Usage

pycircdb is a command-line tool driven by a JSON configuration file. The general invocation pattern is:

pycircdb -c <config.json> <command> [options]

Subcommands can be chained together so that the input circRNAs are looked up once and reused across multiple steps, for example:

pycircdb -c config.json annotate fasta mirna rbp

Global options

-c, --config: Path to the JSON config file containing workflow parameters. Required before any subcommand.
-v, --verbose: Verbosity level: 0 (silent), 1 (high-level, default), 2 (all outputs).
-h, --help: Show the help message and exit.

The configuration file

The config file defines global parameters and the samples to process. A minimal example:

{
    "global_parameters": {
        "max_tasks": 1,
        "output_dir": "results/ci_test/"
    },
    "samples": {
        "ci_test_sample": {
            "file_path": "test/tester.txt",
            "reference": "hg19",
            "zero_based": true
        }
    }
}

The samples dictionary is required. Each sample points to a file of circRNA coordinates and declares its genome reference (e.g. hg19) and whether the coordinates are zero_based.

Commands

annotate

Annotate circRNAs using the configured databases.

pycircdb -c config.json annotate -d arraystar,circbase,circatlas

-d, --database: Comma-separated list of databases to use. Valid options: arraystar, circatlas, circbank, circbase, circpedia, circrna_db, cscd, exorbase.

fasta

Output circRNA sequences in FASTA format.

pycircdb -c config.json fasta -d arraystar,circbase

-d, --database: Comma-separated list of databases to use. Valid options: arraystar, circatlas, circbank, circbase, circpedia, circrna_db, cscd.

mirna

Output miRNA interactions for the identified circRNAs.

pycircdb -c config.json mirna -a miRanda,PITA,TargetScan

-a, --algorithm: Comma-separated list of algorithms to use. Valid options: miRanda, PITA, TargetScan.

rbp

Output RNA-binding protein (RBP) interactions for the identified circRNAs.

pycircdb -c config.json rbp