflicting versions when creating the conda environments, it helps to (temporarily) set the channel priority to flexible.

Publication and authors