DTI-Voodoo: machine learning over interaction networks and ontology-based background knowledge predicts drug–target interactions
Open accessThe DOI matches its Crossref record; 11 of the 11 references it cites were found in a registry or an index.
Abstract
Motivation: In silico drug–target interaction (DTI) prediction is important for drug discovery and drug repurposing. Approaches to predict DTIs can proceed indirectly, top-down, using phenotypic effects of drugs to identify potential drug targets, or they can be direct, bottom-up and use molecular information to directly predict binding affinities. Both approaches can be combined with information about interaction networks. Results: We developed DTI-Voodoo as a computational method that combines molecular features and ontology-encoded phenotypic effects of drugs with protein–protein interaction networks, and uses a graph convolutional neural network to predict DTIs. We demonstrate that drug effect features can exploit information in the interaction network whereas molecular features do not. DTI-Voodoo is designed to predict candidate drugs for a given protein; we use this formulation to show that common DTI datasets contain intrinsic biases with major effects on performance evaluation and comparison of DTI prediction methods. Using a modified evaluation scheme, we demonstrate that DTI-Voodoo improves significantly over state of the art DTI prediction methods. Availability and implementation: DTI-Voodoo source code and data necessary to reproduce results are freely available at https://github.com/THinnerichs/DTI-VOODOO. Supplementary information: Supplementary data are available at Bioinformatics online.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
The Gene Ontology resource: enriching a GOld mine
The Gene Ontology Consortium, Carbon, S., Douglass, E., Good, B. M., Unni, D. R., Harris, N. L., Mungall, C. J., Basu, S., Chisholm, R. L., Dodson, R. J., et al. (2021) · Nucleic Acids Research · doi:10.1093/nar/gkaa1113 · Cited by 3,555
The InterPro protein families and domains database: 20 years on
Blum, M., Chang, H.-Y., Chuguransky, S., Grego, T., Kandasaamy, S., Mitchell, A., Nuka, G., Paysan-Lafosse, T., Qureshi, M., Raj, S., et al. (2021) · Nucleic Acids Research · doi:10.1093/nar/gkaa977 · Cited by 2,231 · cited 1×
“We further employ another mechanism to avoid redundancy and fading signals in stacked graph convolutional networks, using residual connections and a normalization scheme (Li et al. , 2019, 2020) as shown in Supplementary Figure S3.”
Drug target identification using side-effect similarity
Campillos, M., Kuhn, M., Gavin, A.-C., Jensen, L. J., & Bork, P. (2008) · Science · doi:10.1126/science.1158140 · Cited by 1,248 · cited 1×
“Top-down methods use information about physiological effects of drugs for DTI prediction, such as side-effect similarity (Campillos et al., 2008), that is largely complementary to knowledge”
MethodExpansion of the Human Phenotype Ontology (HPO) knowledge base and resources
Köhler, S., Carmody, L., Vasilevsky, N., Jacobsen, J. O. B., Danis, D., Gourdine, J.-P., Gargano, M., Harris, N. L., Matentzoglu, N., McMurry, J. A., et al. (2019) · Nucleic Acids Research · doi:10.1093/nar/gky1105 · Cited by 677
Drug–target interaction prediction: databases, web servers and computational models
Chen, X., Yan, C. C., Zhang, X., Zhang, X., Dai, F., Yin, J., & Zhang, Y. (2016) · Briefings in Bioinformatics · doi:10.1093/bib/bbv066 · Cited by 658 · cited 2×
“Both approaches may be used in conjunction with network inference (Chen et al., 2015).”
MethodGuilt-by-association goes global
Oliver, S. (2000) · Nature · doi:10.1038/35001165 · Cited by 557 · cited 1×
“Network-based DTI prediction methods use the guilt-by-association principle (Oliver, 2000) and assume that a protein is a likely target for a drug if many of the protein’s neighbors in the interaction network are targets of the drug (Gillis and Pavlidis, 2012).”
MethodWan, F., Hong, L., Xiao, A., Jiang, T., & Zeng, J. (2019) · Bioinformatics · doi:10.1093/bioinformatics/bty543 · Cited by 381 · cited 2×
“Furthermore, we added DTINet (Luo et al. , 2017) as method for comparison which has been used to develop a number of methods such as NeoDTI (Wan et al. , 2019) with similar methodology.”
MethodThe mammalian phenotype ontology: enabling robust annotation and comparative analysis
Smith, C. L., & Eppig, J. T. (2009) · WIREs Systems Biology and Medicine · doi:10.1002/wsbm.44 · Cited by 353 · cited 2×
“We utilize the PhenomeNET ontology (Hoehndorf et al. , 2011), an ontology integrating ontologies such as the Human Phenotype Ontology (Köhler et al. , 2018), Gene Ontology (Ashburner et al. , 2000; Carbon et al. , 2020), Mammalian Phenotype Ontology (Smith and Eppig, 2009) and several others.”
BackgroundMethodA review of network-based approaches to drug repositioning
Lotfi Shahreza, M., Ghadiri, N., Mousavi, S. R., Varshosaz, J., & Green, J. R. (2018) · Briefings in Bioinformatics · doi:10.1093/bib/bbx017 · Cited by 330
“Guilt by Association” Is the Exception Rather Than the Rule in Gene Networks
Gillis, J., & Pavlidis, P. (2012) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1002444 · Cited by 214
Cited by
Artificial intelligence in cancer target identification and drug discovery
You, Y., Lai, X., Pan, Y., Zheng, H., Vera, J., Liu, S., Deng, S., & Zhang, L. (2022) · Signal Transduction and Targeted Therapy · doi:10.1038/s41392-022-00994-0 · Cited by 435
Interpretable bilinear attention network with domain adaptation improves drug–target prediction
Bai, P., Miljković, F., John, B., & Lu, H. (2023) · Nature Machine Intelligence · doi:10.1038/s42256-022-00605-1 · Cited by 362
Explainable artificial intelligence for drug discovery and development: a comprehensive survey
Alizadehsani, R., Oyelere, S. S., Hussain, S., Jagatheesaperumal, S. K., Calixto, R. R., Rahouti, M., Roshanzamir, M., & De Albuquerque, V. H. C. (2024) · IEEE Access · doi:10.1109/access.2024.3373195 · Cited by 124
Application of Artificial Intelligence In Drug-target Interactions Prediction: A Review
Liao, Q., Zhang, Y., Chu, Y., Ding, Y., Liu, Z., Zhao, X., Wang, Y., Wan, J., Ding, Y., Tiwari, P., et al. (2025) · npj Biomedical Innovations · doi:10.1038/s44385-024-00003-9 · Cited by 82
DTSyn: a dual-transformer-based neural network to predict synergistic drug combinations
Hu, J., Gao, J., Fang, X., Liu, Z., Wang, F., Huang, W., Wu, H., & Zhao, G. (2022) · Briefings in Bioinformatics · doi:10.1093/bib/bbac302 · Cited by 65
Zhang, P., Ma, J., & Chen, T. (2025) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbbio.2025.3576488 · Cited by 13
Metadata
- Authors
- Hinnerichs, T., & Hoehndorf, R.
- Venue
- Bioinformatics
- Publisher
- Oxford University Press (OUP)
- Published
- 2021-07-28
- Volume
- 37
- Issue
- 24
- Pages
- 4835-4843
- ISSN
- 1367-4803 (ISSN-L)
- 1367-4811
- OpenAlex
- W3158080037
- Cited by
- 26
- References
- 11
- FWCI
- 1.50
- Type
- journal article
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.