Home Pre-workhop Instructions
Lectures
1: Introduction and Key Concepts 2: Variant Detection and Phylogenetic Trees 3: Practical Applications of WGS and Phylogenetics 4: Advanced Applications of WGS
Practical 1: WGS Data Analysis
Obtaining sequencing data Viewing raw sequencing data QC of FASTQ files Cleaning and filtering FASTQs Reference mapping/alignment De novo assembly
Practical 2: Variant Calling and ML Trees
Variant calling and VCFs Building consensus sequences Aligning consensus sequences Maximum Likelihood trees
Practical 3: Timed Phylogenetic Trees
ML timed trees Bayesian trees (BactDating) Bayesian trees (BEAST2)
Practical 4: Transmission and Profiling
Identifying lineages and serotypes Transmission network reconstruction
Practical 5: Mixed Infection, Recombination and ANI
Identifying mixed infection Detecting recombination Average nucleotide identity (ANI)
Practical 6: Phylogeography and Phylodynamics
Phylogeography Inferring population dynamics
Practical 7: Fitness and Selection
Local branching index (LBI) Detecing homoplasy dN/dS GWAS


Average Nucleotide Identity

The average nucleotide identity (ANI) measures the mean sequence identity of all shared orthologous regions between two genomes. It is widely used to define species boundaries in prokaryotes, with a typical cutoff of ~95–96% ANI indicating the same species.

ANI provides higher resolution than 16S rRNA comparisons and is computed using tools like fastANI or pyANI, though these can be quite complex to run. It also can give you a more formal measure how close an assembly is to known reference strains, and potentially whether they are different species.

Here we will use a custom function in Python (computeANI.py) to calculate the ANI from an assembly to different reference strains that have been provided to you.

The data we will use for this exercise are:


  1. Install the required dependencies into your conda environment.
    2. conda install -c bioconda blast
pip install biopython

    These should already be downloaded but we want to make sure they are available to run the next command.

  2. We can now calculate the ANI between our unknown assembly and each reference strain, one at time, with the following command (this will calculate ANI for the unknown assembly and the H37Rv M. tuberculosis strain):
    3. python computeANI.py Unknown.fasta H37Rv.fasta

Run the analysis again for each of the reference strains.

Questions

  1. Which species is the unknown sample most likely to belong to?

  2. Can you gain any more insights from this analysis about the relatedness among all the species tested?


This is the end of the activities in practical session 5. Navigate back to the homepage for other activities here.