Leading Professional Society for Computational Biology and Bioinformatics
Connecting, Training, Empowering, Worldwide

banner

Bio-Ontologies COSI Track Presentations

Attention Conference Presenters - please review the Speaker Information Page available here
Bio-Ontologies Welcoming Remarks
Date: Monday, July 24
Time: 10:00 AM - 10:10 AM
Room: Meeting Hall V
  • Michel Dumontier, Maastricht University, Netherlands

Presentation Overview: Show

A welcome and overview for the 2017 Bio-Ontologies COSI track

Bio-Ontologies KEYNOTE: Ontologies: Necessary, but not sufficient
Date: Monday, July 24
Time: 10:10 AM - 11:10 AM
Room: Meeting Hall V
  • Robert Stevens, Manchester University, United Kingdom

Presentation Overview: Show

The past twenty years of the Bio-Ontologies meeting have seen great advances in the use of ontologies within the domain of biology. The bio-ontology community has produced a broad range of ontologies that capture a wide variety of biological phenomena. At their heart, ontologies allow us to `know what we are talking about', affording consistent annotation and thus querying of biological phenomena across species. This drive towards a common, shared understanding of the entities of the domain has its successes, but despite the successes we see in coverage and use in annotation, there are limitations in how we develop and use ontologies; there is more we can do. Over the next twenty years, these limitations may cease to become inconveniences , and grow to become a block in the uptake of ontologies, and ontological technologies. To avoid this, we must accept that ontologies are necessary for exploiting knowledge computationally within biology, but they are not sufficient to do so. In this talk I will describe where I think the bio-ontologies community is, where it needs to go and how it should do so. Along the way, I will elucidate some of the other necessary conditions for exploiting knowledge in biology. If we achieve these conditions, ontologies will still be used in another 20 years time and, by then, will have contributed significantly to our understanding of biology.

Robert Stevens is a professor of computer science at the University of Manchester. He has a Ph.D. in human computer interaction and a degree in biochemistry. he has used this background to drive a very cross-disciplinary research agenda. His main research interests are in how we develop, interact with and use logic based ontologies to describe, manage and analyse biological data. Along the way he has also done work in workflows in bioinformatics, extracting and representing methods used in analysing biological data and how humans interact with complex information. Robert has long been an advocate of description logic based ontologies and has established the "Pizza Tutorial" as one of the most widely used OWL tutorials. He was co-chair of the Bio-Ontologies SIG at ISMB for eight years, a co-chair of the ontologies and databases track at ISMB, co-chair of the International biomedical Ontology conference in 2012 and co-chair of SWAT4LS in 2015.

Bio-Ontologies Panel: 20 years of Bio-Ontologies at ISMB
Date: Monday, July 24
Time: 11:10 AM - 12:00 PM
Room: Meeting Hall V

    Presentation Overview: Show

    A panel discussion on the past 20 years of Bio-Ontologies.

    Deep Learning with Word Embeddings improves Biomedical Named Entity Recognition
    Date: Monday, July 24
    Time: 2:00 PM - 2:30 PM
    Room: Meeting Hall V
    • Ulf Leser, Humboldt-Universität zu Berlin, Germany
    • David Luis Wiegandt, Humboldt-Universität zu Berlin, Germany
    • Mariana Neves, Hasso-Plattner-Institute, Germany
    • Leon Weber, Humboldt-Universität zu Berlin, Germany
    • Maryam Habibi, Humboldt-Universität zu Berlin, Germany

    Presentation Overview: Show

    Motivation: Text mining has become an important tool for biomedical research. The most fundamental text mining task is the recognition of biomedical named entities (NER), such as genes, chemicals, and diseases. Current NER methods rely on predefined features which try to capture the specific surface properties of entity types, properties of the typical local context, background knowledge, and linguistic information. State-of-the-art tools are entity-specific, as dictionaries and empirically optimal feature sets differ between entity types, which makes their development costly. Furthermore, features are often optimized for a specific gold standard corpus, which makes extrapolation of quality measures difficult.
    Results: We show that a completely generic method based on deep learning and statistical word embeddings (called LSTM-CRF) outperforms state-of-the-art entity-specific NER tools, and often by a large margin. To this end, we compared the performance of LSTM-CRF on 33 data sets covering five different entity classes with that of best-of-class NER tools and an entity-agnostic CRF implementation. On average, F1-score of LSTM-CRF is 5% above that of the baselines, mostly due to a sharp increase in recall.
    Availability: The source code for LSTM-CRF is available at https://github.com/glample/tagger and the links to the corpora are available at https://corposaurus.github.io/corpora.

