KmerAperture

Alignment-free estimation of core and accessory distances in closely related genomes, including SNPs within k of one another

Dependencies

KmerAperture is written in python3 and requires the following python packages:

• numpy
• screed
• pandas
• biopython

Setup

First set up a conda environment with the appropriate dependencies:

conda env create -f environment.yml
conda activate env-KmerAperture

Compile the kmer parser. This is recommended for speed

conda install ocaml -c conda-forge
ocamlopt.opt -O3 -o parser/KmerApertureParser parser/KmerApertureParser.ml -ccopt -static


#mamba can be used instead of conda:
mamba install ocaml

Alternatively KmerAperture can be run with --pyonly and will take longer to read the genomes

Usage

Run the main script KmerAperture referencing your fasta directory and reference genome (required):

KmerAperture --fastas <fasta dir> --reference <ref file fasta>

The following options are also available:

Flag	Short flag	Description	Required	Default val
--fastas	-f	Provide path to query fastas directory	✅
--reference	-r	Path and file name of fasta reference genome	✅
--kmersize	-k	k size		21
--pyonly	-py	Run without fast ocaml kmer parser		False
--proc	-p	Run with multiple processors		1

You may also precluster your genomes (cf below) if you suspect them of being relatively diverse (such as species-wide)

Input

A reference genome and a directory of assembled query genomes (fasta format)

Output

Filename	Description
full_align.align	Full genome SNP alignment. SNPs, invariant sites and query-gaps are included
accessory_coords.json	Coordinates of sequence accessory to each query genome compared with the reference
{referencename}_{k}.csv	Comma separated results for each genomes SNP and indel count compared with reference and accessory size

To use the core genome SNPs for a phylogeny with branch lengths corrected for recombination use BactCore, iqtree and ClonalFrameML:

BactCore full_align.fasta > core_alignment.fasta
iqtree -s core_alignment.fasta -B 1000
ClonalFrameML core_alignment.fasta.treefile core_alignment.fasta kmeraperture

They may first be installed with:

conda install -c conda-forge -c bioconda -c defaults clonalframeml
conda install -c bioconda iqtree
git clone https://github.com/moorembioinfo/BactCore.git
cd BactCore
g++ -std=c++11 -O3 -fopenmp BactCore.cpp -o BactCore

Pre-cluster (suspected) diverse genomes prior to KmerAperture

Install sourmash (https://github.com/sourmash-bio/sourmash):

conda install -c conda-forge -c bioconda sourmash

Run:

python sourmash_precluster.py --fastas <fasta dir> 

The output is a hierarchical dendrogram (UPGMA) of your genomes MinHash (Jaccard) distances. Select a clustering threshold that cuts across long branches (major lineages). Then cluster based on this threshold with:

python sourmash_precluster.py --fastas <fasta dir> --threshold <threshold>

Obtain the threshold for clustering by viewing the dendrogram and selecting a distance that cuts horizontally through blue lines only. You may also assess the reference placement with preclustering and consider a more appropriate reference for genome clusters if needed.

Cite

If you find KmerAperture useful please cite:

KmerAperture: Retaining k-mer synteny for alignment-free estimation of within-lineage core and accessory differences https://www.biorxiv.org/content/10.1101/2022.10.12.511870

@article {Moore2022.10.12.511870,
	author = {Moore, Matthew P and Laager, Mirjam and Ribeca, Paolo and Didelot, Xavier},
	title = {KmerAperture: Retaining k-mer synteny for alignment-free estimation of within-lineage core and accessory differences},
	elocation-id = {2022.10.12.511870},
	year = {2022},
	doi = {10.1101/2022.10.12.511870},
	publisher = {Cold Spring Harbor Laboratory},
	URL = {https://www.biorxiv.org/content/10.1101/2022.10.12.511870},
	eprint = {https://www.biorxiv.org/content/10.1101/2022.10.12.511870.full.pdf},
	journal = {bioRxiv}
}

If you use preclustering, please also cite sourmash: https://joss.theoj.org/papers/10.21105/joss.00027#

@article {Brown2016,  
         doi = {10.21105/joss.00027},  
	 url = {https://doi.org/10.21105/joss.00027},  
	 year = {2016},  
	 publisher = {The Open Journal},  
	 volume = {1}, number = {5},  
	 pages = {27},  
	 author = {C. Titus Brown and Luiz Irber},  
	 title = {sourmash: a library for MinHash sketching of DNA}, journal = {Journal of Open Source Software} }

Licence

Please note that the code for KmerAperture is distributed under the terms of the GNU GPL v3 license, for more details see https://www.gnu.org/copyleft/gpl.html