{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,8]],"date-time":"2026-01-08T09:56:38Z","timestamp":1767866198022,"version":"3.49.0"},"reference-count":45,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2022,7,26]],"date-time":"2022-07-26T00:00:00Z","timestamp":1658793600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/journals\/pages\/open_access\/funder_policies\/chorus\/standard_publication_model"}],"funder":[{"name":"Science Compute Platform"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,9,20]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Adoption of recently developed methods from machine learning has given rise to creation of drug-discovery knowledge graphs (KGs) that utilize the interconnected nature of the domain. Graph-based modelling of the data, combined with KG embedding (KGE) methods, are promising as they provide a more intuitive representation and are suitable for inference tasks such as predicting missing links. One common application is to produce ranked lists of genes for a given disease, where the rank is based on the perceived likelihood of association between the gene and the disease. It is thus critical that these predictions are not only pertinent but also biologically meaningful. However, KGs can be biased either directly due to the underlying data sources that are integrated or due to modelling choices in the construction of the graph, one consequence of which is that certain entities can get topologically overrepresented. We demonstrate the effect of these inherent structural imbalances, resulting in densely connected entities being highly ranked no matter the context. We provide support for this observation across different datasets, models as well as predictive tasks. Further, we present various graph perturbation experiments which yield more support to the observation that KGE models can be more influenced by the frequency of entities rather than any biological information encoded within the relations. Our results highlight the importance of data modelling choices, and emphasizes the need for practitioners to be mindful of these issues when interpreting model outputs and during KG composition.<\/jats:p>","DOI":"10.1093\/bib\/bbac279","type":"journal-article","created":{"date-parts":[[2022,7,26]],"date-time":"2022-07-26T10:19:26Z","timestamp":1658830766000},"source":"Crossref","is-referenced-by-count":23,"title":["Implications of topological imbalance for representation learning on biomedical knowledge graphs"],"prefix":"10.1093","volume":"23","author":[{"given":"Stephen","family":"Bonner","sequence":"first","affiliation":[{"name":"Data Sciences and Quantitative Biology , Discovery Sciences, R&D, , Cambridge, UK"},{"name":"AstraZeneca , Discovery Sciences, R&D, , Cambridge, UK"}]},{"given":"Ufuk","family":"Kirik","sequence":"additional","affiliation":[{"name":"Data Sciences and Quantitative Biology , Discovery Sciences, R&D, , Gothenburg, Sweeden"},{"name":"AstraZeneca , Discovery Sciences, R&D, , Gothenburg, Sweeden"}]},{"given":"Ola","family":"Engkvist","sequence":"additional","affiliation":[{"name":"Molecular AI , Discovery Sciences, R&D, , Gothenburg, Sweeden"},{"name":"AstraZeneca , Discovery Sciences, R&D, , Gothenburg, Sweeden"}]},{"given":"Jian","family":"Tang","sequence":"additional","affiliation":[{"name":"HEC Montreal , Canada"},{"name":"Canadian Institute for Advanced Research (CIFAR) , Canada"},{"name":"Mila - Quebec AI Institute , Montreal, Canada"}]},{"given":"Ian P","family":"Barrett","sequence":"additional","affiliation":[{"name":"Data Sciences and Quantitative Biology , Discovery Sciences, R&D, , Cambridge, UK"},{"name":"AstraZeneca , Discovery Sciences, R&D, , Cambridge, UK"}]}],"member":"286","published-online":{"date-parts":[[2022,7,26]]},"reference":[{"issue":"6","key":"2022092013213390700_ref1","doi-asserted-by":"crossref","first-page":"419","DOI":"10.1038\/nrd4309","article-title":"Lessons learned from the fate of AstraZeneca\u2019s drug pipeline: a five-dimensional framework","volume":"13","author":"Cook","year":"2014","journal-title":"Nat Rev Drug Discov"},{"issue":"3","key":"2022092013213390700_ref2","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1038\/nrd.2017.244","article-title":"Impact of a five-dimensional framework on R&D productivity at AstraZeneca","volume":"17","author":"Morgan","year":"2018","journal-title":"Nat Rev Drug Discov"},{"issue":"6","key":"2022092013213390700_ref3","doi-asserted-by":"crossref","first-page":"333","DOI":"10.1038\/nrg.2016.49","article-title":"Coming of age: ten years of next-generation sequencing technologies","volume":"17","author":"Goodwin","year":"2016","journal-title":"Nat Rev Genet"},{"issue":"11","key":"2022092013213390700_ref4","doi-asserted-by":"crossref","first-page":"2281","DOI":"10.1038\/nprot.2013.143","article-title":"Genome engineering using the CRISPR-Cas9 system","volume":"8","author":"Ran","year":"2013","journal-title":"Nat