DDI — Model Performance

Link-prediction evaluation · DrugBank interaction graph

System online

Model Performance Evaluation

4,795 drugs · 824,249 DDI pairs · DrugBank v5.1

Why cold-start evaluation is what matters here
The GNN's real-world purpose is to predict interactions for drugs with no documented DrugBank pairs — newly approved drugs, rare compounds, or combinations simply never studied together. In a standard 80/20 edge split every test drug still appears in the training graph, so graph heuristics exploit dense connectivity (average degree ~344) and score ~0.97 AUC — not because they generalise, but because the graph is so dense that almost any two test drugs share common neighbours.

In cold-start evaluation, 10 % of drugs are held out entirely — they have zero training edges. Common Neighbours, Adamic-Adar and Jaccard score exactly 0 for every cold pair (AUC = 0.50, pure chance — analytically guaranteed). Logistic Regression survives because it uses node features (CYP450, ATC codes, physicochemical properties) that exist independently of graph topology. The GNN's advantage is to combine those features with partial graph context from nearby warm drugs — AUROC 0.9175 vs LR's 0.8974.
Cold-Start Evaluation — The Real Benchmark
10 % of drugs held out completely (zero training edges) · simulates newly approved drugs entering the formulary
Held-out drugs: 284 / 2,847 (10 %) Cold pairs: 158,642 pos + 158,642 neg Seed: 42 Pos:neg: 1:1
Model Category Cold-Start AUC-ROC Performance Avg Precision
Graph Heuristics (all)
Common Neighbors · Adamic-Adar · Jaccard
graph heuristic 0.5000 (collapses)
0.50
0.5000
Logistic Regression
node features only (CYP450, ATC, physicochemical)
non-graph ML 0.8974
0.8974
0.9032
GNN — HeteroGraphSAGE + NCN
features + graph context · this model
GNN 0.9175
0.9175
0.8824
Graph heuristics evaluate to exactly 0 for cold pairs — cold drugs have no training neighbours by definition, so Common Neighbors = Adamic-Adar = Jaccard = 0 for every pair. AUC = 0.50 is analytically guaranteed.
GNN advantage over LR: AUROC +0.0201 · AUPR −0.0208 (the AUPR gap reflects the difficulty of ranking novel interactions without prior graph context). GNN wins on AUROC, which is the primary metric for ranking novel pairs.
Warm evaluation — for reference only
Standard 80/20 Edge Split (Transductive)
All test drugs appear in the training graph · 80 % train / 20 % test · 1:1 pos:neg · 20 % masked positives (nnPU) removed before split
⚠ Context only — high heuristic scores reflect graph density (avg degree ~344), not generalisation
Model Category Warm AUC-ROC Bar Avg Precision
Feature Cosine Similarity no-graph baseline 0.6041
0.6041
0.5952
Degree Product graph heuristic 0.9534
0.9534
0.9383
Logistic Regression non-graph ML 0.9570
0.9570
0.9474
Common Neighbors graph heuristic 0.9738
0.9738
0.9648
Adamic-Adar graph heuristic 0.9748
0.9748
0.9660
Jaccard graph heuristic 0.9763
0.9763
0.9669
Homo GraphSAGE V1 GNN 0.9615
0.9615
0.9450
Hetero GraphSAGE V1
no CN pooling
GNN 0.9726
0.9726
0.9586
Hetero GraphSAGE + NCN ★
this model
GNN 0.9738
0.9738
0.9589
Graph heuristics score ≥0.97 here because DrugBank is extremely dense (avg degree ~344). Topology-based methods are competitive on warm eval because nearly every drug pair shares common neighbours — this is a dataset property, not generalisation. Feature Cosine (no graph, raw drug features only) scores 0.60 — confirming the graph structure is what drives heuristic performance, not the features alone.
GNN model evolution
Architecture Progression — Warm AUROC / AUPR
Each version adds one structural improvement over the previous
VersionKey additionWarm AUROCWarm AUPRΔ AUROC
Homo GraphSAGE V1 Homogeneous DDI graph only 0.96150.9450
Hetero GraphSAGE V1 + drug→protein edges (hetero graph) 0.97260.9586 +0.0111
Hetero + NCN (ours) ★ + CN pooling of shared DDI neighbours & shared protein targets 0.97380.9589 +0.0012
The heterogeneous graph (drug-protein edges) adds +0.0111 AUROC over the homogeneous baseline. CN pooling (NCN decoder) adds a further +0.0012 — the incremental gain is modest on warm eval but the decoder provides richer mechanistic context for cold-start pairs where common neighbours from shared protein targets still exist.
Prior work comparison
Published Benchmarks vs This Work
Note: datasets and evaluation protocols differ — direct numerical comparison is indicative only
ModelDataset scaleAUROCAUPRNotes
LLM-DDI ~4,000 drugs 0.95710.7346 LLM text encoding; low AUPR
ACDGNN ~500–1,500 drugs 0.98350.9881 Confirmed negatives; smaller, denser subset
Ours (Hetero + NCN) ★ 4,795 drugs · PU learning 0.97380.9589 No confirmed negatives; largest scale
ACDGNN operates on a smaller, confirmed-negative subset (~500–1,500 drugs) which makes the task easier than our setting. LLM-DDI has very low AUPR (0.7346) — we exceed it by +0.2243. Our model operates at full scale (4,795 drugs) under PU learning (no confirmed negatives), which is structurally harder. Feature Cosine similarity alone scores 0.60 AUROC — confirming the +0.37 AUROC contribution of graph structure.
Error analysis — threshold 0.43
Confusion Matrix + Degree-Based Error Analysis
Test set · 131,944 pairs (65,972 pos + 65,972 neg) · threshold 0.43 · warm evaluation
Confusion Matrix
55,406
True Positives
3,055
False Positives
10,566
False Negatives
62,917
True Negatives
Precision: 0.9477
Recall: 0.8398
Avg Combined Degree by Error Type
Error typeAvg degreeRelative
True Positives2,292
False Negatives2,168
False Positives1,851
False Negatives have slightly lower avg degree than True Positives (2,168 vs 2,292). High-degree drugs are also missed, indicating the model struggles with specific interaction patterns rather than purely data-sparse drugs.
False Positives have the lowest avg degree (1,851) — the model over-predicts for moderately connected drugs that share structural or topological similarity without documented DDIs. This suggests pharmacological complexity (multiple CYP450 targets, transporters) is harder to predict than data sparsity. The threshold 0.43 was calibrated on a held-out sample; clinical deployment would require prospective recalibration.
Evaluation Notes
💬