    EXTRACT 2.0: text-mining-assisted interactive annotation of bio-medical named entities and ontology terms
    Date: Monday, July 24
    Time: 2:30 PM - 2:45 PM
    Room: Meeting Hall V
    • Lars Juhl Jensen, Cellular Network Biology Group, Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, 2200 Copenhagen N, Denmark, Denmark
    • Rūdolfs Bērziņš, Cellular Network Biology Group, Novo Nordisk Foundation Center for Protein Research, Faculty of Health and Medical Sciences, University of Copenhagen, 2200 Copenhagen N, Denmark, Denmark
    • Evangelos Pafilis, Institute of Marine Biology, Biotechnology and Aquaculture, Hellenic Centre for Marine Research, Greece
    A Text Mining Approach Characterizes Fusion Proteins and Their Interactions From PubMed Publications
    Date: Monday, July 24
    Time: 2:45 PM - 3:00 PM
    Room: Meeting Hall V
    • Lars Juhl Jensen, University of Copenhagen, Denmark
    • Alessandro Gorohovski, National Technical University of Ukraine (KPI), Ukraine
    • Milana Frenkel-Morgenstern, Bar-Ilan University, Israel
    • Somnath Tagore, Bar-Ilan University, Israel
    Bio-Ontologies KEYNOTE: Sense and similarity: making sense of similarity for ontologies
    Date: Monday, July 24
    Time: 3:00 PM - 4:00 PM
    Room: Meeting Hall V
    • Catia Pesquita, University of Lisbon, Portugal

    Presentation Overview: Show

    Judging the similarity between things allows us to understand the world. Similar things tend to behave similarly, which supports inference. In the life sciences, knowledge can seldom be reduced to mathematical form, so calculating similarity is not trivial. Ontologies provide the scaffolding to compute similarity between entities based on their semantic annotations. However, biomedical entities can be annotated under multiple ontologies to cover distinct domains or due to fragmentation issues. Therefore, an accurate measure of the similarity between biomedical entities depends on the creation of meaningful links between related ontologies. I will discuss the challenges and the evolutions of the last decade in computing the similarity between concepts from different bio-ontologies (ontology matching), and the similarity between bio-ontology annotated entities (semantic similarity), and their interconnectedness. I will look into applications in biological and clinical domains, namely for data mining and semantic data integration.

    Catia Pesquita is an Assistant Professor at the Computer Science department at Faculdade de Ciências, Universidade de Lisboa. She is also a senior researcher at the Large Scale Information Systems Research Lab (LASIGE) where she leads research projects dedicated to several areas including the development of novel semantics based approaches to knowledge engineering and data mining, particularly in the biomedical and clinical domains. She has made significant contributions to the areas of ontology matching and ontology-based semantic similarity. Catia Pesquita has a degree in Biology, an MSc in Bioinformatics and a PhD in Computer Science, awarded by Universidade de Lisboa. She was the recipient of a research award from the Luso-American Foundation for Development. Together with her research team she has also won several awards and competitions in ontology matching.

    Classification and analysis of a large collection of in vivo assay descriptions
    Date: Monday, July 24
    Time: 4:45 PM - 5:00 PM
    Room: Meeting Hall V
    • John Overington, Medicines Discovery Catapult, United Kingdom
    • Magdalena Zwierzyna, BenevolentAI/University College London, United Kingdom

    Presentation Overview: Show

    Testing potential drug treatments in animal models is a crucial part of preclinical drug discovery. Yet, high failure rates for new therapies in the clinic demonstrate a growing need for better understanding of the relevance and role of animal model research. In this study, we use text mining methods and machine learning models to systematically analyze a large collection of drug discovery assay descriptions in rats and mice. First, we parse the assay descriptions and mine them for information about animal experiments: genetic strains, experimental treatments, and phenotypic readouts used in the assays. To automatically organize the extracted information, we construct a semantic space of assay descriptions using a neural network language model, Word2Vec, and show that related animal models and phenotypic terms tend to cluster together in the constructed semantic space. In addition, we show that random forest classifiers trained with features generated by Word2Vec predict the class of drugs tested in different assays with accuracy of 0.89. Finally, we combine information mined from text with structured annotations stored in the ChEMBL database to investigate the patterns of usage of different animal models across a range of experiments, drug classes, and disease areas. Our results demonstrate that text mining and machine learning have a potential to contribute to the ongoing debate on the interpretation and reproducibility of animal studies through enabling access, integration, and large-scale analysis of in vivo drug screening data.

