Knowledge graph prediction of unknown adverse drug reactions and validation in electronic health records
Open accessAbstract
Unknown adverse reactions to drugs available on the market present a significant health risk and limit accurate judgement of the cost/benefit trade-off for medications. Machine learning has the potential to predict unknown adverse reactions from current knowledge. We constructed a knowledge graph containing four types of node: drugs, protein targets, indications and adverse reactions. Using this graph, we developed a machine learning algorithm based on a simple enrichment test and first demonstrated this method performs extremely well at classifying known causes of adverse reactions (AUC 0.92). A cross validation scheme in which 10% of drug-adverse reaction edges were systematically deleted per fold showed that the method correctly predicts 68% of the deleted edges on average. Next, a subset of adverse reactions that could be reliably detected in anonymised electronic health records from South London and Maudsley NHS Foundation Trust were used to validate predictions from the model that are not currently known in public databases. High-confidence predictions were validated in electronic records significantly more frequently than random models, and outperformed standard methods (logistic regression, decision trees and support vector machines). This approach has the potential to improve patient safety by predicting adverse reactions that were not observed during randomised trials.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · see citing works
Incidence of Adverse Drug Reactions in Hospitalized Patients: A Meta-analysis of Prospective Studies
Lazarou, J., Pomeranz, B. H., & Corey, P. N. (1998) · JAMA · doi:10.1001/jama.279.15.1200 · Cited by 4,807 · see citing works
Adverse drug reactions as cause of admission to hospital: prospective analysis of 18 820 patients
Pirmohamed, M., James, S., Meakin, S., Green, C. A., SCOTT, A. K., Walley, T. J., Farrar, K., Park, B. K., & Breckenridge, A. M. (2004) · BMJ · doi:10.1136/bmj.329.7456.15 · Cited by 3,236 · see citing works
QSAR Modeling: Where have you been? Where are you going to?
Cherkasov, A., Muratov, E. N., Fourches, D., Varnek, A., Baskin, I. I., Cronin, M., Dearden, J., Gramatica, P., Martin, Y. C., Todeschini, R., et al. (2014) · Journal of Medicinal Chemistry · doi:10.1021/jm4004285 · Cited by 2,164 · see citing works
DrugBank 4.0: shedding new light on drug metabolism
Law, V., Knox, C., Djoumbou-Feunang, Y., Jewison, T., Guo, A. C., Liu, Y., Maciejewski, A., Arndt, D., Wilson, M., Neveu, V., et al. (2013) · Nucleic Acids Research · doi:10.1093/nar/gkt1068 · Cited by 1,995 · see citing works
Under-Reporting of Adverse Drug Reactions
Hazell, L., & Shakir, S. (2006) · Drug Safety · doi:10.2165/00002018-200629050-00003 · Cited by 1,938 · see citing works
The SIDER database of drugs and side effects
Kuhn, M., Letunic, I., Jensen, L. J., & Bork, P. (2016) · Nucleic Acids Research · doi:10.1093/nar/gkv1075 · Cited by 1,451 · see citing works
Whitebread, S., Hamon, J., Bojanic, D., & Urbán, L. (2005) · Drug Discovery Today · doi:10.1016/s1359-6446(05)03632-9 · Cited by 408 · see citing works
Liu, M., Wu, Y., Chen, Y., Sun, J., Zhao, Z., Chen, X.-W., Matheny, M. E., & Xu, H. (2012) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2011-000699 · Cited by 291 · see citing works
Predicting drug side-effect profiles: a chemical fragment-based approach
Pauwels, E., Stoven, V., & Yamanishi, Y. (2011) · BMC Bioinformatics · doi:10.1186/1471-2105-12-169 · Cited by 233 · see citing works
SSRI-Induced extrapyramidal side-effects and akathisia: implications for treatment
Lane, R. M. (1998) · Journal of Psychopharmacology · doi:10.1177/026988119801200212 · Cited by 213 · see citing works
Fernandes, A. C., Cloete, D., Broadbent, M. T. M., Hayes, R. D., Chang, C., Jackson, R., Roberts, A., Tsang, J., Soncul, M., Liebscher, J., et al. (2013) · BMC Medical Informatics and Decision Making · doi:10.1186/1472-6947-13-71 · Cited by 191 · see citing works
