neoplasmer (ALPHA RELEASE)

Matches names of cancers and cancer-related conditions to ontology classes

Usage (via Docker)

Currently the recommended way to do this is on the command line via the neoplasmer.sh bash script:

./neoplasmer.sh tests/data/vicc_input.txt

You do not need to install anything other than Docker. The shell script is standalone.

The input file is a newline delimited list of terms. See tests/data/ for examples.

The first execution will take a minute or two; some warning messages may be printed, these can be ignored. Two directories will be created: RDF-Cache and .cache

Subsequent executions will be much faster

You can see example results in scratch/vicc-results.tsv

Python client

First, start a service:

./neoplasmer.sh --port 9055

Then install python libs:

pip3 install -r requirements.txt

Run the test client:

python3 bin/neoplasmer-client.py 'lung cancer' 'brain glioma'

TODO: additional documentation on how to this works

Running without docker

Install SWI-Prolog from http://www.swi-prolog.org

Run directly using bin/neoplasmer