    Onassis: Ontology Annotation and Semantic Similarity Software
    Date: Monday, July 24
    Time: 5:00 PM - 5:15 PM
    Room: Meeting Hall V
    • Mattia Pelizzola, Istituto Italiano di Tecnologia, Italy
    • Eugenia Galeota, Istituto Italiano di Tecnologia, Italy

    Presentation Overview: Show

    Large-scale biological data integration systems rely on the availability of well-structured metadata. Despite the development of standards, metadata is often sub-optimal in terms of the amount and quality of the available information. In the field of genomics and epigenomics, the information about the cell type and the disease state (or lack of disease) are two fundamental pieces of information that are often unclearly provided in the metadata of the high-throughput data. While the ability of deciphering the interplay between multi-layered epigenomics marks strongly benefits from the availability of numerous publicly available datasets, the difficultly in retrieving and associating relevant samples still constitutes a major bottleneck.
    Onassis is a user-friendly R package aimed at combining semantically coherent heterogeneous omics datasets from private and/or public repositories. The recognition of domain specific entities not only allows users to retrieve samples related to a given cell type or experimental condition, but also to discover different and not immediately obvious relationships between experiments. Onassis functionalities can be used to retrieve unstructured and poorly annotated sample’s descriptions, recognize concepts from a multitude of biomedical ontologies and to quantify the similarities/divergences between pairs or groups of query studies. In particular the software includes modules to assist on: (i) the retrieval of samples’ metadata from GEO and SRA, (ii) the annotation of these data with concepts belonging to OBO biomedical ontologies, and (iii) the organization of available samples in comparable and coherent groups based on semantic similarity metrics.
    Onassis features have already been proved to be an essential part in the development of a workflow and management system for high-throughput sequencing data1, guaranteeing a seamless integration with semantically coherent publicly available data. Based on the analysis of independent studies profiling the binding of the Myc transcription factor by ChIP-seq, we previously showed that the semantic similarity determined with Onassis was coherent with Myc binding patterns. Furthermore, we showed that is possible to complement the dataset of Myc ChIP-seq experiments, by adding chromatin modification marks and Pol II activity in the same or very similar cell lines and disease states2. This illustrated Onassis ability to meaningfully combine experiments from independent studies, and expand the dataset by retrieving semantically coherent experiments profiling additional marks.
    Illustrating the usefulness of Onassis for the identification of putative disease biomarkers, we adopted this tool to annotate ~14.000 samples profiling DNA methylation in various cell types and disease concepts. Within each sample, we identified Low Methylated Regions (LMRs), thus defining cell-type specific LMRs that are differential between disease states and healthy conditions. We focused on LMRs distal from genes, which can be associated to regulatory regions such as enhancers. We defined a specificity score of each LMR, based on its occurrence in ~70 different concepts from the Cell Line Ontology. Cell-type specific LMRs were further analyzed to identify those that were enriched in samples associated to disease state(s) compared to healthy samples, leveraging on concepts from the Disease Ontology. Functional enrichment analyses of the genes in proximity to the LMRs highlighted the pathways that could be associated with altered DNA methylation signatures in a given disease state.
    Altogether these analyses illustrate the usefulness of Onassis in playing a key role in the integrative analyses of datasets from repositories of large-scale high-throughput data.

    Break Out for Paper Writing: Challenges and Opportunities in Bio-Ontologies
    Date: Monday, July 24
    Time: 5:15 PM - 6:00 PM
    Room: Meeting Hall V
    • Robert Hoehndorf, KAUST, Saudi Arabia
    • Michel Dumontier, Maastricht University, Netherlands

    Presentation Overview: Show

    An interactive session to jointly conceive and write a paper on the past, present, and future of Bio-ontologies.

