card-trick 0.2.0

Utility package to find gene - drug relationships within CARD

Card trick

A software package that extracts information from the ontology relationships from The Comprehensive Antibiotic Resistance Database (CARD) https://card.mcmaster.ca/home

Original code: https://gitlab.com/cgps/card_trick

Forked version code: https://gitlab.com/JFsanchezherrero/card_trick

This forked version retrieves all information for each entry available in CARD ontology information.

Installation

This package requires Python3

pip3 install card-trick

Usage

There are two main modules: search and update

usage: card-trick [-h] [-v] {update,search} ...

positional arguments:
  {update,search}  The following commands are available. Type card_trick
                   <COMMAND> -h for more help on a specific commands
    update         Get latest CARD ontology
    search         search CARD ontology

optional arguments:
  -h, --help       show this help message and exit
  -v, --version    display the version number
  --man            Additional information

Download ontology database

usage: card-trick update [-h] [--path PATH] [-q QUIET]

optional arguments:
  -h, --help            show this help message and exit
  --path PATH           Path to store CARD ontology.
  -q QUIET, --quiet QUIET
                        Do not print process information

The update command card-trick update will download the latest version of the database. If already exists in the path provided (or default) it would update it.

Database in obo format is stored in your home directory in the .card-trick directory or in the directory stated using option --path

Obo format file would be parsed into a csv file for later analysis.

All information for each entry in CARD ontology database are stored in the csv file.

Search ontology database

usage: card-trick search -i INPUT -t {ARO,gene,antibiotic,target,any} 
                         [-f {json,csv,tsv,all}] [-o OUTPUT_NAME] [-p PATH] [-b]
                         [-i2 INPUT_2] [t2 {ARO,gene,antibiotic,target,any}] [-b2] [-q] [-h]

optional arguments:
  -h, --help            show this help message and exit

  -i INPUT, --input INPUT 
                        Input term to search in CARD ontology. Provide a unique 
                        term o several using --batch option

  -f {json,csv,tsv,all}, --format_output {json,csv,tsv,all}
                        Output format. csv, tsv or json

  -o OUTPUT_NAME, --output_name OUTPUT_NAME
                        Output name.

  -p PATH, --path PATH  
                        Path containing CARD ontology. Default is user’s home 
                        directory.

  -t {ARO,gene,antibiotic,target,any}, --term {ARO,gene,antibiotic,target,any}
                        The type of term provided to search.

  -b, --batch           Provide this option if input is a file containing
                        multiple terms, one per line.

  -i2 INPUT_2, --input_2 INPUT_2
                        Input term to search in results retrieved from first
                        input. Provide a unique term o several using --batch
                        option

  -t2 {ARO,gene,antibiotic,target,any}, --term_2 {ARO,gene,antibiotic,target,any}
                        The type of term provided to search for the second
                        input. Default: any

  -b2, --batch_2        Provide this option if input_2 is a file containing
                        multiple terms, one per line.

  -q, --quiet           Do not print process information

The command card-trick search will return matches as a dataframe and print into a default tabular file.

There are several possibilities to search the dataframe. This module can search for:

• gene entries
• ARO terms
• antibiotic to which a gene confers resistance
• antibiotic to target a gene

No output is print to STDOUT as card-trick would retrieve all information from CARD ontology database and might be too much information.

There is a possibility to generate a batch search of multiple terms using option batch.

Also, there is a possibility to link several searches: Use option --input_2, --term_2 or --batch_2 e.g. CTX genes that have a resistance to ceftazidime

Examples

## Different term options:
card-trick search -t antibiotic -i tigecycline
card-trick search -t target -i tigecycline
card-trick search -t gene -i ctx
card-trick search -t ARO -i ARO:3003032
card-trick search -t any -i ctx

## Using path database provided
card-trick search -t gene -i ctx --path /folder/to/card_ontology

## Batch example
card-trick search -t ARO --batch -i batch_entry_file.txt

## Multiple search
card-trick search -t gene -i ctx --path /folder/to/card_ontology -i2 ceftazidime -t2 antibiotic
card-trick search -t gene -i ctx --path /folder/to/card_ontology --batch_2 -i2 file.txt -t2 any 

## Output
card-trick search -t ARO --batch -i batch_entry_file.txt -f tsv -o example_name

Using in a python script or program

Here is a simple code example to download, parse and search the database.

If the database is download and parsed as a csv, just load the information to a pandas dataframe

import card_trick
import pandas as pd

CARD_folder=/path/to/your/database/ontology/card

## uptade database in a path
aro_obo_file = card_trick.ontology_functions.update_ontology(CARD_folder, False)

## get ontology and save it in csv
card_trick.ontology_functions.parse_ontology(aro_obo_file, False)

## load information
csv_file = CARD_folder + '/aro.obo.csv'
card_ontology = pd.read_csv(csv_file, sep=',', index_col=0)

## search for examples IDs
AROS_identified = ('ARO:3000026', 'ARO:3004058')
term = 'ARO'
information_ontology = card_trick_caller.get_info_CARD(AROS_identified, term, card_ontology)