Self-citationDrug-induced adverse events prediction with the LINCS L1000 data
Wang, Z., Clark, N. R., & Ma’ayan, A. (2016) · Bioinformatics · doi:10.1093/bioinformatics/btw168 · Cited by 189 · see citing works
Harpaz, R., Vilar, S., DuMouchel, W., Salmasian, H., Haerian, K., Shah, N. H., Chase, H. S., & Friedman, C. (2012) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2012-000930 · Cited by 172 · see citing works
Drug Side-Effect Prediction Based on the Integration of Chemical and Biological Spaces
Yamanishi, Y., Pauwels, E., & Kotera, M. (2012) · Journal of Chemical Information and Modeling · doi:10.1021/ci2005548 · Cited by 144 · see citing works
Predicting adverse side effects of drugs
Huang, L., Wu, X., & Chen, J. Y. (2011) · BMC Genomics · doi:10.1186/1471-2164-12-s5-s11 · Cited by 133 · see citing works
Predicting Adverse Drug Events Using Pharmacological Network Models
Cami, A., Arnold, A. D., Manzi, S. F., & Reis, B. Y. (2011) · Science Translational Medicine · doi:10.1126/scitranslmed.3002774 · Cited by 131 · see citing works
Extrapyramidal Symptoms Associated With Cyclic Antidepressant Treatment
Gill, H. S., DeVane, C. L., & Risch, S. C. (1997) · Journal of Clinical Psychopharmacology · doi:10.1097/00004714-199710000-00007 · Cited by 107 · see citing works
Liu, M., Hinz, E. M., Matheny, M. E., et al. (2012) · Journal of the American Medical Informatics Association · doi:10.1136/amiajnl-2012-001119 · Cited by 79 · see citing works
Tricyclic antidepressant-induced extrapyramidal side effects.
Vandel, P., Bonin, B., Lévêque, E., Sechter, D., & Bízouard, P. (1997) · European Neuropsychopharmacology · doi:10.1016/s0924-977x(97)00405-7 · Cited by 73 · see citing works
Frid, A. A., & Matthews, E. J. (2009) · Regulatory Toxicology and Pharmacology · doi:10.1016/j.yrtph.2009.11.005 · Cited by 72 · see citing works
Integrative relational machine-learning for understanding drug side-effect profiles
Bresso, E., Grisoni, R., Marchetti, G., et al. (2013) · BMC Bioinformatics · doi:10.1186/1471-2105-14-207 · Cited by 67 · see citing works
Targeted high throughput sequencing in hereditary ataxia and spastic paraplegia
Iqbal, Z., Rydning, S. L., Wedding, I. M., Koht, J. A., Pihlstrøm, L., Rengmark, A. H., Henriksen, S. P., Tallaksen, C. M. E., & Toft, M. (2017) · PLoS ONE · doi:10.1371/journal.pone.0174667 · Cited by 62 · see citing works
Analysis of Chemical and Biological Features Yields Mechanistic Insights into Drug Side Effects
Duran‐Frigola, M., & Aloy, P. (2013) · Chemistry & Biology · doi:10.1016/j.chembiol.2013.03.017 · Cited by 57 · see citing works
Data-driven Approach to Detect and Predict Adverse Drug Reactions
Ho, T.-B., Le, L., Thai, D. T., et al. (2016) · Current Pharmaceutical Design · doi:10.2174/1381612822666160509125047 · Cited by 57 · see citing works
The Mechanism of Drug-induced Akathisia
LOONEN, A. J. M., & Stahl, S. M. (2011) · CNS Spectrums · doi:10.1017/s1092852912000107 · Cited by 56 · see citing works
Boland, M. R., Jacunski, A., Lorberbaum, T., et al. (2015) · WIREs Systems Biology and Medicine · doi:10.1002/wsbm.1323 · Cited by 48 · see citing works
Iqbal, E., Mallah, R., Rhodes, D. R., Wu, H., Romero, A., Chang, N., Dzahini, O., Pandey, C., Broadbent, M. T. M., Stewart, R. J., et al. (2017) · PLoS ONE · doi:10.1371/journal.pone.0187121 · Cited by 48 · see citing works
Self-citationGESSE: Predicting Drug Side Effects from Drug–Target Relationships
Pérez‐Nueno, V. I., Souchet, M., Karaboga, A. S., et al. (2015) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5b00120 · Cited by 25 · see citing works
ARWAR: A network approach for predicting Adverse Drug Reactions
Rahmani, H., Weiß, G., Méndez‐Lucio, O., & Bender, A. (2015) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2015.11.005 · Cited by 22 · see citing works
Thrombosis of cerebral veins following intravenous application of clomipramine.
