A5037662171- Biomedical Text Mining and Ontologies
- Bioinformatics and Genomic Networks
- Machine Learning in Bioinformatics
- Semantic Web and Ontologies
- Topic Modeling
- Artificial Intelligence in Healthcare
- Health, Environment, Cognitive Aging
- Protein Structure and Dynamics
- Traumatic Brain Injury and Neurovascular Disturbances
- RNA and protein synthesis mechanisms
- Cardiac Arrest and Resuscitation
- Genomics and Rare Diseases
- Occupational Health and Safety Research
King Abdullah University of Science and Technology
2018-2022
Kootenay Association for Science & Technology
2020
Abstract Motivation Ontologies are widely used in biology for data annotation, integration and analysis. In addition to formally structured axioms, ontologies contain meta-data the form of annotation axioms which provide valuable pieces information that characterize ontology classes. Annotation commonly include class labels, descriptions or synonyms. Despite being a rich source semantic information, generally unexploited by ontology-based analysis methods such as similarity measures. Results...
We propose the Onto2Vec method, an approach to learn feature vectors for biological entities based on their annotations biomedical ontologies. Our method can be applied a wide range of bioinformatics research problems such as similarity-based prediction interactions between proteins, classification interaction types using supervised learning, or clustering.
Ontologies have long been employed in the life sciences to formally represent and reason over domain knowledge, they are almost every major biological database. Recently, ontologies increasingly being used provide background knowledge similarity-based analysis machine learning models. The methods combine still novel actively developed. We an overview that use compute similarity incorporate them methods; particular, we outline how semantic measures ontology embeddings can exploit biomedical...
Abstract Background The current COVID-19 pandemic and the previous SARS/MERS outbreaks of 2003 2012 have resulted in a series major global public health crises. We argue that interest developing effective safe vaccines drugs to better understand coronaviruses associated disease mechenisms it is necessary integrate large exponentially growing body heterogeneous coronavirus data. Ontologies play an important role standard-based knowledge data representation, integration, sharing, analysis....
The number of available protein sequences in public databases is increasing exponentially. However, a significant percentage these lack functional annotation, which essential for the understanding how biological systems operate. Here, we propose novel method, Quantitative Annotation Unknown STructure (QAUST), to infer functions, specifically Gene Ontology (GO) terms and Enzyme Commission (EC) numbers. QAUST uses three sources information: structure information encoded by global local...
Abstract Motivation Over the past years, significant resources have been invested into formalizing biomedical ontologies. Formal axioms in ontologies developed and used to detect ensure ontology consistency, find unsatisfiable classes, improve interoperability, guide extension through application of axiom-based design patterns encode domain background knowledge. The knowledge may also potential provide for machine learning predictive modelling. Results We use ontology-based methods evaluate...
Abstract Motivation There are now over 500 ontologies in the life sciences. Over past years, significant resources have been invested into formalizing these biomedical ontologies. Formal axioms developed and used to detect ensure ontology consistency, find unsatisfiable classes, improve interoperability, guide extension through application of axiom-based design patterns, encode domain background knowledge. At same time, extended their amount human-readable information such as labels...
Abstract Motivation Ontologies are widely used in biomedicine for the annotation and standardization of data. One main roles ontologies is to provide structured background knowledge within a domain as well set labels, synonyms, definitions classes domain. The two types information provided by have been extensively exploited natural language processing machine learning applications. However, they commonly separately, thus it unknown if joining sources can further benefit data analysis tasks....
Motivation: Ontologies are widely used in biology for data annotation, integration, and analysis. In addition to formally structured axioms, ontologies contain meta-data the form of annotation axioms which provide valuable pieces information that characterize ontology classes. Annotations commonly include class labels, descriptions, or synonyms. Despite being a rich source semantic information, generally unexploited by ontology-based analysis methods such as similarity measures. Results: We...