gCons

A desktop tool for genomic consensus analysis across multiple FASTA genomes.

gCons identifies common zones — genomic regions shared by a configurable proportion of input genomes — using k-mer based comparison. Results are exported as a consensus FASTA file and visualised interactively.

---

How it works

gCons chains two external tools:

1. redoak finds k-mers shared across genomes and reports how many genomes contain each one.
2. gkampi maps each common k-mer to its exact position within every genome.

From this positional data, gCons builds chains of consecutive k-mers that appear consistently across genomes (controlled by the α and β thresholds), which it calls common zones. Each zone is exported as a nucleotide sequence anchored to the reference genome.

FASTA files
    │
    ▼
redoak ──► shared k-mer list
    │
    ▼
gkampi ──► positional index (per genome)
    │
    ▼
Zone detection (α, β filtering)
    │
    ├──► ResulFasta.fasta   (consensus sequences)
    ├──► resultat.csv       (zone positions)
    └──► Interactive plot   (Tkinter/matplotlib)

---

Requirements

• Python 3.10+
• redoak and gkampi binaries in the working directory
• tkinter (see note below)

pip install -r requirements.txt

macOS (Homebrew): tkinter requires a separate install.

brew install python-tk@3.x   # replace 3.x with your Python version

Ubuntu/Debian:

sudo apt install python3-tk

---

Usage

python gCons_controleurTK.py

Fill in the four fields and click Run analysis:

Field	Description
FASTA file paths	Space-separated paths. The first file is the reference genome.
k-mer size	Length of the k-mers used for comparison (e.g. 21).
Alpha (%)	Minimum percentage of genomes that must share a k-mer for it to be considered common.
Beta (%)	Minimum percentage of genomes that must share a consecutive k-mer link for a zone to be extended.

Example:

FASTA files  →  genome1.fasta genome2.fasta genome3.fasta
k-mer size   →  21
Alpha        →  80
Beta         →  60

This finds k-mers present in at least 80% of genomes, then chains them into zones where at least 60% of genomes share consecutive links.

---

Outputs

File	Description
ResulFasta.fasta	Consensus sequences, one entry per common zone, annotated with reference coordinates and scaffold membership.
resultat.csv	Raw zone positions in the reference genome (semicolon-separated).
commun.txt	Intermediate redoak output filtered by alpha.
result<n>.csv	Raw gkampi positional output for genome n.

---

Project structure

gCons/
├── gCons_controleurTK.py   # UI entry point
├── gCons_fonctionsTK.py    # Pipeline logic and visualisation
├── requirements.txt
├── redoak                  # External binary (not versioned)
├── gkampi                  # External binary (not versioned)
└── data/
    └── example/            # Sample FASTA files for testing

---

Parameters — choosing α and β

Both thresholds are expressed as a percentage of the total number of input genomes and are rounded to the nearest integer.

• High α (e.g. 90%): only k-mers nearly universal across all genomes are retained. Yields fewer, more conserved zones.
• Low α (e.g. 50%): includes k-mers present in just half the genomes. Yields more zones, potentially less conserved.
• β ≤ α is the natural constraint: a chain link cannot be more prevalent than the k-mer itself.

---

Internship context

This tool was developed during a research internship at LIRMM (Laboratoire d'Informatique, de Robotique et de Microélectronique de Montpellier) as part of a genomic analysis pipeline for inter-genome comparison using k-mer methods (redoak, gkampi).