Eikmeier, G., Kuhlmann, R., & Gastpar, M. (1988) · Journal of Neurology Neurosurgery & Psychiatry · doi:10.1136/jnnp.51.11.1461 · Cited by 11 · see citing works
Shao, Z., Hirayama, Y., Yamanishi, Y., et al. (2015) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.5b00376 · Cited by 9 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Artificial Intelligence for Drug Discovery: Are We There Yet?
Hasselgren, C., & Oprea, T. I. (2024) · Annual Review of Pharmacology and Toxicology · doi:10.1146/annurev-pharmtox-040323-040828 · Cited by 236 · see citing works
“incorporate side effect information from package inserts, target information, and indication information have shown promise in predicting adverse events (146).”
BackgroundThe Science of Prognosis in Psychiatry: A Review
Fusar-Poli, P., Hijazi, Z., Stahl, D., & Steyerberg, E. W. (2018) · JAMA Psychiatry · doi:10.1001/jamapsychiatry.2018.2530 · Cited by 197 · see citing works
- Authors
- Bean, D. M., Wu, H., Iqbal, E., Dzahini, O., Ibrahim, Z. M., Broadbent, M., Stewart, R., & Dobson, R. J. B.
- Affiliations
- King's College London (GB)
- South London and Maudsley NHS Foundation Trust (GB)
- Corresponding author
- Dobson, Richard J. B.
- ORCID
- Venue
- Scientific Reports
- Published
- 2017-11-27
- Volume
- 7
- Issue
- 1
- Pages
- 16416-16416
- ISSN
- 2045-2322 (ISSN-L)
- PMC
- PMC5703951
- OpenAlex
- W2769065924
- Cited by
- 127
- References
- 32
- Citations by year
- 2026: 14 · 2025: 23 · 2024: 16 · 2023: 18 · 2022: 22 · 2021: 11 · 2020: 14 · 2019: 8 · 2018: 1
- FWCI
- 4.68
- Type
- journal article
- Language
- en
- Update notices
- correction (2018-03-06)
- Open access
- gold
- doi.org
- Locations
- Scientific Reports (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Research Portal (King's College London) (submittedVersion, cc-by)
- Edinburgh Research Explorer (University of Edinburgh) (submittedVersion, other-oa)
- UCL Discovery (University College London) (submittedVersion)
- Funders
- National Institute for Social Care and Health Research (MR/K006584/1)
- South London and Maudsley NHS Foundation Trust
- Wellcome Trust (MR/K006584/1)
- Cancer Research UK
- National Institute for Health and Care Research (RP-PG-0407-10314, 05/40/04, NF-SI-0616-10066)
- British Heart Foundation
- King's College London
- University College London
- Versus Arthritis
- Medical Research Council (MR/K006584/1, K006584/1, MC_PC_13041)
- Engineering and Physical Sciences Research Council
- Economic and Social Research Council
- Topics
- Life Sciences › Pharmacology, Toxicology and Pharmaceutics › Toxicology › Pharmacovigilance and Adverse Drug Reactions
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Computational Drug Discovery Methods
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Biomedical Text Mining and Ontologies
- Keywords
- knowledge graphs, protein targets, drug indications, electronic health records, patient safety
- MeSH terms
- Machine Learning, Algorithms, Humans, Prognosis, Reproducibility of Results, Databases, Factual, Knowledge Bases, Electronic Health Records, Public Health Surveillance, Drug-Related Side Effects and Adverse Reactions
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
