Bio-Ontologies (#bioontologies; http://www.bio-ontologies.org.uk/) is an ISMB Community of Special Interest (COSI) that covers the latest and most innovative research in the application of ontologies and more generally the organization, presentation, and dissemination of knowledge in biomedicine and the life sciences. Join our mailing list and our slack team! Have something to say? Tweet to @bioontologies or send an email to the organizers - firstname.lastname@example.org.
Bio-Ontologies will be in its 24th year at ISMB 2021 which will be held as an ONLINE CONFERENCE between July 25-29, 2021. Bio-ontologies strives to provide a vibrant environment for reporting novel methods and sharing experiences on the construction and application of ontologies in health care and the life sciences. The COSI offers a constructive environment to nurture discussion of innovative and scientifically sound work that ranges from preliminary to completed, from both young and experienced investigators alike. Bio-Ontologies participants also benefit from a strongly interdisciplinary setting, where ISMB attendees intermingle with members from American Medical Informatics Association (AMIA) and the W3C’s Healthcare and Life Sciences Community Group (HCLSCG), thereby increasing impact through broader dissemination and enabling new and exceptional collaborations. This year's event will be integrated into the main conference and will feature papers accepted for the ISMB proceedings.
Program is available at ISMB: https://www.iscb.org/cms_addon/conferences/ismb2020/tracks/bio-ontcosi
Actionable Machine Learning for Drug Discovery andDevelopment
Marinka Zitnik, Harvard University
The success of machine learning depends heavily on the choice of features on which the algorithms are applied. For that reason, much of the efforts go into the engineering of informative features. In this talk, I describe our research in learning deep representations that are actionable and allow endpoint users to ask what-if questions and receive robust predictions that can be interpreted meaningfully. These methods specify deep graph neural functions that can flexibly embed data points into an embedding space, optimized to reflect the topology of input data. Our recent theoretical results allow us to augment the embeddings and make them robust and counterfactually fair. I will describe how these methods enabled the repurposing of drugs for an emerging pathogen and led to downstream validation in human cells. Last, I will highlight Therapeutics Data Commons (https://tdcommons.ai), a platform with AI/ML-ready datasets and tasks for therapeutics together with an ecosystem of tools, libraries, leaderboards, and community resources.
Marinka Zitnik is an Assistant Professor at Harvard University with appointments in the Department of Biomedical Informatics, Broad Institute of MIT and Harvard, and Harvard Data Science. Dr. Zitnik's research is at the interface of machine learning and biomedical informatics, focusing on graph representation learning, knowledge graphs, data fusion, and their applications to network biology and therapeutics. Dr. Zitnik has published extensively in top ML venues and leading interdisciplinary journals. She has organized numerous workshops and tutorials in the nexus of AI, deep learning, drug discovery, and medical AI at leading conferences, where she is also in the organizing committees. Her research won numerous best paper and research awards from the International Society for Computational Biology, Bayer Early Excellence in Science Award, a Rising Star Award in Electrical Engineering and Computer Science, and a Next Generation Award in Biomedicine.
Aligning Human and Machine Intelligence
Matthias Samwald, University of Vienna
Recent years were marked by outstanding advances in the capabilities of artificial intelligence (AI). Deep learning enabled rapid progress on many benchmarks that were previously deemed difficult to tackle for machine learning algorithms, often achieving human-level performance. Models such as GPT-3 demonstrated problem-solving capabilities across a wide range of intelligence tasks. Given these developments, AI holds the potential to radically transform society through accelerating technological, biomedical, and societal knowledge creation and innovation. The realization of this potential increasingly hinges on our understanding of how to best define, train, measure, interconnect and utilize AI capabilities. The development of a shared ontology of AI tasks and capabilities is instrumental towards achieving such an understanding. In this talk, I will present recent work on the Intelligence Task Ontology, a large-scale ontology with broad coverage of artificial intelligence tasks, benchmarks, and datasets. I will demonstrate how such ontological models can form the foundation for harnessing AI in biomedical research and clinical decision-making.
Matthias Samwald is an associate professor for biomedical informatics and artificial intelligence at the Medical University of Vienna. His work is focused on speeding up scientific progress and the development of novel, personalized therapies through advanced AI technologies and knowledge-based systems. He co-authored over 100 peer-reviewed publications in the areas of personalized medicine, NLP, graph machine learning, information retrieval, biomedical ontologies, and clinical decision support systems.
Call for Submissions - DEADLINE Thursday, May 6, 2021 (Key Dates as per ISMB)
We invite the submission of 1-page poster abstracts, and 2-4 page short papers for oral presentation. Please prepare your submission using the templates at the bottom of the site. Topics include but not limited to:
Ontologies to support COVID-19 research
Ontologies in annotation and metadata standards
Machine learning and ontologies
Text mining and ontologies
Ontology evolution, quality, and evaluation
Ontologies, knowledge representation, and reasoning
Submissions can be made on the ISMB site (https://www.iscb.org/ismbeccb2021-submit). All submissions accepted for oral presentation can optionally also be presented as a poster. Accepted submissions require that one author register for the conference and present the work. The presenter should identify themselves as the corresponding author during the submission process. The presentations are normally allocated between 10-20 minutes.
All submissions to Bio-Ontologies are reviewed. Review criteria include :
Relevance of the topic to conference attendees.
Significance of the problem, such that reviewers understand that it is a problem that is important and worth solving.
Novelty of the approach, including critical discussion of related work.
Soundness and potential for reproducibility of the study and/or methods.
Quality of writing, including readability and clearly stated contribution(s)
Submissions accepted for oral presentation will be considered for invitation to a Special Issue in the Journal of Biomedical Semantics (JBMS). Research that has been accepted as a poster presentation may also be invited for further development to the JBMS special issue.