Abricate
:mag_right: :pill: Mass screening of contigs for antimicrobial and virulence genes
Install / Use
/learn @tseemann/AbricateREADME
ABRicate
Mass screening of contigs for antimicrobial resistance or virulence genes. It comes bundled with multiple databases: NCBI, CARD, ARG-ANNOT, Resfinder, MEGARES, EcOH, PlasmidFinder, Ecoli_VF and VFDB, VICTORS, BacMet.
Is this the right tool for me?
- It only supports contigs, not FASTQ reads
- It only detects acquired resistance genes, NOT point mutations
- It uses a DNA sequence database, not protein
- It needs BLAST+ >= 2.7 and
any2fastato be installed - It's written in Perl :camel:
If you are happy with the above, then please continue! Otherwise consider using Ariba, Resfinder, RGI, SRST2, AMRFinderPlus, etc.
Quick Start
% abricate 6159.fasta
Using database resfinder: 2130 sequences - Mar 17, 2017
Processing: 6159.fna
Found 3 genes in 6159.fna
#FILE SEQUENCE START END STRAND GENE COVERAGE COVERAGE_MAP GAPS %COVERAGE %IDENTITY DATABASE ACCESSION PRODUCT RESISTANCE
6159.fna NC_017338.1 39177 41186 + mecA_15 1-2010/2010 =============== 0/0 100.00 100.000 ncbi AB505628 n/a FUSIDIC_ACID
6159.fna NC_017338.1 727191 728356 - norA_1 1-1166/1167 =============== 0/0 99.91 92.367 ncbi M97169 n/a FOSFOMYCIN
6159.fna NC_017339.1 10150 10995 + blaZ_32 1-846/846 =============== 0/0 100.00 100.000 ncbi AP004832 betalactamase BETA-LACTAM;PENICILLIN
Installation
Bioconda
If you use Conda follow the instructions to add the Bioconda channel:
conda install -c conda-forge -c bioconda abricate
abricate --check
abricate --list
Source
If you install from source, Abricate has the following package dependencies:
any2fastafor sequence file format conversion- BLAST+ >2.7.0 for
blastn,makeblastdb,blastdbcmd - Perl modules:
LWP::Simple,Bio::Perl,JSON,Path::Tiny git,unzip,gzipfor updating databases
Most of these are easy to install on an Ubuntu-based system:
sudo apt-get install bioperl ncbi-blast+ gzip unzip git \
libjson-perl libtext-csv-perl libpath-tiny-perl liblwp-protocol-https-perl libwww-perl
git clone https://github.com/tseemann/abricate.git
./abricate/bin/abricate --check
./abricate/bin/abricate --setupdb
./abricate/bin/abricate ./abricate/test/assembly.fa
Input
Abricate takes any sequence file that any2fasta can convert to FASTA files (eg. Genbank,
EMBL), and they can be optionally gzip or bzip2 compressed.
abricate assembly.fa
abricate assembly.fa.gz
abricate assembly.gbk
abricate assembly.gbk.bz2
It can take multiple files at once too:
abricate assembly.*
abricate /mnt/ncbi/bacteria/*.gbk.gz
Or you can provide it a "file of file names" (a "FOFN"):
% cat test/fofn.txt
assembly.fa
assembly.fa.gz
assembly.gbk
assembly.gbk.bz2
% abricate --fofn test/fofn.txt
It does not accept raw FASTQ reads; please use Ariba or SRTS2 for that.
Output
Abricate produces a tap-separated output file with the following columns:
Column | Example | Description
-------|---------|------------
FILE | Ecoli.fna | The filename this hit came from
SEQUENCE | contig000324 | The sequence in the filename
START | 23423 | Start coordinate in the sequence
END | 24117 | End coordinate
STRAND | + | Strand + or -
GENE | tet(M) | AMR gene name
COVERAGE | 1-1920/1920 | What proportion of the gene is in our sequence
COVERAGE_MAP | =============== | A visual represenation of the hit. ==aligned, .=unaligned, /=has_gaps
GAPS | 1/4 | Openings / gaps in subject and query - possible psuedogene?
%COVERAGE | 100.00% | Proportion of gene covered
%IDENTITY | 99.95% | Proportion of exact nucleotide matches in the covered part
DATABASE | ncbi | The database this sequence comes from
ACCESSION | NC_009632:49744-50476 | The genomic source of the sequence
PRODUCT | aminoglycoside O-phosphotransferase APH(3')-IIIa | Gene product (if available)
RESISTANCE | TETRACYCLINE;FUSIDIC_ACID | putative antibiotic resistance phenotype, ;-separated
Caveats
- Does not find mutational resistance, only acquired genes.
- Gap reporting incomplete
- Sometimes two heavily overlapping genes will be reported for the same locus
- Possible coverage calculation issues
Databases
ABRicate comes with some pre-downloaded databases:
You can check what you have installed with the --list command.
This lists the available databases in TSV (or CSV with --csv) and three
columns:
% abricate --list
DATABASE SEQUENCES DBTYPE DATE
ecoh 597 nucl 2025-Apr-14
bacmet2 746 nucl 2025-Apr-14
ncbi 7456 nucl 2025-Apr-14
victors 4545 nucl 2025-Apr-14
vfdb 4394 nucl 2025-Apr-14
megares 7424 nucl 2025-Apr-14
card 6048 nucl 2025-Apr-14
plasmidfinder 488 nucl 2025-Apr-14
resfinder 3174 nucl 2025-Apr-14
argannot 2224 nucl 2025-Apr-14
ecoli_vf 2701 nucl 2025-Apr-14
The default database is ncbi.
You can choose a different database using the --db option:
% abricate --db vfdb --quiet 6159.fa
6159.fna NC_017338.1 2724620 2726149 aur 1-1530/1530 =============== 0/0 100.00 99.346 vfdb NP_647375 zinc metalloproteinase aureolysin
6159.fna NC_017338.1 2766595 2767155 icaR 1-561/561 =============== 0/0 100.00 98.930 vfdb NP_647402 N-acetylglucosaminyltransferase
6159.fna NC_017338.1 2767319 2768557 icaA 1-1239/1239 =============== 0/0 100.00 99.677 vfdb NP_647403 n/a
6159.fna NC_017338.1 2768521 2768826 icaD 1-306/306 =============== 0/0 100.00 99.020 vfdb NP_647404 n/a
6159.fna NC_017338.1 2768823 2769695 icaB 1-873/873 =============== 0/0 100.00 99.542 vfdb NP_647405 n/a
6159.fna NC_017338.1 2769682 2770734 icaC 1-1053/1053 =============== 0/0 100.00 98.955 vfdb NP_647406 n/a
6159.fna NC_017338.1 2771040 2773085 lip 1-2046/2046 =============== 0/0 100.00 98.778 vfdb NP_647407 triacylglycerol lipase precursor
Combining reports across samples
ABRicate can combine results into a simple matrix of gene presence/absence.
An absent gene is denoted . and a present gene is represented by its '%COVERAGE`.
This can be individual abricate reports, or a combined one.
# Run abricate on each .fa file
% abricate 1.fna > 1.tab
% abricate 2.fna > 2.tab
# Combine
% abricate --summary 1.tab 2.tab
#FILE NUM_FOUND aac(6')-aph(2'')_1 aadD_1 blaZ_32 blaZ_36 erm(A)_1 mecA_15 norA_1 spc_1 tet(M)_7
1.tab 8 100.00 100.00 .