    Bio-ontologies KEYNOTE: Mining the Ultimate Phenome Repository
    Date: Tuesday, July 25
    Time: 8:30 AM - 9:30 AM
    Room: Meeting Hall V
    • Nigam Shah, Stanford University, United States

    Presentation Overview: Show

    Controlled terms are widely used to annotate gene products and literature abstracts. Subsequent analysis of the resulting annotations enable significant insights that advance molecular biology. In this presentation we will review how the use of ontologies in annotating, and analyzing, Electronic Health Records (EHR) make it possible to examine the outcomes of decisions made by doctors during clinical practice to generate evidence from the collective experience of millions of patients. I will describe the role that bio-ontologies play in supporting the Informatics Consult service at Stanford, and in enabling an advanced search engine that build cohorts of patients with specific phenotypes in sub-second response times..
    Dr. Nigam Shah is an associate professor of Medicine (Biomedical Informatics) at Stanford University, Assistant Director of the Center for Biomedical Informatics Research, and a core member of the Biomedical Informatics Graduate Program. Dr. Shah's research focuses on combining machine learning and prior knowledge in medical ontologies to enable use cases of the learning health system. Dr. Shah received the AMIA New Investigator Award for 2013, was elected into the American College of Medical Informatics (ACMI) in 2015 and is inducted into the American Society for Clinical Investigation (ASCI) in 2016. He holds an MBBS from Baroda Medical College, India, a PhD from Penn State University and completed postdoctoral training at Stanford University. More at: https://med.stanford.edu/profiles/nigam-shah

    A Machine-Compiled Database of Genome-Wide Association Studies
    Date: Tuesday, July 25
    Time: 10:30 AM - 11:00 AM
    Room: Meeting Hall V
    • Michael Snyder, Stanford University, United States
    • Serafim Batzoglou, Stanford University, United States
    • Christopher Re, Stanford University, United States
    • Alexander Ratner, Stanford University, United States
    • Braden Hancock, Stanford University, United States
    • Jialin Ding, Stanford University, United States
    • Volodymyr Kuleshov, Stanford University, United States

    Presentation Overview: Show

    Tens of thousands of genotype/phenotype associations have been discovered to date, yet not all of them are available to scientists in a useful, easy to access form. Here, we describe GwasDB, a machine reading system for automatically extracting these associations from the scientific literature in the form of a structured database. Our system reveals that existing manually-curated repositories are incomplete, and produces >2,000 previously undocumented associations, which represents about 20% of the size of the largest existing repository of open-access papers. Our results highlight both the importance and the feasibility of using machine reading algorithms to make scientific findings easily accessible.

    MetaCrowd: Crowdsourcing Biomedical Metadata Quality Assessment
    Date: Tuesday, July 25
    Time: 11:00 AM - 11:15 AM
    Room: Meeting Hall V
    • Michel Dumontier, Maastricht University, Netherlands
    • Amrapali Zaveri, Maastricht University, Netherlands

    Presentation Overview: Show

    The ability to efficiently search and filter datasets depends on access to high quality metadata. While most biomedical repositories require data submitters to provide a minimal set of metadata, some such as the Gene Expression Omnibus (GEO) allow users to specify additional metadata in the form of textual key-value pairs (e.g. sex: female). However, since there is no structured vocabulary to guide the submitter regarding the metadata terms to use, consequently, the 44,000,000+ key-value pairs in GEO suffer from numerous quality issues including redundancy, heterogeneity, inconsistency, and incompleteness. Such issues hinder the ability of scientists to hone in on datasets that meet their requirements and point to a need for accurate, structured and complete description of the data. Current methods for quality assessment are not only time consuming but also error prone and are not scalable over the vast amount of biomedical metadata that needs curation. Moreover, they lack the ability to give detailed insights about the completeness of the metadata. Importantly, with a scarcity of domain experts to curate the rapidly increasing data in GEO, there is a need for more efficient methods for curating the metadata such as crowdsourcing. Thus, in our approach, MetaCrowd, we apply crowdsourcing as a means to assess the quality of the biomedical metadata, particularly in the GEO dataset. We report on preliminary results and propose a hybrid solution of combining clustering and crowdsourcing methods for large-scale biomedical metadata quality assessment

