We specialized in: Radiopharmaceuticals
AI for pharmaceutical discovery, development, and de-risking.
Our AI platform
One AI stack. Three ways to work with us.
BuildDevelopers & Researchers
For chemists, protein designers, molecular biologists, and AI drug-discovery teams.
EPiQ™
Advanced platform
Models, APIs, workflows, and infrastructure for advanced users and technical teams. Generate in the platform. Start from a biological target, create lead candidates, and optimize them with Epitorix's state-of-the-art models in one modular workspace.
Xenoclea
Guided workspace
Prompt-based, interactive discovery for users who want power without managing infrastructure. Say what you need to bind. Plain language is the whole interface. Nothing to configure, nothing to install.
Xenoclea is built into EPiQ™.
De-riskInvestigators
For pharma teams seeking earlier visibility into development, legal, and regulatory risk.
PharmPhax™
Biological & clinical risk
Bring your molecule. Epitorix combines AI analysis with expert human review to optimize the peptide and expose off-target, cell-type, organ, and biodistribution risks before costly development.
PharmacoLaw
Legal, patent & regulatory intelligence
PharmacoLaw includes PharmPhax plus patent, regulatory, and litigation intelligence for current or developing drugs.
PharmacoLaw includes PharmPhax™.
InvestInvestors
For investors and strategic partners evaluating programs and platform value.
Therapeutic Pipeline
Drug programs
A portfolio of internal therapeutic programs advancing from computational discovery toward development.
Epitorix AI models
Platform & infrastructure
The underlying models, datasets, and infrastructure powering Xenoclea, EPiQ and internal programs.
One intelligence layer. Multiple ways to engage with Epitorix. Build therapies. Back innovation. De-risk development.
Who we are
Epitorix is an intelligence platform and product company for pharma. We build AI models, applications, and internal drug programs that help teams discover therapies, evaluate risk, and make better decisions.
Driven by biology. Powered by AI.
Ligand ranking
Epitorix developed a deep-learning model that ranks any ligand input. Give it 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.
Measured in internal benchmarks. A preprint describing the evaluation is forthcoming.
Your ligands
Ranked
…
Every class
It works across small molecules, peptides, non-canonical peptides, macrocycles, antibodies, nanobodies, mini-proteins and DNA/RNA ligands.
Not just ranking
Epitorix can generate and rank. You are equally free to bring your own ligands and have them ranked.
Kd evaluation at scale
Epitorix predicts Kd at scale: Pearson R = 0.70 for small molecules and 0.84 for peptides, screening thousands in minutes. High speed ranking at ~1,000 molecules/s.
Science behind the scene
Traditional development can take years and millions of dollars before major safety risks become visible. Epitorix predicts biodistribution, organ risk, and cell-type risk earlier, enabling faster and lower-cost candidate optimization before clinical trials.
Traditional Approach
3-5 Years
to reach preclinical
$5-$10M
Up to 100x
lower cost
Up to 65x
faster
Epitorix
4-6 Weeks
to reach preclinical
Up to $100K
Generation
Traditional mRNA/phage display creates noisy, non-site-specific libraries and can take months. Epitorix generates ~10,000 high-quality site- and target-specific candidates with far less biological noise.
Traditional
mRNA/Phage Display
AI Structure-Guided
Generation
Higher accuracy and target specificity
Off-target
Traditional microarrays are limited by protein coverage, biological noise, cost, and long timelines. Epitorix predicts protein off-targets for targets, small molecules, and peptides in about 1 hour with >90% accuracy.
Highly Accurate Off-Target Prediction
In retrospective analyses, our model achieved over 90% accuracy in detecting known off-target interactions.
F1 · Binder
R · Peptide
R · Small-Molecule
Traditional methods: expensive and limited number of panels
Ranking
SPR and BLI require synthesis and wet-lab testing and can take months. Molecular dynamic and FEP+ need multiple hours per prediction. Epitorix predicts Kd at scale: Pearson R = 0.70 for small molecules and 0.84 for peptides, screening thousands in minutes.
Peptides
R = 0.84
Small molecules
R = 0.70
Comparable accuracy to costly computational methods.
Biodistribution and organ/cell at risk
Traditional programs often discover toxicity only after clinical testing begins. Epitorix predicts PK, PD, biodistribution, cell-type risk, and organ risk to optimize radiotherapeutics before side effects emerge.
More about EPiQ™
PharmPhax™
A Reliable System to De-Risk Radiopharmaceutical Industrial Investment
We put certainty and intelligence at your fingertips
0%
Accuracy
Services & Products
Drug discovery begins with a target. Describe yours in plain language, and the designs, their structures and their data come back together.
Advanced platform
Models, APIs, workflows, and infrastructure for advanced users and technical teams. Generate in the platform. Start from a biological target, create lead candidates, and optimize them with Epitorix's state-of-the-art models in one modular workspace.
Guided workspace
Prompt-based, interactive discovery for users who want power without managing infrastructure. Say what you need to bind. Plain language is the whole interface. Nothing to configure, nothing to install.
Biological & clinical risk
Bring your molecule. Epitorix combines AI analysis with expert human review to optimize the peptide and expose off-target, cell-type, organ, and biodistribution risks before costly development.
Legal, patent & regulatory intelligence
PharmacoLaw includes PharmPhax plus patent, regulatory, and litigation intelligence for current or developing drugs.
Drug programs
A portfolio of internal therapeutic programs advancing from computational discovery toward development.
Platform & infrastructure
The underlying models, datasets, and infrastructure powering Xenoclea, EPiQ and internal programs.
Pipeline & Show cases
We are generating the next pharmaceutical using our pipeline
Design the Lead
Completed · Q1 2026
Optimize the Lead
Completed · Q1 2026
In Vitro Binding Assay
Completed · Q1 2026
In Vivo Validation
In Progress · Q2 2026
Translational & Regulatory Step
Upcoming · Q3 2026
Clinical Trial
Upcoming · Q4 2026
Target
In Vivo ValidationHER2
Experimental Kd
2.1 nM
In Progress · Q2 2026
Target
In Vivo ValidationTROP2
Experimental Kd
8.5 nM
In Progress · Q2 2026
Target
In Vivo ValidationCD38
Experimental Kd
19.5 nM
In Progress · Q2 2026
Target
In Vivo ValidationPD-L1
Experimental Kd
45 nM
In Progress · Q2 2026
/FAQ
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.
Contact
Move from disease biology to a collaborative end-to-end program—from target discovery through lead generation—combining Epitorix intelligence with the scientists guiding the development path.
