Search

Acknowledgements: This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) UK through the ACTION on cancer Grant (EP/R022925/1, EP/R022941/1). The computational resources were provided by the Swedish National Infrastructure for Computing (SNIC) at the Chalmers University of Technology partially funded by the Swedish Research Council through Grant Agreement No. 2018-05973. Prof. King acknowledges the support of the Knut and Alice Wallenberg Foundation Wallenberg Autonomous Systems and Software Program (WASP). N. F. Grinberg would like to acknowledge funding from the Wellcome Trust (WT107881) and the MRC (MC_UU_00002/4)., We present an extension to the federated ensemble regression using classification algorithm, an ensemble learning algorithm for regression problems which leverages the distribution of the samples in a learning set to achieve improved performance. We evaluated the extension using four classifiers and four regressors, two discretizers, and 119 responses from a wide variety of datasets in different domains. Additionally, we compared our algorithm to two resampling methods aimed at addressing imbalanced datasets. Our results show that the proposed extension is highly unlikely to perform worse than the base case, and on average outperforms the two resampling methods with significant differences in performance.

The Ontology for Biomedical Investigations (OBI) is an ontology that provides terms with precisely defined meanings to describe all aspects of how investigations in the biological and medical domains are conducted. OBI re-uses ontologies that provide a representation of biomedical knowledge from the Open Biological and Biomedical Ontologies (OBO) project and adds the ability to describe how this knowledge was derived. We here describe the state of OBI and several applications that are using it, such as adding semantic expressivity to existing databases, building data entry forms, and enabling interoperability between knowledge resources. OBI covers all phases of the investigation process, such as planning, execution and reporting. It represents information and material entities that participate in these processes, as well as roles and functions. Prior to OBI, it was not possible to use a single internally consistent resource that could be applied to multiple types of experiments for these applications. OBI has made this possible by creating terms for entities involved in biological and medical investigations and by importing parts of other biomedical ontologies such as GO, Chemical Entities of Biological Interest (ChEBI) and Phenotype Attribute and Trait Ontology (PATO) without altering their meaning. OBI is being used in a wide range of projects covering genomics, multi-omics, immunology, and catalogs of services. OBI has also spawned other ontologies (Information Artifact Ontology) and methods for importing parts of ontologies (Minimum information to reference an external ontology term (MIREOT)). The OBI project is an open cross-disciplinary collaborative effort, encompassing multiple research communities from around the globe. To date, OBI has created 2366 classes and 40 relations along with textual and formal definitions. The OBI Consortium maintains a web resource (http://obi-ontology.org) providing details on the people, policies, and issues being addressed in association with OBI. The current release of OBI is available at http://purl.obolibrary.org/obo/obi.owl. This work was supported by the National Institute of Allergy and Infectious Diseases (https:// www.niaid.nih.gov/Pages/default.aspx) R01AI081062 to YH, the National Institutes of Health (http://www. nih.gov/) R01GM093132, U01 DK 072473, P41 HG003619 and HHSN272201400030C to CJS, HHSN272201400028C to RHS, HHSN272201200010C and 1U19AI118626 to BP, the California Institute for Regenerative Medicine (https:// www.cirm.ca.gov) GC1R-06673-B to RHS, the National Science Foundation Division of Biological Infrastructure (http://www.nsf.gov) 1458400 to MCC