FLUXMATERIA — ADMET
ADMET screening that tells you what it knows — and what it doesn't.
Run full developability panels on candidate sets. Get predictions with confidence indicators. Export decision packets for downstream review.
FLUXMATERIA — ADMET
Run full developability panels on candidate sets. Get predictions with confidence indicators. Export decision packets for downstream review.
Absorption, distribution, metabolism, excretion, toxicity endpoints
Physics-only predictions with no ML or AI correlations
Know how much to trust each prediction
Screen entire candidate sets efficiently
Decision packets with full provenance
Side-by-side candidate evaluation
Most ADMET tools give you a number. FluxMateria gives you a number and tells you how much to trust it.
Prediction is reliable for this chemical space. Proceed with screening logic.
Prediction is informative but verify experimentally for critical decisions.
Molecule is outside reliable prediction space. Consider alternative assays.
Paste SMILES, upload a list, or pull from Workspace
Select ADMET endpoints (or use default panel)
Batch computation with progress tracking
Inspect results with confidence highlighting
Side-by-side candidate evaluation
Decision packet with full provenance
Validated against curated datasets with verified labels from FDA, peer-reviewed literature, and quality-controlled databases.
| Module | Metric | FluxMateria | Validation Scale | Status |
|---|---|---|---|---|
| Permeability (Caco-2) | Pearson r | +0.964 | 40,318 compounds | ✓ Strong |
| Metabolism (CLint) | Pearson r | +0.753 | 38,482 compounds | ✓ Strong |
| PPB (Ensemble) | Pearson r | +0.714 | 14,031 compounds | ✓ Strong |
| hERG (Ensemble) | Pearson r | +0.436 | 5,280 compounds | ✓ Competitive |
| BBB Permeability | Accuracy | 81.8% | 7,803 compounds | ✓ Competitive |
| Solubility | logS MAE | 0.709 | 1,128 compounds | ✓ Competitive |
| DILI Hepatotoxicity | 3-Class Acc | 73.8% | 500 stratified | ✓ Competitive |
| CYP Inhibition (v4.1) | 5-Isoform Acc | 82.6% | 5 CYP isoforms | ✓ Strong |
253K+ compounds validated across 8 ensemble endpoints. Full interpretability preserved.
Single-threaded CPU performance. No GPU required.
| Endpoint | Throughput | Method |
|---|---|---|
| BBB, Solubility | ~300 mol/sec | Physics descriptors |
| DILI Hepatotoxicity | ~2 mol/sec | Similarity ensemble (143K ref) |
| Permeability (ensemble) | ~26 mol/sec | Tanimoto similarity (40K ref) |
| PPB (ensemble) | ~21 mol/sec | Tanimoto similarity (14K ref) |
| Metabolism (ensemble) | ~8 mol/sec | Tanimoto similarity (38K ref) |
| hERG (ensemble) | ~1.3 mol/sec | 3D docking + similarity |
All measurements single-threaded on CPU. Ensemble endpoints trade speed for accuracy via structural similarity to validated reference compounds.
No account required. Paste a SMILES string and see the full panel with confidence indicators.
Run Demo →Upload a CSV of molecules and see results across your candidate set.
Request Batch Access →