Ontology Mapping Exercise

This repository is a programming exercise whose aim it is to map ontology terms and ids when given an existing ontology file. Documentation will exist in this README, and a Jupyter Notebook within the folder /documentation/ will walk through the whole exerecise. The standalone script is called exercise.py

Original Exercise Prompt

• The task is to write a script / a set of scripts in Python 3.x that takes an ontology file as an input and provides a table with mappings for each concept to terms from specified ontologies.

The script(s) should do the following steps:

1. Read in the provided sample turtle file
2. Load into an appropriate data structure
3. Extract IDs and Preferred Labels (preferably using SPARQL) from concepts only in the “skin cancer” branch!
4. For each concept, retrieve mappings (using APIs)
5. Write the combined result into a tab delimited file, in which you list the mappings you retrieved for the different target ontologies using the different methods. One line per concept. Use ‘;’ for concatenating multiple mappings into one field – if applicable

• A sample output file could look like this:

Concept	PT	OLS_mappings	OXO_mappings
DOID:162	cancer	MeSH:D009369; EFO:0000311	MeSH:D009369; EFO:0000311

How to run exercise.py

This script was run using Anaconda. You will need the same environment to run the script.
Steps to run the script:

• Install Anaconda for Python 3.8.
• Then, on the command line run:
  conda env create -f environment.yml
• Make sure you conda activate the environment you just created
• Run the python program as usual:
  python exercise.py

Additional info: the package rdfpandas

The pandas package of course has many built-in functions to convert csv and other delimited data to DataFrames. But it does not handle RDF serializations well.

The rdfpandas package addresses this issue by being able to handle Turtle RDF files, which is the file we are using in this exercise.

from rdfpandas.graph import to_dataframe
import pandas as pd
import rdflib

g = rdflib.Graph()
g.parse('to_df_test.ttl', format = 'ttl')
df = to_dataframe(g)
df.to_csv('test.csv', index = True, index_label = "@id")