Extract COG Sequences

This script extracts nucleotide and amino acid sequences corresponding to a list of COGs (Clusters of Orthologous Groups) from annotated microbial genomes. It works with Prokka-annotated genomes, eggNOG-mapper output, and genome FASTA files, and organizes the extracted sequences into per-COG folders.

---

Features

• Parses .emapper.annotations files from eggNOG-mapper to identify gene-to-COG mappings.
• Matches genes to coordinates in Prokka GFF files and extracts nucleotide and amino acid sequences.
• Organizes outputs into folders per COG (with gene name included).
• Supports different overwrite policies:
  • ask: prompts for each existing file
  • always: always overwrite existing files
  • never: never overwrite
  • dry-run: simulate what would be done, without writing anything
• Logs all steps to cog_extraction_log.txt and prints progress to terminal.

---

Required Input Files

1. Genome annotations (Prokka):

   ./prokka/<genome_id>/<genome_id>.gff
   ./prokka/<genome_id>/<genome_id>.faa
   

2. eggNOG-mapper outputs:

   ./eggnog/*.emapper.annotations
   

3. Genome nucleotide FASTA files (.fna):

   ./genomes_selected/*.fna
   

4. COG list (cog_gene_list.txt)
   A tab-delimited file with:

   • a required column cog
   • and an optional column gene (used for folder naming)

   Example:

   cog	gene
   COG0605	Superoxide_dismutase
   COG2032	CU/Zn_SOD
   COG2077
   

   • If gene is missing or empty, the COG ID will be used as the gene name.
   • This file must exist in the same directory as the script.

5. (Optional) Genome list (genome_list.txt)
   If present, this file restricts the analysis to the listed genomes.

   • Format: plain text, one genome ID per line.

   • Example:

     GCA_00012345.1
     GCA_00067890.1
     

   • If this file is not present, the script will use all genome folders in the Prokka directory.

---

Output

• Nucleotide sequences:

  ./cog_sequences_nt/COG0605_Superoxide_dismutase/GCA_00012345_COG0605.fna
  

• Amino acid sequences:

  ./cog_sequences_aa/COG0605_Superoxide_dismutase/GCA_00012345_COG0605.faa
  

• Log file:

  cog_extraction_log.txt
  

---

Usage

python Extract_COG_sequences.py [ask|always|never|dry-run]

• If no argument is passed, the default is ask.

---

Examples

Run using all genome folders (default behavior):

python Extract_COG_sequences.py always

Run using a genome ID list file:

1. Create a genome_list.txt file in the same directory as the script:

   GCA_00012345.1
   GCA_00067890.1
   

2. Then run the script:

   python Extract_COG_sequences.py ask

Perform a dry run to preview actions:

python Extract_COG_sequences.py dry-run

---

Dependencies

• Python 3.6+

• Biopython (pip install biopython)

• A properly structured input directory tree as described above

• Date: July 2025
  Institution: Okinawa Institute of Science and Technology (OIST)
  Author: Fatima Li-Hau (with assistance from ChatGPT)

Citation

If you use this code in your research, please cite it as:

Li-Hau, F. (2025). Extract COG Sequences Tool (Version 1.0) [Computer software]. Okinawa Institute of Science and Technology. https://github.com/microfafa-gh/Utilities/

Please also consider citing the tools used in this workflow, such as:

• Cock et al., 2009. Biopython: freely available Python tools for computational molecular biology and bioinformatics. Bioinformatics.
• Seemann, T. 2014. Prokka: rapid prokaryotic genome annotation. Bioinformatics.
• Huerta-Cepas et al., 2017. eggNOG-Mapper: functional annotation of orthologs. Molecular Biology and Evolution.

---