Predicting drug-drug interactions using heterogeneous graph neural networks: HGNN-DDI
Bibliographic record
Abstract
This research centers on predicting drug-drug interactions (DDIs) using a novel approach involving graph neural networks (GNNs) with integrated attention mechanisms. In this method, drugs and proteins are depicted as nodes within a heterogeneous graph. This graph is characterized by different types of edges symbolizing not only DDIs but also drug-protein interactions (DPIs) and protein-protein interactions (PPIs). To analyze the chemical structures of drugs, we employ a pretrained model named ChemBERTa, which utilizes simplified molecular input line entry system (SMILES) strings. The similarity between drug structures based on their SMILES strings is determined using the RDkit tool. Our model is designed to establish and link heterogeneous graph neural networks, taking into account the DPIs and PPIs as key input data. For the final prediction of interaction types between various drugs, we use the Multi-Layer Perception (MLP) technique. This model's primary objective is to enhance the accuracy of DDI predictions by factoring in additional data on both drug-protein and protein-protein interactions. The forecasted DDIs are presented with associated probabilities, offering valuable insights to healthcare professionals. These insights are crucial for assessing the potential risks and advantages of combining different drugs, particularly for patients with diseases at different stages of progression.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 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) | 0.001 | 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".