BejaLab fosmid assembly and annotation pipeline

This repository contains a Snakemake workflow for fosmid processing from reads or pre-assembled contigs.

Input data

The workflow reads config/samples.xlsx with columns:

• id: sample identifier (used in output paths).
• illumina1, illumina2: paired-end read files relative to raw/.
• nanopore: long-read file relative to raw/.
• pre_assembled: assembly FASTA relative to pre_assembled/.
• prefix: required; used as locus_tag_prefix and prepended to LOCUS in final GenBank.
• ref_organism: organism for PGAP submol.yaml (genus_species); defaults to Escherichia coli when empty.
• organism: organism string written into the final transformed .gbk.

Assembly source selection for each sample is:

1. illumina1/illumina2 -> SPAdes
2. otherwise nanopore -> Flye
3. otherwise pre_assembled

Workflow overview

Depending on available inputs, the workflow performs:

• read trimming for Illumina data
• vector filtering on reads using a middle region of the vector
• assembly (spades or flye)
• vector trimming from assembled contigs
• PGAP annotation
• taxonomy classification with Metabuli (metabuli classify)

Default final targets:

• output/{id}/{id}.gbk
• output/{id}/report.xlsx with sheets: Assembly, Annotation, Classification, Genes, Vector

How to create an instance

snakedeploy deploy-workflow https://github.com/BejaLab/fosmids my_new_fosmid_project --branch main

or

snakedeploy deploy-workflow https://github.com/BejaLab/fosmids my_new_fosmid_project --tag v1.0.0 # or another one of https://github.com/BejaLab/fosmids/tags

How to run

1. Fill config/samples.xlsx.
2. Place input files in raw/ and/or pre_assembled/ according to metadata.
3. Run Snakemake (profile enables Conda):

snakemake

snakemake --use-conda --cores 10