APRILE: Exploring the Molecular Mechanisms of Drug Side Effects with Explainable Graph Neural Networks
Bibliographic record
Abstract
Abstract The majority of people over the age of 65 take two or more medications. While many individual drug side effects are known, polypharmacy side effects due to novel drug combinations poses great risk. Here, we present APRILE: an explainable artificial intelligence (XAI) framework that uses graph neural networks to explore the molecular mechanisms underlying polypharmacy side effects. Given a list of side effects and the pairs of drugs causing them, APRILE identifies a set of proteins (drug targets or non-targets) and associated Gene Ontology (GO) terms as mechanistic ‘explanations’ of associated side effects. Using APRILE, we generate such explanations for 843,318 (learned) and 93,966 (novel) side effect–drug pair events, spanning 861 side effects (472 diseases, 485 symptoms and 9 mental disorders) and 20 disease cate-gories. We show that our two new metrics—pharmacogenomic information utilization and protein-protein interaction information utilization—provide quantitative estimates of mechanism complexity. Explanations were significantly consistent with state of the art disease-gene associations for 232/239 (97%) side effects. Further, APRILE generated new insights into molecular mechanisms of four diverse categories of adverse drug reactions: infection, metabolic diseases, gastrointestinal diseases, and mental disorders, including paradoxical side effects. We demonstrate the viability of discovering polypharmacy side effect mechanisms by training an XAI framework on massive biomedical data. Consequently, it facilitates wider and more reliable use of AI in healthcare.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".