ToxSpotter is an AI platform that predicts 27 toxicity endpoints for any molecule you can draw. It was developed to make medicines safe for astronauts on the Moon, Mars and deep-space missions. Now available to de-risk your candidates here on Earth.
Nezu Biotech is part of the European Space Agency Business Incubation Centre (ESA-BIC).
Toxicity is one of the leading reasons candidate molecules are abandoned, and it is usually discovered after the money is already spent: after weeks of in vivo work.
ToxSpotter models are trained on carefully curated, high-quality toxicity data and validated against the reference assays they replicate, delivering high agreement with laboratory outcomes across our endpoint panel.
Twenty-seven endpoints spanning nine toxicity families: cardiac, hepatic, genotoxic, endocrine, cellular stress, acute systemic, dermal sensitization, environmental, and dosing. All assessed in one submission.
We process thousands of structures at a time. Screening that would occupy a laboratory for months returns as a finished report, so your series ranking is ready while the chemistry is still flexible.
In silico screening costs a fraction of a comparable assay panel, which makes it viable at the stage where it changes the most: before synthesis, before scale-up, before you commit a budget to a molecule.
You receive a visual report your whole team can read and a detailed table your toxicologists can interrogate. No software to install, no dashboard to learn, no data-science hire required.
Every molecule is positioned against thousands of approved drugs, and against those approved for the same indication as yours. Instead of asking “is this score bad?”, you can see exactly where your candidate sits among medicines that already passed.
Cardiac
Liver
Genotoxicity & DNA damage
Dosing
Endocrine & nuclear receptors
Cellular stress
Skin sensitization
Acute systemic toxicity
Environmental toxicity
Bring us a handful of molecules and a question you can't currently answer. We'll walk you through what a ToxSpotter report would tell you about them.
ToxSpotter is a decision-support tool for the discovery stage, not a regulatory submission package. Regulatory filings such as an IND or CTA require GLP-compliant laboratory studies under ICH M3(R2) and equivalent guidance, and no in silico method currently substitutes for them.
No. You send structures; we return a report written for chemists, toxicologists and project leads. All modelling happens on our side, and a specialist walks you through your first report.
Never. Client structures are kept strictly separate from our training data and are not used to improve our models, under written confidentiality terms agreed before you send anything.
Chemical structures (SMILES or SDF), and if you want indication-matched benchmarking, the therapeutic area you're working in. Nothing about your assays, your formulations or your programme is required.
From a single compound to thousands. Large series are where ToxSpotter earns the most, because ranking a whole library is exactly what the laboratory can't afford to do.
Performance is reported per endpoint, not as one headline figure, because endpoints differ in how well they're modelled. Across our panel, agreement with the corresponding laboratory assays runs at 70% - 95%.
Predictions are strongest for molecules within the chemical space the models were trained on. For highly unusual structures, confidence drops, and we tell you when a molecule sits outside that space rather than reporting a number as if it were certain. Those compounds are good candidates for early lab confirmation.
All of them. The endpoint panel covers dermal sensitization and environmental toxicity alongside pharmaceutical endpoints, which makes it directly useful for cosmetic ingredients, agrochemicals and industrial chemicals as well as drug candidates.
ToxSpotter was developed as part of the ESA Business Incubation Centre, for a setting where a medicine's safety has to be established without a hospital, a resupply flight or a second chance. That constraint shaped how conservatively we build and validate the models. Terrestrial pipelines get the benefit.
Processing is fast; the schedule is set by scope and agreed upfront during onboarding. Most engagements deliver in days rather than the months a comparable assay panel would take.
A specialist walks you through it, and you keep both the visual report and the full data table. Many clients come back with the next series once they've seen how the ranking held up in the lab.
Nezu Biotech GmbH – Quality, Innovation and Speed
Nezu Biotech GmbH
Tiergartenstraße 15
69121 Heidelberg
Germany
Amtsgericht Mannheim, HRB 753340
Steuernummer: 32496 / 39686
Geschäftsführer
Dr. Tiago Jose da Silva Lopes
Founder & CEO
www.nezubiotech.com/contact/
