PPARATUS AND A SYSTEM FOR CALCULATING SIMILARITIES BETWEEN DRUGS AND USING THE SIMILARITIES TO EXTRAPOLATE SIDE EFFECTS

摘要

A system for calculating similarities between drugs and using the similarities to extrapolate side effects of a specified drug comprising: a similarity-computation apparatus configured to use open data to compute similarities between drugs, with: a data integration module configured to obtain data related to a set of drugs from a plurality of open data sources and to integrate the data to generate RDF triples; an RDF module configured to store an RDF graph of the RDF triples: and a similarity calculation module configured to retrieve a list of all the drugs present in the RDF graph and to calculate the similarity of each drug with every other drug in the RDF graph, for storage in a similarity database; and a side effect determining apparatus configured to extrapolate side effects of a specified drug from open data related to the set of drugs, with: a connection to the similarity database; a user interface configured to allow a user to specify a drug from the set of drugs; a neighbours ranking module configured to use the drug similarities to obtain nearest neighbours of the specified drug in terms of the similarities; and a side effects propagation module configured to collect known side effects from the nearest neighbours and to combine and rank the side effects according to their co-occurrence among the neighbours.

主权项

1. A similarity-computation apparatus configured to use open data to compute similarities between drugs, the apparatus comprising:
a data integration module configured to obtain data related to a set of drugs from at plurality of open data sources and to integrate the data to generate RDF triples; an RDF module configured to store an RDF graph of the RDF triples; and a similarity calculation module configured to retrieve a list of all the drugs present in the RDF graph and to calculate the similarity of each drug with every other drug in the RDF graph.