Utilities for molecular prior-knowledge processing

ID translation, taxonomy, reference lists, orthologous genes, and more. Also available as a web service: https://utils.omnipathdb.org/

For most users: use omnipath-client

If you just want to translate IDs, resolve organisms, or access reference lists, install the lightweight client:

pip install omnipath-client

from omnipath_client.utils import map_name, translate_column
map_name('TP53', 'genesymbol', 'uniprot')  # {'P04637'}

The client queries the utils.omnipathdb.org web service -- no database setup required, same API as the local package.

Install omnipath-utils directly only if you need to run the service locally or build the database yourself.

Quick Start

from omnipath_utils.mapping import map_name, map_names
from omnipath_utils.taxonomy import ensure_ncbi_tax_id
from omnipath_utils.reflists import is_swissprot

# Translate gene symbol to UniProt
map_name('TP53', 'genesymbol', 'uniprot')
# {'P04637', ...}

# Translate multiple
map_names(['TP53', 'EGFR'], 'genesymbol', 'uniprot')
# {'P04637', 'P00533', ...}

# Resolve organism
ensure_ncbi_tax_id('human')   # 9606
ensure_ncbi_tax_id('hsapiens') # 9606

# Check if reviewed
is_swissprot('P04637')  # True

Installation

pip install omnipath-utils

With database and web service:

pip install "omnipath-utils[server]"

Web Service

omnipath-utils build --organisms 9606
omnipath-utils serve --port 8082

curl "http://localhost:8082/mapping/translate?identifiers=TP53,EGFR&id_type=genesymbol&target_id_type=uniprot"

Documentation

https://saezlab.github.io/omnipath-utils

License

The package is under the GNU GPLv3 license. This doesn't affect the web service and data, where each original resource carries its own license, and is potentially available for commercial use.