FastANI usage

The most practical protocol to calculate ANI values | Nanzhen Qiao | Dr. Michael Gänzle’s lab | 20250313

ANI (average nucleotide identity)

An ANI value of 95% (94–96%) has been almost consistently used in recent years to describe new bacterial species.

The final figure you will get 👉

Ref: https://doi.org/10.1128/aem.01034-23

Protocol 🧑‍🔧

Step 1. Prepare the genomes you will calculate.

Better format them as .fna files.

Make sure the number of genomes is correct.

ls -l | grep "^-" | wc -l

Step 2. Generate the genome list.

ls | grep ".fna$" > list.txt

Step 3. Run FastANI.

If you need to install FastANI:

conda create --name fastANI
conda activate fastANI
conda install bioconda::fastani

#if conda is not wroking, try mamba
mamba install fastANI

# check if fastANI is working
fastANI -h

fastANI --ql list.txt --rl list.txt -t 4 -o all_all_fastani_output.txt
# -t, threads
# --ql, query_list
# --rl, reference_list
# -o, output_file

Step 4. Turn the output file into a matrix.

Need two files here: list.txt & all_all_fastani_output.txt.

Run fastani_output_to_matrix.py(a Python script), and the matrix will be outputted as fastani_matrix.csv.

import pandas as pd
import numpy as np

list_file = './list.txt'
fastani_file = './all_all_fastani_output.txt'
output_matrix_file = './fastani_matrix.csv'

# Load genome list
def load_genomes(list_file):
    with open(list_file, 'r') as f:
        genomes = [line.strip() for line in f]
    return genomes

# Load FastANI data and construct matrix
def construct_matrix(fastani_file, genomes):
    genome_index = {genome: i for i, genome in enumerate(genomes)}
    matrix_size = len(genomes)
    fastani_matrix = np.full((matrix_size, matrix_size), np.nan)  # Initialize with NaN

    with open(fastani_file, 'r') as f:
        for line in f:
            cols = line.strip().split('\t')
            if len(cols) < 3:
                continue
            g1, g2, ani_value = cols[0], cols[1], float(cols[2])
            if g1 in genome_index and g2 in genome_index:
                i, j = genome_index[g1], genome_index[g2]
                fastani_matrix[i, j] = ani_value
                fastani_matrix[j, i] = ani_value  # Ensure symmetry

    # Fill diagonal with 100.0 (self-comparison)
    np.fill_diagonal(fastani_matrix, 100.0)
    return pd.DataFrame(fastani_matrix, index=genomes, columns=genomes)

# Save matrix for Excel (CSV format)
def save_matrix(matrix, output_file):
    matrix.to_csv(output_file, sep=',')

if __name__ == '__main__':
    genomes = load_genomes(list_file)
    fastani_matrix = construct_matrix(fastani_file, genomes)
    save_matrix(fastani_matrix, output_matrix_file)
    print(f"FastANI matrix saved to {output_matrix_file}")

Step 5. Integrate the matrix into the phylogenetic tree on iTOL.

• Log into iTOL (https://itol.embl.de/itol.cgi) and make sure the phylogenetic tree is ready.
• Download the spreadsheet to edit big phylogenetic trees easily (https://itoleditor.letunic.com/download/iTOL_annotation_editor_v1_8_Excel.xlsm).
• Log in → select the tree you are working on → create a new dataset → choose Heatmap → Label the dataset and pick up a legend color → paste the data.
• If needed, sort the data on the iTOL website.

Step 6. 🎉 🎈

🎉