Simulate the whole chain, not one link.
Give the platform a patient and a regimen. It runs the full causal arc — genome, exposure, engagement, pathway, disease, biomarkers, and safety — in sequence, each module feeding the next, every value carrying its source. Here's what happens at each step.
If a value can't be found and trusted, the model reports the gap — it doesn't invent one to fill it.
What you bring, what comes back
A simulation is a question about one patient and one regimen. You describe both; the platform returns a connected, sourced forecast across the whole arc — not a single curve from a single tool.
A patient and a regimen
The patient: genome, physiology, existing conditions, current medications. The regimen: the drugs, their doses, the schedule. As much or as little as you have — the model is explicit about what each gap costs.
A sourced forecast, end to end
Exposure and target-engagement profiles, pathway and disease response, biomarker trajectories, and a benefit-risk read — each value traceable to where it came from, each forecast carrying its confidence.
Seven modules, one connected pass
Each station takes the one before it as input. Change anything upstream and everything downstream re-runs — the way the body actually works. Every module names the sources it drew from.
-
1
Pharmacogenomics
Receives the patient's genome.
Star-allele calling and metabolizer phenotypes across CYP2D6, CYP2C19, CYP2C9, and other pharmacogenes — as continuous activity scores, not binary bins.
Sources (selection)PharmVar · CPIC · DPWG · PharmGKB
-
2
Pharmacokinetics
Receives the metabolizer phenotype, plus the regimen.
Concentration-time profiles for every drug, adjusted for renal and hepatic function, age, weight, albumin, and enzyme activity — with formulation-specific absorption, drug-drug interactions, and population variability.
Sources (selection)PK-DB · DrugBank · FDA labels · EMA SmPC
-
3
Pharmacodynamics
Receives the exposure reaching the target tissue.
Target engagement and receptor occupancy over time — Emax modeling, hysteresis, therapeutic-window assessment, and combination pharmacology, including synergy and antagonism.
Sources (selection)ChEMBL · BindingDB · Guide to Pharmacology
-
4
Pathway modulation
Receives target engagement.
Engagement propagated through biological networks, weighted by disease stage — the same drug moves an early-stage patient differently than a late-stage one.
Sources (selection)Reactome · KEGG · WikiPathways · STRING
-
5
Disease progression
Receives the pathway response.
Stage-aware progression against calibrated natural-history baselines, so the effect reads relative to what would have happened anyway — not against an untreated abstraction.
Sources (selection)ClinicalTrials.gov · Open Targets · DisGeNET
-
6
Biomarkers
Receives the modeled disease trajectory.
The readouts a clinician actually tracks — HbA1c, LDL-C, eGFR, NT-proBNP, MMSE, CDR-SB, and more — with their temporal dynamics.
Sources (selection)GTEx · Human Protein Atlas · GEO
-
7
Toxicity & benefit-risk
Receives efficacy and safety signals from the full chain.
Organ-specific risk across hepatic, cardiac, renal, neurological, and hematological dimensions — off-target pharmacology, hERG liability, ADR likelihood — folded into one benefit-risk view with the patient's own modifiers carried through.
Sources (selection)SIDER · LiverTox · ToxCast / Tox21 · FAERS
Honest by construction
When a module can't find a value it can stand behind, it reports the gap rather than fabricating one. A gap you can see is worth more than a number you can't trust — and over thousands of parameters, that discipline is what keeps the whole result sound.
One change upstream, felt all the way down
OVIVO has no average patient. Take this one, on a drug cleared through CYP2D6, and follow the chain:
Poor metabolizer → higher exposure than the label assumes → target engagement pushed past the therapeutic window → a benefit-risk read that flips from favorable to cautionary.
A population-average tool reports the label dose and moves on. OVIVO forecasts the difference, mechanism by mechanism — and shows the step where it happened.
| Factor | This patient |
|---|---|
| CYP2D6 | Poor metabolizer |
| Renal function | Stage 3 CKD |
| Concomitant | Statin (CYP3A4) |
| Disease stage | Early |
A foundation that keeps growing
The sources named at each station are a selection. Underneath sits a large, curated set of authoritative public databases — and it grows every week as we integrate new ones, each held to the same provenance discipline as everything else.
In a partnership, your sources join it too: proprietary datasets, internal assays, and project-specific data, integrated into a dedicated environment and traced exactly like the rest.
A selection · the full set grows weekly · partner sources integrated per project
Where partners point it
The same chain, aimed at the decision in front of you — with the reasoning shown, so the answer is one you can defend to a board or a regulator.
De-risk a pipeline
See where a program is most likely to fail — wrong dose, wrong cohort, a thin therapeutic window — before the spend goes in.
DDI & combinations
Quantify drug-drug interactions and combination pharmacology, synergy and antagonism, for a specific regimen.
Dose & formulation
Compare doses, schedules, and formulations on exposure and target engagement, patient by patient.
Continue or kill
Get a mechanistic read on whether to advance, repurpose, or stop — with the chain of reasoning behind it.