FAQ
Clear answers to common questions.
What EPiQ™ and PharmPhax™ do, how the performance figures are measured, who owns generated molecules, and where the pipeline stands.
Company
Radiopharmaceutical development is slow, expensive, and fragmented. Critical questions about target suitability, binding affinity, off-target interactions, biodistribution, organ exposure, and cell-specific toxicity are often evaluated through separate experimental workflows. Epitorix integrates these decisions into an AI-driven discovery and de-risking ecosystem designed to identify stronger candidates and potential development risks earlier.
Three audiences, one set of models. Developers and researchers design and rank peptides, antibodies, nanobodies and mini-proteins in EPiQ and Xenoclea. Investigators use PharmPhax and PharmacoLaw to expose biological, clinical, patent and regulatory risk before capital is committed. Investors and strategic partners back the Therapeutic Pipeline — four AI-designed programs now being tested in vivo — or the Epitorix AI models beneath them.
A platform company whose own programs are the proof. Epitorix builds proprietary AI models and the products on them — EPiQ with Xenoclea for design, PharmPhax and PharmacoLaw for de-risking — and runs a Therapeutic Pipeline of four AI-designed programs, generated, ranked and computationally de-risked with those same models.
The programs are where the models meet experiment: the lead candidates are now being tested in vivo. Investors and strategic partners can back a single program, or the models and infrastructure beneath all four.
Epitorix designs radiopharmaceuticals with its own models, offers the products built on them, and runs its own programs. AI for pharmaceutical discovery, development, and de-risking: Epitorix builds its own models, the products on them, and the drug programs that put both to the test. Specialized in radiopharmaceuticals.
Platform
EPiQ™ is Epitorix's modular AI workspace for target discovery, binding-site analysis, molecular generation, affinity ranking, optimization, off-target prediction, cell-type risk assessment, organ-risk prediction, and biodistribution modeling. Users can apply individual modules or connect them into an end-to-end workflow.
Xenoclea is the conversational AI orchestrator built into EPiQ, and the entry point for scientists who want Epitorix's models without touching infrastructure. Describe what you need to bind in plain language; Xenoclea plans the work across Epitorix's models and returns ranked designs with predicted 3D structures and the data behind them. Nothing to configure or install.
EPiQ itself is the modular workspace: direct access to each model, run alone or connected into one end-to-end workflow. Xenoclea also ranks — given 100 generated ligands, from any generative model or your own library, it recovers more than 80% of the true top ligands in internal benchmarks.
Fifteen generator engines run in EPiQ, selected per target rather than defaulted to one: BoltzProt1, RFpeptides, PepFlow, HYDRA, DiffPepBuilder, ColabDesign, BoltzGen, BoltzDesign1, CyclicMPNN, PepHAR, EpitorixRL, EvoDiff, PepTune, EpitorixDesignPeptide and EpitorixDesignPeptideSC. Published engines and Epitorix's own — EpitorixRL, EpitorixDesignPeptide and EpitorixDesignPeptideSC among them — sit side by side.
Generation is one stage. EPiQ's models also cover target discovery, binding-site analysis, affinity ranking, optimization, off-target prediction, cell-type and organ risk, and biodistribution modeling.
Yes. Eligible academics receive unlimited access to Epitorix's top-tier models. Ask about eligibility through the contact form.
The same models design the candidate, predict where it goes and what it hits, and rank what to test next. Every product runs on them.
Its own models design peptide, antibody, nanobody and mini-protein candidates, and the same models predict what would fail them — off-target binding, biodistribution, organ and cell-type risk, PK and PD — before a trial does.
Design and rank peptides, antibodies, nanobodies and mini-proteins in one workspace. Architectures designed by Epitorix, trained on data we generated ourselves and validated at the bench — for peptides, antibodies, nanobodies and mini-proteins.
Every class: small molecules, peptides, non-canonical peptides, macrocycles, antibodies, nanobodies, mini-proteins and DNA/RNA ligands.
Give Xenoclea 100 generated ligands — from any generative model, or from your own library — and in internal benchmarks it recovers more than 80% of your true top ligands.
Bring your own library or ask for new designs; either way, one ranked list. Bring candidates and a target, or ask for new designs, and one ranked list comes back with structures.
Products
PharmPhax™ is an industrial-grade translational de-risking report for existing radiopharmaceutical candidates. Using Epitorix AI models, it analyzes potential off-target interactions, biodistribution, PK, PD, organ exposure, and cell-type-specific risk to identify possible causes of failure before clinical trials begin.
In retrospective analyses of radiopharmaceuticals with documented clinical outcomes, our models detected previously reported development and safety liabilities with up to 93% accuracy. PharmPhax™ is designed to help companies investigate risks earlier, optimize candidates, and make more informed investment decisions before committing substantial capital to clinical development.
PharmacoLaw reads biology and law in one report. Biological and clinical risk sit beside patent, regulatory and legal exposure for a current or developing drug, so investigators and investors see the whole liability picture before committing capital. Everything in PharmPhax is included; both reports are requested through the contact form.
Choose EPiQ™ when you need direct access to modular AI models. Choose PharmPhax™ when you already have a candidate and need a structured translational-risk assessment. Choose EPITwin™ when you need a collaborative development program guided by Epitorix scientists. Choose Studio when you need scientifically rigorous molecular visualization, animation, or biological communication.
Testing a candidate for the liabilities that end programs — off-target binding, biodistribution, organ and cell-type exposure, PK and PD — before trials begin, so failures happen on paper instead.