Protoc"},{"issue":"1","key":"2022092013213390700_ref5","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/ncomms13091","article-title":"Multi-omic data integration enables discovery of hidden biological regularities","volume":"7","author":"Ebrahim","year":"2016","journal-title":"Nat Commun"},{"issue":"1","key":"2022092013213390700_ref6","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s13059-017-1215-1","article-title":"Multi-omics approaches to disease","volume":"18","author":"Hasin","year":"2017","journal-title":"Genome Biol"},{"key":"2022092013213390700_ref7","article-title":"A Review of Biomedical Datasets Relating to Drug Discovery: A Knowledge Graph Perspective","author":"Bonner","year":"2021"},{"key":"2022092013213390700_ref8","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab159","article-title":"Utilizing graph machine learning within drug discovery and development","author":"Gaudelet","year":"2021","journal-title":"Brief Bioinform"},{"issue":"1","key":"2022092013213390700_ref9","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41598-020-74922-z","article-title":"Preclinical validation of therapeutic targets predicted by tensor factorization on heterogeneous graphs","volume":"10","author":"Paliwal","year":"2020","journal-title":"Sci Rep"},{"key":"2022092013213390700_ref10","volume-title":"International Conference on Machine Learning (ICML)","author":"Trouillon","year":"2016"},{"issue":"D1","key":"2022092013213390700_ref11","doi-asserted-by":"crossref","first-page":"D995","DOI":"10.1093\/nar\/gkw1072","article-title":"Pharos: collating protein information to shed light on the druggable genome","volume":"45","author":"Nguyen","year":"2017","journal-title":"Nucleic Acids Res"},{"key":"2022092013213390700_ref12","doi-asserted-by":"crossref","first-page":"793","DOI":"10.1145\/3292500.3330961","volume-title":"Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","author":"Zhang","year":"2019"},{"key":"2022092013213390700_ref13","author":"Newman","year":"2018"},{"key":"2022092013213390700_ref14","doi-asserted-by":"crossref","DOI":"10.1109\/TPAMI.2021.3124805","article-title":"Bringing light into the dark: A large-scale evaluation of knowledge graph embedding models under a unified framework","author":"Ali","year":"2021","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"2022092013213390700_ref15","volume-title":"International Conference on Learning Representations","author":"Ruffinelli","year":"2019"},{"key":"2022092013213390700_ref16","article-title":"A survey on knowledge graphs: Representation, acquisition, and applications","author":"Ji","year":"2021","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"issue":"12","key":"2022092013213390700_ref17","doi-asserted-by":"crossref","first-page":"2724","DOI":"10.1109\/TKDE.2017.2754499","article-title":"Knowledge graph embedding: A survey of approaches and applications","volume":"29","author":"Wang","year":"2017","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"issue":"5439","key":"2022092013213390700_ref18","doi-asserted-by":"crossref","first-page":"509","DOI":"10.1126\/science.286.5439.509","article-title":"Emergence of scaling in random networks","volume":"286","author":"Barab\u00e1si","year":"1999","journal-title":"Science"},{"key":"2022092013213390700_ref19","doi-asserted-by":"crossref","first-page":"1423","DOI":"10.1145\/3485447.3512189","volume-title":"Proceedings of the ACM Web Conference 2022","author":"Wang","year":"2022"},{"key":"2022092013213390700_ref20","article-title":"Adversarial learning for debiasing knowledge graph embeddings","author":"Arduini","year":"2020"},{"key":"2022092013213390700_ref21","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1109\/ASONAM49781.2020.9381459","volume-title":"2020 IEEE\/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)","author":"Bourli","year":"2020"},{"key":"2022092013213390700_ref22","first-page":"1059","volume-title":"Conference on Uncertainty in Artificial Intelligence","author":"Mohamed","year":"2020"},{"key":"2022092013213390700_ref23","first-page":"2787","volume-title":"Advances in neural information processing systems","author":"Bordes","year":"2013"},{"key":"2022092013213390700_ref24","volume-title":"7th biennial conference on innovative data systems research. CIDR Conference","author":"Mahdisoltani","year":"2014"},{"key":"2022092013213390700_ref25","doi-asserted-by":"crossref","first-page":"975","DOI":"10.1145\/3340531.3411910","volume-title":"Proceedings of the 29th ACM International Conference on Information & Knowledge Management","author":"Liu","year":"2020"},{"key":"2022092013213390700_ref26","doi-asserted-by":"crossref","first-page":"1109","DOI":"10.1145\/3447548.3467276","volume-title":"Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining","author":"Liu","year":"2021"},{"key":"2022092013213390700_ref27","doi-asserted-by":"crossref","first-page":"1435","DOI":"10.1145\/3340531.3411872","volume-title":"Proceedings