chewBBACA stands for "BSR-Based Allele Calling Algorithm". The "chew" part could be thought of as "Comprehensive and Highly Efficient Workflow" but at this point still it needs a bit of work to make that claim so we just add "chew" to add extra coolness to the software name. BSR stands for BLAST Score Ratio as proposed by Rasko DA et al.
chewBBACA is a comprehensive pipeline including a set of functions for the creation and validation of whole genome and core genome MultiLocus Sequence Typing (wg/cgMLST) schemas, providing an allele calling algorithm based on Blast Score Ratio that can be run in multiprocessor settings and a set of functions to visualize and validate allele variation in the loci.
chewBBACA performs the schema creation and allele calls on complete or draft genomes resulting from de novo assemblers.
The code contained in this repository concerning release v2.0.5. has been published in Microbial Genomics under the title:
chewBBACA: A complete suite for gene-by-gene schema creation and strain identification - Link to paper
When using chewBBACA please use the following citation:
Silva M, Machado M, Silva D, Rossi M, Moran-Gilad J, Santos S, Ramirez M, Carriço J. 15/03/2018. M Gen 4(3): doi:10.1099/mgen.0.000166
As from 25/01/2018 this repo is now fully on python 3 (tested on >=3.4). The previous python 2 version can be found at https://github.com/B-UMMI/chewBBACA/tree/chewbbaca_py2
As from 02/02/2018 chewBBACA is a python package, install with pip instead of git clone.
As from 02/02/2018 prodigal training files to be used are now on a separate repository. You can find them at https://github.com/mickaelsilva/prodigal_training_files.
Many Thanks to Stefano Morabito and Arnold Knijn (https://github.com/aknijn) for EURL VTEC in ISS, Rome ! https://toolshed.g2.bx.psu.edu/repository?repository_id=88fd7663075eeae9&changeset_revision=093352878303
- New alleles also have a timestamp added to the allele name.
- Corrected bug from 2.0.15 when no prodigal training file provided.
- Added prodigal training files to the package. They are now at CHEWBBACA/prodigal_training_files.
- when using the function
PrepExternalSchema, older behavior would remove any locus with a single translation error while the latest change(2.0.12) would not change the original source fasta, this would make the schema unusable. It is now enforced that the alleles that do not translate are removed from the fasta, be sure to backup your data before using this function.
- corrected bug when -h on allele call
- new option for the schema creation. A schema can be created based on a single fasta file, jumping the prodigal gene prediction running. Use
--CDSand provide a sinfle fasta file on the-iinput.
- cgMLST profile extraction function (ExtractCgMLST) more efficient (thanks Dillon Barker)
- new option for the allele call, size threshold previously hardcoded at 0.2 can now be changed using the
--stoption. Size threshold is important for the definition of ASM and ALM (alleles smaller/larger than mode).
- blast results during allele call are not saved as a file, instead are piped directly for processing
- new option for the allele call, if genome fasta input is already a fasta of CDS use the
--CDSoption
- corrected bug that prevented usage of latest blast version (>=2.7.0)
- version flag can now be used
--version - instead of calling the main script
chewBBACA.pyyou can now usechewie(if installed trought pip).
- AlleleCall : -i option accepts a single fasta file now
Check the wiki pages ...
...for a much more thorough chewBBACA walkthrough. Below you can find a list of commands for a quick usage of the software.
Use BBACA gitter...
... if you have any pressing question. Chat can be faster and better than email for troubleshooting purposes
Important Notes before starting:
- chewBBACA define an allele as a complete Coding DNA Sequence, with start and stop codon according to the NCBI genetic code table 11 identified using Prodigal 2.6.0 . It will automatically exclude any allele for which the DNA sequence does not contain start or stop codons and for which the length is not multiple of three.
- All the referenced lists of files must contain full path for the files.
- Make sure that your fasta files are UNIX format. If they were created in Linux or MacOS systems they should be in the correct format, but if they were created in Windows systems, you should do a a quick conversion using for example dos2unix.
An extensive tutorial repository ...
...is available as example on how to run an analysis pipeline using chewBBACA.
A ready to use docker image ...
...automatically built from the latest version of chewBBACA in Ubuntu 16.04.
Available training files ...
...use for a better result, species specific. Also inside the package now! e.g. --ptf Acinetobacter_baumannii.trn will now also work!
Installing chewBBACA
Install using pip
pip3 install chewbbaca
You need to install the following dependencies. Prodigal and BLAST must be added to the PATH variables.
Python dependencies:
- numpy>=1.14.0
- scipy>=0.13.3
- biopython>=1.70
- plotly>=1.12.9
- SPARQLWrapper>=1.8.0
- pandas>=0.22.0
Main dependencies:
- BLAST 2.5.0+ ftp://ftp.ncbi.nih.gov/blast/executables/blast+/2.5.0/ or above
- Prodigal 2.6.0 or above
Other dependencies (for schema evaluation only):
Create your own wgMLST schema based on a set of genomes fasta files. The command is the following:
chewBBACA.py CreateSchema -i ./genomes/ -o OutputFolderName --cpu 4
Parameters
-i Folder containing the genomes from which schema will be created. Alternatively a file
containing the path to the list of genomes. One file path (must be full path)
to any fasta/multifasta file containing all the complete or draft genomes you want to call alleles for.
-o prefix for the output folder for the schema
--cpu Number of cpus to use
--bsr (Optional) Minimum BSR for defining locus similarity. Default at 0.6.
