Identification of microbe–disease signed associations via multi-scale variational graph autoencoder based on signed message propagation
MSignVGAE achieves AUROC and AUPR values of 0.9742 and 0.9601, respectively. Case studies on three diseases demonstrate that MSignVGAE can effectively capture a comprehensive distribution of associations by leveraging signed information.
Authors
- Hongxia Hao (ORCID: https://orcid.org/0000-0002-2324-0803)
- Liang Yu (ORCID: https://orcid.org/0000-0002-8351-3332)
- Huan Zhu
Institutions
- Xidian University (CN)
Publication Details
- Journal
- BMC Biology
- Published
- 2024-08-15
- DOI
- https://doi.org/10.1186/s12915-024-01968-0
- Citations
- 47
- Primary Topic
- Bioinformatics and Genomic Networks
- Type
- article
- Field-Weighted Citation Impact
- 10.99
Funders
- National Natural Science Foundation of China