OpenBioLink: a benchmarking framework for large-scale biomedical link prediction

Breit, Anna; Ott, Simon; Agibetov, Asan; Samwald, Matthias

doi:10.1093/bioinformatics/btaa274

Cited by 50 publications

(52 citation statements)

References 8 publications

(8 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We demonstrated that this approach outperforms competing methods, and more detailed analysis showed it is particularly beneficial when the data attributes are informative. Future work will evaluate the efficacy of our approach for more comprehensive knowledge graphs, such as the recent OpenBioLink (Breit et al , 2020) dataset. Further, it is well known (Kadlec et al , 2017) that hyperparameter tuning can lead to significant performance gains for many methods; thus, future work will also incorporate hyperparameter optimization strategies.…”

Section: Discussionmentioning

confidence: 99%

“…From a practical point of view, it is expensive to identify novel relations and most of them are still unknown. Recently, a variety of similar biomedical KGs have been proposed, such as (Breit et al, 2020) and (Mohamed et al, 2020). Due to the similarity of the source databases, we do not further consider those here; nevertheless, DOUBLER can be directly applied to those KGs, as well.…”

Section: Biomedical Knowledge Graphmentioning

confidence: 99%

See 1 more Smart Citation

DOUBLER: Unified Representation Learning of Biological Entities and Documents for Predicting Protein–Disease Relationships

Sztyler¹,

Malone²

2020

Preprint

View full text Add to dashboard Cite

Motivation: We propose a system that learns consistent representations of biological entities, such as proteins and diseases, based on a knowledge graph and additional data modalities, like structured annotations and free text describing the entities. In contrast to similar approaches, we explicitly incorporate the consistency of the representations into the learning process. In particular, we use these representations to identify novel proteins associated with diseases; these novel relationships could be used to prioritize protein targets for new drugs. Results: We show that our approach outperforms state-of-the-art link prediction algorithms for predicting unknown protein-disease associations. Detailed analysis demonstrates that our approach is most beneficial when additional data modalities, such as free text, are informative. Availability: Code and data are available at: https://github.com/nle-sztyler/research-doubler

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Biomedical Knowledge Graphmentioning

confidence: 99%

DOUBLER: Unified Representation Learning of Biological Entities and Documents for Predicting Protein–Disease Relationships

Sztyler¹,

Malone²

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Third, in the field of research, professional findings and co-authors interactions in any research area [9]. Certainly, the link prediction method can be used in biology and biomathematics [6], for example, in gene expression network and health care a, specialists predicting about outmost probability which reversals near future possibilities and managing them in proportion by communication with related people through online and off-line social network.…”

Section: Figure 1 Example Of Link Predictionmentioning

confidence: 99%

Structural Importance-based Link Prediction Techniques in Social Network

Samad¹,

Azam²,

Qadir³

2021

EAI Endorsed Transactions on Industrial Networks and Intelligen

View full text Add to dashboard Cite

“…The first network, OpenBioLink, is a large-scale KG generated from an integrative effort designed to establish a benchmark dataset for link prediction [39]. The second is an In-House network that is comprised of tens of thousands of interactions from eight databases that we have harmonized for this work including PathMe [40][41][42][43], BioGrid [44], IntAct [45], and PathwayCommons [46] for protein-protein relations, DrugBank [47] for drug-protein relations, and DisGeNet [26] for protein-indication interactions.…”

Section: Case Scenariosmentioning

confidence: 99%

Drug2ways: Reasoning over causal paths in biological networks for drug discovery

et al. 2020

View full text Add to dashboard Cite

Elucidating the causal mechanisms responsible for disease can reveal potential therapeutic targets for pharmacological intervention and, accordingly, guide drug repositioning and discovery. In essence, the topology of a network can reveal the impact a drug candidate may have on a given biological state, leading the way for enhanced disease characterization and the design of advanced therapies. Network-based approaches, in particular, are highly suited for these purposes as they hold the capacity to identify the molecular mechanisms underlying disease. Here, we present drug2ways, a novel methodology that leverages multimodal causal networks for predicting drug candidates. Drug2ways implements an efficient algorithm which reasons over causal paths in large-scale biological networks to propose drug candidates for a given disease. We validate our approach using clinical trial information and demonstrate how drug2ways can be used for multiple applications to identify: i) single-target drug candidates, ii) candidates with polypharmacological properties that can optimize multiple targets, and iii) candidates for combination therapy. Finally, we make drug2ways available to the scientific community as a Python package that enables conducting these applications on multiple standard network formats.

show abstract

OpenBioLink: a benchmarking framework for large-scale biomedical link prediction

Cited by 50 publications

References 8 publications

DOUBLER: Unified Representation Learning of Biological Entities and Documents for Predicting Protein–Disease Relationships

DOUBLER: Unified Representation Learning of Biological Entities and Documents for Predicting Protein–Disease Relationships

Structural Importance-based Link Prediction Techniques in Social Network

Drug2ways: Reasoning over causal paths in biological networks for drug discovery

Contact Info

Product

Resources

About