PlasCliques allows the profiling and comparison of plasmid content between bacterial isolates. It does this by assessing the presence of discrete groups of plasmid genes (cliques), found to be consistantly observed together over a large diverse dataset of plasmids in Enterobacteriaceae (Tokolyi *et al*., in preparation). With the provided database, draft bacterial genomes can have their plasmid cliques profiled using the findCliques.py script, and their output visually compared. If the user wishes to create their own clique database, createCliques.py can be used.

Install

Download and run locally after ensuring the documented dependencies are installed.

git clone https://github.com/atokolyi/PlasCliques.git

Discovering plasmid cliques in bacterial isolates

This script allows the discovery of plasmid cliques in bacterial assemblies. For Enterobacteriaceae, the provided database can be used as input, otherwise a custom clique database can be created using the scripts in the next section.

Quick usage

This will output a list of plasmid cliques in each isolate.

./findCliques.py -c EnteroCLDB -a assemblies_*.fasta

Heatmap visualisation

Using the grid output option allows us to export a matrix of the cliques present in each isolate, which can then be used to create a summary visualisation (.html) of the output.

./findCliques.py -c EnteroCLDB -a assemblies_*.fasta -b pc_out.grid
Rscript misc/makeHeatmap.R pc_out.grid

Command line options

Short	Long	Description (bold=required)
-v	--version	Print version number and exit
-h	--help	Print command line options and exit
-a	--assemblies	Assembled bacterial contigs as input (.fasta)
-c	--cliquesDB	Folder containing database of cliques
-s	--minCliqueSize	Minimum size (integer) of cliques included, default=2
-m	--match	Match file to match isolate backbones against a DB of large plasmids (.match)
-l	--outLevel	Integer level of output verbosity, 0: Cliques (default), 1: Previous + categories, 2: Previous + genes in cliques, 3: Previous + gene descriptions
-b	--bulkOut	For matrix output, specify an output file (.grid)

Creating a custom database of cliques

While PlasCliques comes with a database of cliques discovered in Enterobactereaceae plasmids, you may want to add additional plasmids, investigate a different bacterial family or perhaps have in mind an altogether different use of gene cliques (e.g. in phage, virus, human etc.). This script allows you to discover co-occurring clusters of genes (cliques) in your isolates. The output of this script is a clique database, which can then be used as input to findCliques.py, to discover the presence of these cliques in novel isolates.

A brief overview of the clique finding process is that we extract genes from the isolate assemblies, cluster these genes them by similarity, optionally add annotations to the gene clusters, detect the presence of these clusters in the original isolates to create a presence/absence matrix and then use this to calculate co-occurrence similarities between genes. This yeilds a network from which we can extract highly co-occurring groups of genes (cliques).

Quick usage

This will create folder containing a clique database (EnteroCLDB) based on the co-occurrence of the clustered genes (gene_clusters.clstr) in the investigated genomes (genomes_*.ffn).

./createCliques.py -g genomes_*.ffn -d gene_clusters.clstr -c EnteroCLDB

Command line options

Short	Long	Description (bold=required)
-v	--version	Print version number and exit
-h	--help	Print command line options and exit
-d	--cdhit	The cd-hit cluster file (.clstr)
-g	--genome	Prokka output of genomes (.ffn)
-c	--cliquesDB	Output folder containing database of discovered cliques
-j	--threshold	Jaccard inclusion threshold for COG co-occurrence weights, default=0.9
-p	--cpus	Number of parallel processes, default=1
-a	--annot	Database of gene annotations and categories (format: gene**\tcategoryID\tdescription\n**)

Dependencies

Software versions in use, these or later should work fine.

R (3.5.2)

• heatmaply (0.15.2)
• htmlwidgets (1.3)

install.packages('heatmaply')
install.packages('htmlwidgets')

python3 (3.6.6)

• Biopython (1.72)
• tqdm (4.26.0)
• joblib (0.12.5)
• python-igraph (0.7.1.post6)

pip3 install biopython
pip3 install tqdm
pip3 install joblib
pip3 install python-igraph