    Enabling community editing of assay terms in OBI while ensuring consistent use of design patterns with spreadsheet templates
    Date: Tuesday, July 25
    Time: 11:15 AM - 11:30 AM
    Room: Meeting Hall V
    • Obi Consortium, Ontology for Biomedical Investigations, United States
    • Bjoern Peters, La Jolla Institute for Allergy and Immunology, United States
    • Randi Vita, La Jolla Institute for Allergy & Immunology, United States
    • James A. Overton, Knocean Inc., Canada

    Presentation Overview: Show

    The Ontology for Biological Investigations (OBI) defines more than 2500 terms, including more than 700 assays types. Not only do assays form the largest group of terms in OBI, they also serve as the primary axis around which other parts of OBI are organized. Here we describe how OBI developers created a template system to facilitate community editing of the assay terms with the goal of (1) unifying logical design patterns and (2) enhancing the ability of a broader user base to productively contribute to ontology development with a less steep learning curve.

    Assessing the quality of manually curated drug indication and usage information via ontology term mappings
    Date: Tuesday, July 25
    Time: 11:30 AM - 11:45 AM
    Room: Meeting Hall V
    • Tudor Oprea, University of New Mexico, United States
    • Linda Rieswijk, Maastricht University, Netherlands
    • Michel Dumontier, Maastricht University, Netherlands
    • Kody Moodley, Maastricht University, Netherlands

    Presentation Overview: Show

    We have developed a new protocol to manually curate indications and usage context for drugs from structured product labels (SPLs) on the DailyMed (dailymed.nlm.nih.gov) online drug label catalogue. This is needed because resources that list drug indications in a computable way (using established bio-ontology terms) generally suffer from inconsistencies in the quality of such information. This, in turn, affects downstream uses for the information such as drug repositioning and clinical decision support. We view our curation effort as a helpful preliminary step towards enriching existing drug resources with more detailed, higher quality metadata about drug indications and usage. However, before we can augment these resources with our metadata, we would like to assess the quality thereof. We have very recently begun our analysis and we are on the brink of preliminary results.

    Bio-Ontologies 1 minute madness
    Date: Tuesday, July 25
    Time: 11:45 AM - 12:00 PM
    Room: Meeting Hall V

      Presentation Overview: Show

      A brief opportunity to present something in 1 minute!

      BIOSSES: A Semantic Sentence Similarity Estimation System for the Biomedical Domain
      Date: Tuesday, July 25
      Time: 2:00 PM - 2:30 PM
      Room: Meeting Hall V
      • Arzucan Ozgur, Bogazici University, Turkey
      • Gizem Soğancıoğlu, Boğaziçi University, Turkey
      • Hakime Öztürk, Boğaziçi University, Turkey

      Presentation Overview: Show

      Motivation: The amount of information available in textual format is rapidly increasing in the biomedical domain. Therefore, natural language processing (NLP) applications are becoming increasingly important to facilitate the retrieval and analysis of these data. Computing the semantic similarity between sentences is an important component in many NLP tasks including text retrieval and summarisation. A number of approaches have been proposed for semantic sentence similarity estimation for generic English. However, our experiments showed that such approaches do not effectively cover biomedical knowledge and produce poor results for biomedical text. Methods: We propose several approaches for sentence-level semantic similarity computation in the
      biomedical domain, including string similarity measures and measures based on the distributed vector representations of sentences learned in an unsupervised manner from a large biomedical corpus. In addition, ontology-based approaches are presented that utilize general and domain-specific ontologies. Finally, a supervised regression based model is developed that effectively combines the different similarity computation metrics. A benchmark data set consisting of 100 sentence pairs from the biomedical literature is manually annotated by five human experts and used for evaluating the proposed methods.

      Results: The experiments showed that the supervised semantic sentence similarity computation approach obtained the best performance (0.836 correlation with gold standard human annotations) and improved over the state-of-the-art domain-independent systems up to 42.6% in terms of the Pearson correlation metric.

