EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Node-degree aware edge sampling mitigates inflated classification performance in biomedical random walk-based graph representation learning

Luca Cappelletti iD, Lauren Rekerle iD, Tommaso Fontana iD, Peter Hansen iD, Elena Casiraghi iD, Vida Ravanmehr iD, Christopher J Mungall iD, Jeremy J Yang iD, Leonard Spranger iD, Guy Karlebach iD, J Harry Caufield iD, Leigh Carmody iD und 5 weitere

Bioinformatics Advances · 05.01.2024 · 4 (1)

DOI: 10.1093/bioadv/vbae036 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Bioinformatics Advances, online erschienen am 04.03.2024. Verfasst von Luca Cappelletti und 16 weiteren Autor:innen. Beteiligte Einrichtung: AnacletoLab, Dipartimento di Informatica, Università degli Studi di Milano, Milano 20133, Italy. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, National Institutes of Health, National Cancer Institute.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Motivation Graph representation learning is a family of related approaches that learn low-dimensional vector representations of nodes and other graph elements called embeddings. Embeddings approximate characteristics of the graph and can be used for a variety of machine-learning tasks such as novel edge prediction. For many biomedical applications, partial knowledge exists about positive edges that represent relationships between pairs of entities, but little to no knowledge is available about negative edges that represent the explicit lack of a relationship between two nodes. For this reason, classification procedures are forced to assume that the vast majority of unlabeled edges are negative. Existing approaches to sampling negative edges for training and evaluating classifiers do so by uniformly sampling pairs of nodes. Results We show here that this sampling strategy typically leads to sets of positive and negative examples with imbalanced node degree distributions. Using representative heterogeneous biomedical knowledge graph and random walk-based graph machine learning, we show that this strategy substantially impacts classification performance. If users of graph machine-learning models apply the models to prioritize examples that are drawn from approximately the same distribution as the positive examples are, then performance of models as estimated in the validation phase may be artificially inflated. We present a degree-aware node sampling approach that mitigates this effect and is simple to implement. Availability and implementation Our code and data are publicly available at https://github.com/monarch-initiative/negativeExampleSelection.

Autor:innen und Institutionen

  • AnacletoLab, Dipartimento di Informatica, Università degli Studi di Milano, Milano 20133, Italy.
  • The Jackson Laboratory for Genomic Medicine, CT 06032, United States.
  • Division of Environmental Genomics and Systems Biology, Lawrence Berkeley National Laboratory, Berkeley, CA 94710, United States.
  • Department of Internal Medicine and UNM Comprehensive Cancer Center, UNM School of Medicine, Albuquerque, NM 87102, United States.
  • Institute of Bioinformatics, Freie Universität Berlin, Berlin, 14195, Germany.
  • Institute for Systems Genomics, University of Connecticut, Farmington, CT 06032, United States.

ORCID-Profile: Luca Cappelletti, Lauren Rekerle, Tommaso Fontana, Peter Hansen, Elena Casiraghi, Vida Ravanmehr, Christopher J Mungall, Jeremy J Yang, Leonard Spranger, Guy Karlebach, J Harry Caufield, Leigh Carmody, Ben Coleman, Tudor I Oprea, Justin Reese, Giorgio Valentini, Peter N Robinson

Förderung

  • NCI NIH HHS
  • National Institutes of Health
  • National Cancer Institute

Zitieren

APA

Cappelletti, L., Rekerle, L., Fontana, T., Hansen, P., Casiraghi, E., Ravanmehr, V., Mungall, C. J., Yang, J. J., Spranger, L., Karlebach, G., Caufield, J. H., Carmody, L., Coleman, B., Oprea, T. I., Reese, J., Valentini, G., & Robinson, P. N. (2024). Node-degree aware edge sampling mitigates inflated classification performance in biomedical random walk-based graph representation learning. Bioinformatics Advances, 4(1). https://doi.org/10.1093/bioadv/vbae036

Vancouver

Cappelletti L, Rekerle L, Fontana T, Hansen P, Casiraghi E, Ravanmehr V, et al. Node-degree aware edge sampling mitigates inflated classification performance in biomedical random walk-based graph representation learning. Bioinformatics Advances. 2024;4(1). doi:10.1093/bioadv/vbae036

BibTeX
@article{Cappelletti2024,
  author = {Cappelletti, Luca and Rekerle, Lauren and Fontana, Tommaso and Hansen, Peter and Casiraghi, Elena and Ravanmehr, Vida and Mungall, Christopher J and Yang, Jeremy J and Spranger, Leonard and Karlebach, Guy and Caufield, J Harry and Carmody, Leigh and Coleman, Ben and Oprea, Tudor I and Reese, Justin and Valentini, Giorgio and Robinson, Peter N},
  title = {Node-degree aware edge sampling mitigates inflated classification performance in biomedical random walk-based graph representation learning},
  journal = {Bioinformatics Advances},
  year = {2024},
  volume = {4},
  number = {1},
  doi = {10.1093/bioadv/vbae036}
}
Kostenlos starten →