A pipeline to annotate genome assemblies automatically using Earlgrey and Braker4, complete with functional annotation (Interpro)

This pipeline is designed to be run on the Polar2020 cluster. The parameters you need to fill in are managed via a custom YAML config file. The pipeline automatically routes the workflow based on the files you provide, so you don't need to pass any complicated mode flags.

Note: do not modify or edit nextflow.config unless you know what you're doing. These contain the hardcoded dependency paths (Dfam, Interpro, Singularity caches, etc.) specific to Polar2020.

1. Copy the template: copy params_template.yaml to params.yaml.
2. Fill in the paths to your genome assembly, masked assembly (optional), and path to a directory with all rna-seq reads (optional). If you are skipping a file (e.g., you don't have RNA reads), leave it as null.
3. Run with Nextflow, passing your config file with -params-file.

nextflow run artorias111/FeatureFlow -params-file params.yaml
# you can name params.yaml to whatever you'd like, this is an example

FeatureFlow dynamically builds the pipeline based on what you put in params.yaml.

• Full Pipeline: Provide genome_assembly, protein_ref, and rna_reads. The pipeline will mask the genome with Earlgrey, run Braker4 with both RNA and protein evidence, and finish with functional annotation.
• Skip Masking: Provide a masked_genome alongside your genome_assembly. The pipeline will skip Earlgrey and feed your provided masked genome directly into Braker4.
• Protein-Only Mode: Leave rna_reads as null or empty. Braker4 will automatically fall back to protein-only evidence for structural annotation.

Results are executed in the work directory, but you have access to clean symlinks of the actual files organized in the results directory, so you're not lost in the sea of hex-coded directories in work.

• Earlgrey (https://github.com/TobyBaril/EarlGrey)
• Braker4 (https://github.com/Gaius-Augustus/BRAKER4)
• InterProScan (https://github.com/ebi-pf-team/interproscan)
• AGAT (https://github.com/NBISweden/AGAT)
• BUSCO (https://gitlab.com/ezlab/busco)