ExpEvoAnalyzer

A snakemake pipeline to analyse experimental evolution data.

ExpEvoAnalyzer uses shovill for assembly, bakta for genome annotation, SKA2/bwa-mem + samtools/bcftools for variant calling, and SnpEff for variant annotation.

ExpEvoAnalyzer can either be run on assemblies or paired-end reads, and works with gzipped files.

ExpEvoAnalyzer filters variants caused by assembly errors by mapping reference reads back to the original assembly, and filtering any shared variants.

Dependencies:

• python>=3.9
• bakta
• ska2
• spades
• pandas
• biopython
• snakemake>=9.6
• snpeff
• shovill
• pyvcf
• gzip
• bwa
• samtools
• bcftools

To install:

Install the required packages using conda/mamba:

git clone https://github.com/samhorsfield96/ExpEvoAnalyzer.git
cd ExpEvoAnalyzer
mamba env create -f environment.yml
mamba activate expevoanalyzer

Running:

ExpEvoAnalyzer expects all reads to be trimmed (adapters removed) if provided. ExpEvoAnalyzer can also be run with assemblies.

Update config.yaml to specify workflow and directory paths.

• output_dir: path to output directory. Does not need to exist prior to running.
• input_dir: path to directory containing paired-end FASTQ files (gzipped or uncompressed).
• reference_reads_R1: path to FASTQ of read 1 for the reference isolate. Can be provided in absence of reference_assembly. If provided with a different reference_assembly, will be treated as and "ancestral" sequence, whereby any variants shared with it in the populaiton will be ignored.
• reference_reads_R2: path to FASTQ of read 2 for the reference isolate. Can be provided in absence of reference_assembly. If provided with a different reference_assembly, will be treated as and "ancestral" sequence, whereby any variants shared with it in the populaiton will be ignored.
• reference_assembly: path to FASTA of reference assembly (if already generated). Can be provided in absence of reference_reads_R1 and reference_reads_R1.
• reference_assembly_gbk: (OPTIONAL) path to Genbank file (gbff) of previously annotated genome. Must match reference_assembly if used.
• alignment_method: must be one of ska (aligning to closely-related reference) or bwa (aligning to distantly-related reference)
• shovill_params: additional parameters for shovill e.g. --trim
• ksize: integer, k-mer size used for SKA2
• snpEFF_config: path to SnpEff config file, usually scripts/snpEff.config
• bakta_db: path to bakta database. This needs to be downloaded and updated prior to running (see here)
• bakta_extra_settings: enables addition of extra bakta parameters e.g. adding your own trusted protein annotations using --protein file.fasta. This can be generated from a gbff using the script extract_fasta_from_gbff.py

Run snakemake:

snakemake --cores <cores>

Outputs:

All workflows output the following directories/files:

• shovill: assembly files of reference strain.
• bakta: bakta annotations of reference strain.
• reads_list: input files for SKA2.
• vcfs: VCF files generated by SKA2/bwa, non-reference read files mapped to reference assembly.
• snpeff: annotated VCFs generated by SnpEff.
• annotated_variants.csv: presence/absence matrix of variants in non-reference files, annotated by SnpEff
• logs: log files.

If running ska, you'll get an additional directory:

• ska_build: SKA2 sketches of non-reference read files.

If running bwa, you'll get an alternative additional directory

• bwa_mem: alignment files generated by bwa mem.