Scalient Bio — AlloTox
Alzheimer's research · AI toxicology

Predicting which preclinical
studies are worth running.

Scalient Bio is building AlloTox — the prediction engine for Alzheimer's preclinical toxicology. We harmonise every available model into one ranked decision. 30% of the time, neither rodent nor non-rodent model predicts human toxicity. We built the layer that asks: will this study tell you anything you don't already know?

£22k
annual triage cost with AlloTox — vs £87k with expert judgement alone (4.0× ours)
0.42%
catch rate needed to break even — one correct triage in 237 pays for itself
8
independent evidence classes — more than any competitor in the market

AlloTox runs your pre-clinical evidence and drug lead through a pre-harmonised chain of prediction engines — literature agents, QSAR toolboxes, PK simulators — then synthesises the disagreement between them into a single auditable Go / No-go report. The signal is disagreement. Every new model a competitor ships makes our product better, not worse.

AD NME — Alzheimer's new molecular entity
Small molecule NME
All NME
All drugs — fail to reach market

Despite massive investments, most drugs still never make it to market.

AD candidates have the worst success rate of any drug class. The bottleneck is not the science — it is the allocation. Which study gets the animal slot?

Our answer

Meet Scalient Bio —
the prediction engine.

Meet the team.

Scientific & regulatory lead
Sophie Bloom
SB
AI architecture & technical lead
Adrian Asheri
AA
Data analytics & evidence
Arina Redina
AR