A targeted approach to assess the plasmidome of bacteria.
If you use this tool, please cite : G. Royer, J.-W. Decousser, C. Branger, M. Dubois, C. Médigue, E. Denamur, D. Vallenet. PlaScope: a targeted approach to assess the plasmidome from genome assemblies at species level. Microbial Genomics, 2018 Sep;4(9).
And don't forget the publications related to its dependencies :
- Bankevich A, Nurk S, Antipov D, Gurevich AA, Dvorkin M et al. SPAdes: a new genome assembly algorithm and its applications to single-cell sequencing. J Comput Biol 2012;19:455–477
- Kim D, Song L, Breitwieser FP, Salzberg SL. Centrifuge: rapid and sensitive classification of metagenomic sequences. Genome Res 2016;26:1721–1729
This method enables you to classify contigs from a WGS assembly according to their location (i.e. plasmid or chromosome). It is based on a smart tool called Centrifuge, initially developed as a metagenomic classifier. We propose here an application on E. coli plasmidome, with a specific database build on one hand on completely finished genomes of E. coli from the NCBI, and on the other hand on a custom plasmid database. In fact 3 databases of plasmid have been merged together :
- plasmids used to create
PlasmidFinder(http://aac.asm.org/content/58/7/3895.long) - plasmids proposed by Orlek et al. (https://www.sciencedirect.com/science/article/pii/S2352340917301567?via%3Dihub)
- plasmids from Branger et al., Extended-spectrum ß-lactamase-genes are spreading on a wide range of Escherichia coli plasmids existing prior the use of third generation cephalosporins, Microbial Genomics, 2018, In press
We also propose a Klebsiella database that has been evaluated on a clinical dataset of 12 Klebsiella pneumoniae strains.
We think that this method can easily be applied to other bacterial species since you have got enough reference data (e.g. Staphylococcus aureus, Enterococcus sp. ...).
You must install these dependencies before you start :
- SPAdes 3.10.1 or later if you want to run the assembly (= mode 1) (header of contigs must be the same as in version 3.10.1, e.g >NODE_1_length_506801_cov_117.065)
- Centrifuge 1.0.3
PlaScope is essentially a wrapper script (called plaScope.sh) around SPAdes and Centrifuge.
It's written in bash and awk and should work on Linux and Mac OS X both with GNU awk and BSD awk.
To install it, simply download the sources and decompress them.
Don't forget to add the location of plaScope.sh to your PATH.
The easiest way to install PlaScope and its dependencies is through BioConda.
Once you have created and activated a conda environment, simply type:
$ conda install plascopeNote that several versions of awk are available in conda so you can further control the environment.
Another way to use PlaScope is through the IFB Cloud.
Just create an account and launch the PlaScope appliance.
You can choose between two modes:
- Mode 1: SPAdes assembly then contig classification
- Mode 2: contig classification only (if you already assembled your genome with SPAdes or Unicycler)
$ ./plaScope.sh -h
usage: plaScope.sh [OPTIONS] [ARGUMENTS]
General options:
-h, --help display this message and exit
-v, --version display version number and exit
-n, --no-banner don't print beautiful banners
-t number of threads[OPTIONAL] [default : 8]
-o output directory [OPTIONAL] [default : current directory]
--sample Sample name [MANDATORY]
--db_dir path to centrifuge database [MANDATORY]
--db_name centrifuge database name [MANDATORY]
Mode 1: SPAdes assembly + contig classification
-1 forward paired-end reads [MANDATORY]
-2 reverse paired-end reads [MANDATORY]
Mode 2: contig classification of a fasta file (only if you already have your SPAdes or Unicycler assembly!)
--fasta SPAdes assembly fasta file [MANDATORY]
-a Specify the assembler used: spades or unicycler. Default=spades.
Example mode 1:
plaScope.sh -1 my_reads_1.fastq.gz -2 my_reads_2.fastq.gz -o output_directory --db_dir path/to/DB --db_name chromosome_plasmid_db --sample name_of_my_sample
Example mode 2:
plaScope.sh --fasta my_fastafile.fasta -o output_directory --db_dir path/to/DB --db_name chromosome_plasmid_db --sample name_of_my_sample -a unicycler
Github:
https://github.com/GuilhemRoyer/PlaScopePlaScope uses a database (see this section) made of 3 files.
The argument --db_dir is the path to the directory where these 3 files are located.
The argument --db_name is the common part between the file names (see examples).
To get the E. coli database, please download the following file on Zenodo: https://doi.org/10.5281/zenodo.1311641
After extracting the tar.gz file, you will have 3 files : chromosome_plasmid_db.1.cf, chromosome_plasmid_db.2.cf and chromosome_plasmid_db.3.cf.
All these files are required for PlaScope.
In this case, the --db_name to use is "chromosome_plasmid_db".
To get the Klebsiella database, please download the following file on Zenodo: https://doi.org/10.5281/zenodo.1311647
After extracting the tar.gz file, you will have 3 files : Klebsiella_PlaScope.1.cf, Klebsiella_PlaScope.2.cf and Klebsiella_PlaScope.3.cf.
All these files are required for PlaScope.
In this case, the --db_name to use is "Klebsiella_PlaScope".
This database has not been extensively benchmarked. We only have assessed its performances by searching for plasmids and resistance genes location on a set of 12 Klebsiella pneumoniae strains from https://academic.oup.com/jac/article/73/7/1796/4966148.
You can also design your own database as explained on this page.
You need to prepare four files:
- database.fna : a multifasta file of your database
>Chromosome_1
ATGGATAAGTTGCTGAACAAAAAGAT......
>Chromosome_2
GAGTGAACGGATGAAACAGAAAGACC......
>Plasmid_1
TCTCGAATGATAAAGGCTATGATGGC......
- nodes.dmp : an artificial taxonomy (i.e. root, chromosome, plasmid)
1 | 1 | root
2 | 1 | chromosome
3 | 1 | plasmid
- seqid_to_taxid.map : a mapping file between the sequences and their taxonomic assignment
Chromosome_1 2
Chromosome_2 2
Plasmid_1 3
- names.dmp : a file mapping taxonomy IDs to a name
1 | root | | |
2 | chromosome | | |
3 | plasmide | | |
Then, build your database as follow:
centrifuge-build -p 10 --conversion-table seqid_to_taxid.map --taxonomy-tree nodes.dmp --name-table names.dmp database.fna chromosome_plasmid_db
- Guilhem Royer (CEA-Genoscope, now at Pasteur): design, implementation, evaluation
- David Valllenet (CEA-Genoscope): design
- Julian Paganini (UMC Utrecht): new feature: accept unicycler assemblies