gws_microbial_genomics

🔍 Introduction

Bakta is a powerful tool for the rapid and standardized annotation of bacterial genomes, MAGs and plasmids, from both isolates and metagenome-assembled genomes (MAGs).

It provides dbxref-rich, sORF-inclusive, taxon-independent annotations in machine-readable formats such as JSON, GFF3, GenBank, EMBL, TSV, and FASTA, ensuring compatibility with downstream workflows.

Unlike protein-only functional annotators (e.g., EggNOG-mapper), Bakta is a full annotation pipeline, comparable to Prokka, DFAST, and PGAP, capable of:

• Predicting CDS (coding DNA sequences) and non-coding RNAs (tRNA, rRNA, tmRNA, ncRNA)
• Detecting CRISPR arrays and origins of replication (oriC/V)
• Adding functional descriptions and stable cross-references to major databases (RefSeq, UniRef100, UniParc), facilitating FAIR-compliant and reproducible analyses.

This makes Bakta a complete solution for researchers working with bacterial genome annotation, comparative genomics, and downstream bioinformatics pipelines.

🧰 Prerequisites

• Access to Constellab and a valid Digital Lab environment
• Installed bricks: gws_microbial_genomics version ≥ 0.1.1
• Input file: A genome assembly in FASTA format (contigs, plasmids, MAGs) Bakta Database: A pre-downloaded Bakta DB (db-full, db-light, or db) generated using Build/Update Bakta Databasetask.

🧪 Use Case Steps

1. Import your genome FASTA into Constellab.
2. Link it to the Task: "Procaryotes Genome Annotation".
3. Configure Parameters: prefix: Output prefix (default: FASTA stem). genus, species, strain: (optional) Organism metadata. translation_table: Choose genetic code (default: The Bacterial, Archaeal and Plant Plastid Code, NCBI 11). replicon_type & replicon_topology: Apply to all contigs if desired (e.g., plasmid + circular). complete_genome: Mark sequences as complete (optional). threads: Number of CPU threads to allocate.
4. Run the Task:

📂 Output

Bakta produces a set of standardized files for downstream use:

✅ Example Use Cases

• Annotating new bacterial isolates before submission to NCBI/ENA.
• Adding functional context to MAGs in metagenomic studies.
• Comparing plasmid vs chromosome content.
• Generating publication-ready genome maps.

🧬 Comparative Summary: Bakta vs eggNOG-mapper

EXAMPLE:

With Bakta

• Input: ecoli_contigs.fna
• Outputs: ecoli.gff3, ecoli.gbff, ecoli.faa, ecoli.ffn, etc.
• What you get in practice:

-            “On contig_12, from 10543 to 11890: a CDS named gyrA”

-            “On contig_3: a tRNA-Leu gene”

-            A GenBank file you can use for comparison, submission, and visualization.

With eggNOG-mapper (after Bakta)

• Typical input: ecoli.faa (the proteins predicted by Bakta)
• Output: annotation.tsv
• What you get in practice:

-            For the gyrA protein: functional and ontology assignments such as COG category, GO terms, EC number (if applicable), KEGG pathway (e.g., DNA replication), etc.

In short: you move from “here is the gene in the genome” to “here is what it does and which pathways it belongs to.”