The Human Phenotype Ontology (HPO) is a standardized vocabulary of terms related to diseases. The importance and the specificity of HPO terms are estimated employing the Information Content (IC). Thus, the analysis of annotated data is a critical challenge for bioinformatics. There exist several approaches to support ontology curators in maintaining and analysing data. Among these, the use of Association Rules (AR) can improve the quality of annotations. In this paper, we present an algorithm for the parallel extraction of Weighted Association Rules (WAR) from HPO terms and annotations, able to face high dimension of data. Experiments performed on real and synthetic datasets show good speed-up and scalability.
Parallel Learning of Weighted Association Rules in Human Phenotype Ontology
Agapito G.;Cannataro M.;Guzzi P. H.;Milano M.
2020-01-01
Abstract
The Human Phenotype Ontology (HPO) is a standardized vocabulary of terms related to diseases. The importance and the specificity of HPO terms are estimated employing the Information Content (IC). Thus, the analysis of annotated data is a critical challenge for bioinformatics. There exist several approaches to support ontology curators in maintaining and analysing data. Among these, the use of Association Rules (AR) can improve the quality of annotations. In this paper, we present an algorithm for the parallel extraction of Weighted Association Rules (WAR) from HPO terms and annotations, able to face high dimension of data. Experiments performed on real and synthetic datasets show good speed-up and scalability.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