--ptf (Optional but recommended, contact for new species) path to file of prodigal training file to use.
Outputs:
One fasta file per gene in the -o directory that is created.
The fasta file names are the given according the FASTA annotation for each coding sequence.
Optional:
Information about each locus is almost non existant at this point, the only information directly given by the schema creation is where are located each identified protein on the genome (proteinID_Genome.tsv file). A function was added to fetch information on each locus based on the uniprot SPARQL endpoint.
chewBBACA.py UniprotFinder -i schema_seed/ -t proteinID_Genome.tsv --cpu 4
Parameters
-i Folder containing the reference genes of the schema.
-t proteinID_Genome.tsv output from the schema creation
--cpu Number of cpus to use
Outputs:
A tsv file with the information of each fasta (new_protids.tsv), location on the genome, a name for which the protein sequence was submitted on uniprot and a link to that identified protein.
Then run is the following:
chewBBACA.py AlleleCall -i ./genomes/ -g genes/ -o OutPrefix --cpu 3
Parameters
-i Folder containing the query genomes. Alternatively a file
containing the list with the full path of the location of the query genomes.
-g Folder containing the reference genes of the schema. Alternatively a file
containing the list with the full path of the location of the reference genes.
-o prefix for the output directory. ID for the allele call run.
--cpu Number of cpus to use
-b (optional)Blastp full path. In case of slurm system BLAST version being outdated it may
be hard to use a different one, use this option using the full path of the updated blastp executable
--ptf (Optional but recommended, contact for new species) path to file of prodigal training file to use.
Outputs files:
./< outPrefix >_< datestamp>/< outPrefix >/results_statistics.txt
./< outPrefix >_< datestamp>/< outPrefix >/results_contigsInfo.txt
./< outPrefix >_< datestamp>/< outPrefix >/results_Alleles.txt
./< outPrefix >_< datestamp>/< outPrefix >logging_info.txt
./< outPrefix >_< datestamp>/< outPrefix >RepeatedLoci.txt
Usage:
chewBBACA.py TestGenomeQuality -i alleles.tsv -n 12 -t 200 -s 5 -o OutFolder
-i raw output file from an allele calling (i.e. results_Alleles.txt)
-n maximum number of iterations. Each iteration removes a set of genomes over the defined threshold (-t) and recalculates all loci presence percentages.
-t maximum threshold, will start at 5. This threshold represents the maximum number of missing loci allowed, for each genome independently, before removing it (genome).
-s step to add to each threshold (suggested 5)
-o Folder for the analysis files
The output consists in a plot with all thresholds and a removedGenomes.txt file where its informed of which genomes are removed per threshold when it reaches a stable point (no more genomes are removed).
Example of an output can be seen here . This example uses an
original set of 714 genomes and a scheme consisting of 3266 loci, using a parameter -n 12,-s 5 and -t 300.
Creating a clean allelic profile for PHYLOViZ
Clean a raw output file from an allele calling to a phyloviz readable file.
Basic usage:
chewBBACA.py ExtractCgMLST -i rawDataToClean.tsv -o output_folders
-i raw output file from an allele calling
-o output folder (created by the script if not existant yet)
-r (optional) list of genes to remove, one per line (e.g. the list of gene detected by ParalogPrunning.py)
-g (optional) list of genomes to remove, one per line (e.g. list of genomes to be removed selected based on testGenomeQuality results)
-p (optional) minimum percentage of loci presence (e.g 0.95 to get a matrix with the loci that are present in at least 95% of the genomes)
Create an html to help visualize your schema
See an example here
Basic usage:
chewBBACA.py SchemaEvaluator -i genes/ -ta 11 -l rms/ratemyschema.html --cpu 3 --title "my title"
-i directory where the genes .fasta files are located or alternatively a .txt file containing the full path for each gene .fasta file per line
-ta (optional) which translation table to use (Default: 11 in case of bacteria)
--title (optional) title to appear on the final html.
-l Location/name of the final html output
--cpu number of cpu to use, will be used for mafft and clustalw2
A: Depending on the variability of the strains used to create the schema and the number of CPUs you have selected, the computing time used will vary. The more variable the strains, the more BLAST comparisons will be made, meaning more time will be needed for finishing the analysis.
A: chewBBACA should allow you to continue where you stopped, just re-run the same command and you should be prompted to continue the allele call or use the flag --fc.
Q: I ran all the steps and my cgMLST loci size is smaller than traditional MLST, does this even work?
A: You probably forgot to eliminate from the analysis genomes responsible for a considerable loss of loci. Try to run again step 4, remove some of those genomes and check if the cgMLST loci number rises.
A: Yes. Be sure to have a single fasta for each locus and use the "PrepExternalSchema" function.
A: At the moment:
- Acinetobacter baumannii
- Campylobacter jejuni
- Enterococcus faecium
- Escherichia coli
- Haemophilus influenzae
- Legionella pneumophila
- Listeria monocytogenes
- Salmonella enterica enteritidis
- Streptococcus agalactiae
- Staphylococcus aureus
- Staphylococcus haemolyticus
- Yersinia enterocolitica
get them at https://github.com/mickaelsilva/prodigal_training_files
A: You can propose a new one to be added to the repository or create your own training files. To create a training file do:
prodigal -i myGoldStandardGenome.fna -t myTrainedFile.trn -p single
Silva M, Machado M, Silva D, Rossi M, Moran-Gilad J, Santos S, Ramirez M, Carriço J. 15/03/2018. M Gen 4(3): doi:10.1099/mgen.0.000166