Comparative analysis of microbial genomes on Oxford nanopore

Tutorial

Create the params.json file describing your task, following the example

{
    "sample_files": [
     "wt.fastq.gz",
     "mutant.fastq.gz"
    ],
    "exclude_samples": [],
    "consensus_ref_sample": "wt",
    "output_dir": "data/",
    "medaka_model": "r1041_e82_400bps_sup_v4.2.0",
    "medaka_variant_model": "r1041_e82_400bps_sup_variant_v4.2.0",

    "ref_fa": "data/ref.fasta",
    "ref_gff": "data/ref.gff",
    "ref_id": "REF"
}

Then run the pipeline as

nextflow run https://github.com/BioinfoSupport/nanopore_compare.git -params-file params.json -profile local

You can use a preconfigured profile for SLURM executor by

nextflow run https://github.com/BioinfoSupport/nanopore_compare.git -params-file params.json -profile baobab

Pipeline structure

For each pair of sample (FASTQ, or unaligned BAM/CRAM) and reference (which is either a supplied reference or a consensus sequence of one of the designated samples) the overall logic is the following

Then this calling pipeline is applied to each pair of provided sample and reference, or sample and consensus sequence of a selected WT sample as reference

Possible optional parameters

    "gene_annotation_bed": "data/saureus/Saureus8325_gene_annotations.bed"

This parameter allows to use gene annotations that do not follow from the ref_gff file provided. Use this in case the pipeline fails at GFF_TO_GENE_BED step

    "regions_bed": "regions.bed"

This will add also a manual note to teh final VCF and CSV files. You can save regions from IGV and use it here. Fully optional.

Output structure

TODO list

• Make a simpler/correct annotate consensus vcf by qualities
• Symmetric calling on consensus
  • Potential attempt in medaka_compare project
• Liftover cases (is it needed at all?)
  • Problematic case: SNP coincides with original reference, how to represent this