      Availability: A web-based system for biomedical semantic sentence similarity computation, the source code, and the annotated benchmark data set are available at: http://tabilab.cmpe.boun.edu.tr/BIOSSES/

      Using ontologies to find and correct errors in database content
      Date: Tuesday, July 25
      Time: 2:30 PM - 2:45 PM
      Room: Meeting Hall V
      • Bjoern Peters, La Jolla Institute for Allergy and Immunology, United States
      • James A. Overton, Knocean Inc., Canada
      • Randi Vita, La Jolla Institute for Allergy and Immunology, United States

      Presentation Overview: Show

      The Immune Epitope Database (IEDB iedb.org) implemented external ontologies and other taxonomically structured vocabularies into its curation and search interfaces to simplify curation practices, improve the user query experience, and facilitate interoperability between the IEDB and other databases. This work led to more accurate curation and improved search capabilities. Each time we gained new information about our data and identified errors that we would not have found otherwise. Here we describe these experiences in finding and correcting errors with the hope that it will inspire other projects to do the same.

      EFO 3: “Your” experimental factor ontology
      Date: Tuesday, July 25
      Time: 2:45 PM - 3:00 PM
      Room: Meeting Hall V
      • Helen Parkinson, European Bioinformatics Institute, United Kingdom
      • Simon Jupp, European Bioinformatics Institute, United Kingdom
      • Danielle Welter, European Bioinformatics Institute, United Kingdom
      • Sirarat Sarntivijai, European Bioinformatics Institute, United Kingdom

      Presentation Overview: Show

      The Experimental Factor Ontology (EFO) is an open-source ontology established in 2009 to aid the annotation of data at the EBI Expression Atlas . EFO has grown from ~2,600 classes to 19,799 classes (as of March 15th 2017). The growing number of classes in EFO reflects the expanding user-base from multiple databases and serves a wider set of use-cases. EFO now covers various life science domains, both animal and plant in many projects i.e., Open Targets , GWAS Catalog , EXCELERATE , and Genomics England . EFO also facilitates knowledge sharing in multi-organisation collaborative work such as ENCODE Consortium , NASA GeneLab , and GSK (Sarntivijai et al., 2016). The growing user-base has necessitated the develop- ment of a new operational pipeline for EFO that presents a more generalised approach to application ontology building.

      Conventions to make ontology term labels predictable and unique
      Date: Tuesday, July 25
      Time: 3:00 PM - 3:15 PM
      Room: Meeting Hall V
      • Obo Operations Committee, OBO, United States
      • Randi Vita, La Jolla Institute for Allergy & Immunology, United States
      • James A. Overton, Knocean Inc., Canada
      • Bjoern Peters, La Jolla Institute for Allergy and Immunology, United States
      • Chris Mungall, Lawrence Berkeley National Laboratory, United States

      Presentation Overview: Show

      While the official identifier of an OWL ontology class or relationship is its IRI, humans rely on textual labels instead. An ontology labeling scheme must clearly convey to human users what a label is referring to when encountering it in an ontology editor, ontology lookup service, or dataset annotated with ontology terms. We here define a set of test-able conventions on how labels should be constructed in order to harmonize labeling within and across ontologies in general, and those in the OBO Foundry in particular.

      Enhancing evidence from literature in Open Targets – a platform for drug target validation
      Date: Tuesday, July 25
      Time: 3:15 PM - 3:45 PM
      Room: Meeting Hall V
      • Johanna McEntyre, EMBL-EBI, United Kingdom
      • Ian Dunham, EMBL-EBI, United Kingdom
      • Senay Kafkas, EMBL-EBI, United Kingdom

      Presentation Overview: Show

      Evaluating the different aspects of target-disease associations such as gene expression level changes and genetic associations is crucial for understanding the mechanism of disease. However, evidence on target-disease associations is available from various resources and integrating it with drug information would provide comprehensive insights for drug repurposing and development. The Open Targets Platform integrates such evidence with the aim of assisting scientists to identify and prioritise drug targets. Currently, more than 2.5 million target-disease associations are covered by the platform. Since Open Targets aims to deliver one of the most comprehensive resources relevant to target validation, we are enhancing the target-disease evidence in the platform with information on the known target-drug associations from ChEMBL and drug-disease associations from text mining the literature. Results show that text mining helps to enhance the existing evidence substantially. In this study, we present our approach and initial results. All data is available from ftp://ftp.ebi.ac.uk/pub/databases/pmc/otar/drug_disease/.

      Bio-Ontologies Closing Remarks
      Date: Tuesday, July 25
      Time: 3:45 PM - 4:00 PM
      Room: Meeting Hall V
      • Michel Dumontier, Maastricht University, Netherlands

      Presentation Overview: Show

      Closing remarks for Bio-ontologies 2017