PharmPhax™ is an industrial-grade translational de-risking report for existing radiopharmaceutical candidates. Using Epitorix AI models, it analyzes potential off-target interactions, biodistribution, PK, PD, organ exposure, and cell-type-specific risk to identify possible causes of failure before clinical trials begin.
The proprietary models, datasets and compute underneath everything Epitorix ships: architectures designed in house, trained on data we generate ourselves and validated at the bench. For investors and strategic partners who want the models rather than the applications.
Investors and strategic partners can back a single program, or the models and infrastructure beneath all four.
Science
Every figure on this site carries its basis in the same block: retrospective analyses and internal benchmarks for the model figures, in vitro binding assays for the pipeline Kd values. A preprint describing the evaluation is forthcoming.
In internal benchmarks, predicted pKd correlates with measured pKd at Pearson R 0.84 for peptides and 0.70 for small molecules, at about 1,000 molecules per second; and given 100 generated ligands, ranking recovers more than 80% of the true top ligands. In retrospective analyses, proteome-wide off-target prediction reaches over 90% recall on known off-target interactions in about an hour. In retrospective analyses of radiopharmaceuticals with documented clinical outcomes, our models detected previously reported development and safety liabilities with up to 93% accuracy.
Epitorix is a research and development platform. Its predictions are not intended for patient-level or clinical decision-making.
Four AI-designed radiopharmaceutical programs, each generated, ranked and computationally de-risked with Epitorix's own models — the same models offered through EPiQ and PharmPhax. Our lead candidates are now being tested in vivo.
The targets, the experimental affinities and the stage-by-stage status are held until reveal and will be published on the pipeline page. Investors and partners can request the pipeline data and methods under NDA through the contact form.
Design happens in software; confirmation happens at the bench. EPiQ connects to 10+ labs across the United States — Epi-labs — reachable from the workspace with one drag-and-drop, so design and experiment run in one loop.
Epitorix's in-house models are validated at the bench, and the lead candidates of the four pipeline programs are now being tested in vivo.
No. All four programs are preclinical and investigational: the lead candidates are now being tested in vivo, and the translational and regulatory step and a first clinical trial remain ahead.
Epitorix's pipeline programs are investigational and preclinical. No Epitorix product has been approved by the FDA, the EMA or any other regulator, and no clinical safety or efficacy has been established.
Radiopharmaceutical development is slow, expensive, and fragmented. Critical questions about target suitability, binding affinity, off-target interactions, biodistribution, organ exposure, and cell-specific toxicity are often evaluated through separate experimental workflows.
Toxicity often surfaces after trials begin. In retrospective analyses of radiopharmaceuticals with documented clinical outcomes, our models detected previously reported development and safety liabilities with up to 93% accuracy.
SPR and BLI measure it in months; molecular dynamics and FEP+ take hours per ligand pair. Epitorix predicts Kd at about 1,000 molecules per second — Pearson R 0.84 for peptides, 0.70 for small molecules, in internal benchmarks.
Pearson R 0.84 (peptides) and 0.70 (small molecules) for predicted vs measured pKd in internal benchmarks, at about 1,000 molecules per second.
A microarray is limited by what is on it. Epitorix screens binders, peptides and small molecules against the whole proteome computationally, in about an hour — over 90% recall on known off-target interactions.
In retrospective analyses, proteome-wide off-target prediction reaches over 90% recall on known off-target interactions in about an hour.
Biodistribution, organ and cell-type risk, PK and PD — predicted before a trial reveals them. The Epitorix route: Predict biodistribution & PK · Identify organ & cell-type risk · Optimize candidate before clinic · Advance to preclinical studies with greater confidence.
Epitorix is a research and development platform. Its predictions are not intended for patient-level or clinical decision-making.
Predicted organ and cell-type risk for one candidate: Kidneys (Proximal tubular cells) high; Liver (Hepatocytes) moderate; Marrow (Hematopoietic stem & progenitor cells) moderate; Salivary glands (Salivary acinar cells) low. Predicted risk for one candidate, schematic — the ranking is candidate-specific.
Phage and mRNA display take months and return noise. Epitorix generates ~10,000 site- and target-specific candidates — about six months saved, success rates up about 60% versus display libraries, in internal benchmarks.
Figures from retrospective analyses and internal benchmarks; a preprint describing the evaluation is forthcoming.
The difference is when you learn a molecule will fail: after 3–5 years and $5–10M, or after 4–6 weeks and up to $100K of in-silico design and de-risking.
Epitorix estimates. ‘Preclinical’ means a designed, ranked and de-risked candidate ready to enter preclinical testing; in vitro and in vivo studies, GLP toxicology, GMP manufacturing and regulatory review follow before any first-in-human study.
Ownership & use
As between Epitorix and the user, the user retains rights to the molecular outputs generated through their authorized EPiQ™ account, subject to the applicable platform agreement. Epitorix does not claim ownership of a user's generated outputs solely because they were created using the platform.
No. EPiQ™ is an AI-based generative platform, and AI models may independently generate identical or substantially similar molecules for different users, particularly when users provide similar targets, binding sites, constraints, or design objectives. Epitorix does not guarantee that every generated structure will be exclusive, novel, or patentable.
No. Epitorix is a research and development platform. Its predictions are not intended to independently diagnose, treat, prevent, or manage disease, and they should not be used as the sole basis for patient-level or clinical decisions.
Contact
Bring us a target.
A target to design against, a candidate to de-risk, or the pipeline data.
Laboratory bench · liquid handler and assay plates