of the 29th ACM International Conference on Information & Knowledge Management","author":"Tang","year":"2020"},{"key":"2022092013213390700_ref28","volume-title":"Thirty-Fifth Conference on Neural Information Processing Systems","author":"Kojaku","year":"2021"},{"key":"2022092013213390700_ref29","first-page":"3289","volume-title":"Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence, IJCAI-19","author":"Rahman","year":"2019"},{"key":"2022092013213390700_ref30","article-title":"The probability of edge existence due to node degree: a baseline for network-based predictions","author":"Zietz","year":"2020","journal-title":"Manubot, Tech Rep"},{"key":"2022092013213390700_ref31","doi-asserted-by":"crossref","DOI":"10.7554\/eLife.26726","article-title":"Systematic integration of biomedical knowledge prioritizes drugs for repurposing","volume":"6","author":"Himmelstein","year":"2017","journal-title":"Elife"},{"key":"2022092013213390700_ref32","doi-asserted-by":"crossref","first-page":"83","DOI":"10.1016\/j.ymeth.2014.11.020","article-title":"DISEASES: Text mining and data integration of disease\u2013gene associations","volume":"74","author":"Pletscher-Frankild","year":"2015","journal-title":"Methods"},{"issue":"D1","key":"2022092013213390700_ref33","doi-asserted-by":"crossref","first-page":"D1005","DOI":"10.1093\/nar\/gky1120","article-title":"The NHGRI-EBI GWAS Catalog of published genome-wide association studies, targeted arrays and summary statistics 2019","volume":"47","author":"Buniello","year":"2019","journal-title":"Nucleic Acids Res"},{"key":"2022092013213390700_ref34","first-page":"gkw943","article-title":"DisGeNET: a comprehensive platform integrating information on human disease-associated genes and variants","author":"Pi\u00f1ero","year":"2016","journal-title":"Nucleic Acids Res"},{"key":"2022092013213390700_ref35","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"Wang","year":"2014"},{"key":"2022092013213390700_ref36","volume-title":"International Conference on Learning Representations","author":"Sun","year":"2019"},{"key":"2022092013213390700_ref37","volume-title":"Proceedings of the International Conference on Learning Representations (ICLR)","author":"Yang","year":"2015"},{"issue":"2","key":"2022092013213390700_ref38","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3424672","article-title":"Knowledge graph embedding for link prediction: A comparative analysis","volume":"15","author":"Rossi","year":"2021","journal-title":"ACM Transactions on Knowledge Discovery from Data (TKDD)"},{"key":"2022092013213390700_ref39","first-page":"100036","article-title":"Understanding the performance of knowledge graph embeddings in drug discovery","volume-title":"Artificial Intelligence in the Life Sciences","author":"Bonner","year":"2022"},{"key":"2022092013213390700_ref40","doi-asserted-by":"crossref","first-page":"57","DOI":"10.18653\/v1\/W15-4007","volume-title":"Proceedings of the 3rd workshop on continuous vector space models and their compositionality","author":"Toutanova","year":"2015"},{"key":"2022092013213390700_ref41","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"Dettmers","year":"2018"},{"issue":"6","key":"2022092013213390700_ref42","doi-asserted-by":"crossref","first-page":"793","DOI":"10.1093\/jb\/mvq044","article-title":"Regulatory mechanisms involved in the control of ubiquitin homeostasis","volume":"147","author":"Kimura","year":"2010","journal-title":"The Journal of Biochemistry"},{"key":"2022092013213390700_ref43","doi-asserted-by":"crossref","DOI":"10.1093\/bioinformatics\/btaa274","article-title":"OpenBioLink: A benchmarking framework for large-scale biomedical link prediction","author":"Breit","year":"2020","journal-title":"Bioinformatics"},{"key":"2022092013213390700_ref44","doi-asserted-by":"crossref","first-page":"375","DOI":"10.1007\/978-3-030-77385-4_22","volume-title":"European Semantic Web Conference","author":"Liu","year":"2021"},{"key":"2022092013213390700_ref45","article-title":"Task-Driven Knowledge Graph Filtering Improves Prioritizing Drugs for Repurposing. Technical","author":"Ratajczak","year":"2021","journal-title":"Report"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/23\/5\/bbac279\/45937606\/bbac279.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/23\/5\/bbac279\/45937606\/bbac279.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,20]],"date-time":"2022-09-20T17:43:34Z","timestamp":1663695814000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbac279\/6649936"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,7,26]]},"references-count":45,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2022,9,20]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbac279","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2022,9]]},"published":{"date-parts":[[2022,7,26]]},"article-number":"bbac279"}}