Best paper of the year in Medical Informatics for AnacletoLab
The International Medical Informatics Association (IMIA) has selected the paper:
M. Notaro, M. Schubach, P.N. Robinson, G. Valentini. Prediction of Human Phenotype Ontology terms by means of hierarchical ensemble method, BMC Bioinformatics, vol. 18 (1), 2017
as one of the best 5 papers of the year for the Section "Knowledge Representation and Management".
The novelty of this work, realized by AnacletoLab, Computational Biology and Bioinformatics Lab of the Computer Science Dept. of UNIMI, in collaboration with the Jackson Lab for Genomic Medicine, CT, USA and with the Berlin Institute of Health, consists in the integration of biomedical ontologies in the architecture of machine learning models for the prediction of genes associated with pathological phenotypes. This integration, realized through a hierarchical ensemble of learning machines, allows to identify in silico new "candidate genes" that can be associated with pathological human phenotypes