Liver Injury
Accurately predict drug induced liver toxicity
20-25% of clinical trials fail due to drug-induced liver injury, Axiom wants to eliminate these failures
 Pfizer 2D HepG2 |  Axiom AI model | |
|---|---|---|
| ROC-AUC | -- | 0.89 |
| Sensitivity / True Positive Rate | 34% | 75% |
| Specificity / True Negative Rate | 91% | 90% |
Massive proprietary data training large-scale AI models
We did the lab work to build the world's largest human liver dataset then used it to train highly accurate models.
Step 1
Automated lab created the world’s largest primary human liver dataset
-
130,000+ molecules exposed to primary human liver cells.
-
High content imaging of key cell organelles paired with biochemical assays.
-
Learn more about Axiom's training set.
Step 2
Quantify biology with unprecedented sensitivity
-
10+ models of unique biology (confluency, ER-stress, mitochondrial toxicity, and more).
-
Bile canalicular networks, efflux/uptake transporters, steatosis coming soon.
-
Points of departure dramatically more sensitive than traditional assays.
Step 3
Learn how chemistry affects biology
-
Molecules induce diverse cellular phenotypes.
-
Axiom's models learn which molecules induce which toxic cell phenotypes.
-
For a new molecule, our models predict its cellular phenotype and dose-dependent relationship with toxicity.
Step 4
Precise clinical risk assessment
-
Understand the relationship between human exposure and human toxicity.
-
Accurately compute the therapeutic index.
-
Reason about toxicity at physiologically relevant concentrations.
Inside our model: how it works
Axiom uses molecular structure, properties, and cell biology to predict clinical toxicity outcomes.
Assess toxicity in a modern web application
Study clinical liver injury risk with powerful data visualization and predictions.
Liver Services
Axiom offers a variety of services for understanding drug induced liver injury.
