PTH Meridian — Bioinformatics Division

Computational health.

Disease does not happen without cause. Finding that cause — precisely, computationally, before more people suffer — is the mission of PTH Meridian's health division. We build tools that surface the biological signal hidden inside complex clinical, genomic and molecular data.

01

The health stack

Computational tools that ask the hard questions about disease causation and molecular analysis — and find answers in data that was always there.

// 01
HqA
HARD QUESTIONS ANSWERED — DISEASE CAUSATION ENGINE

Disease causation mapping at the computational level. HqA surfaces the biological, environmental and systemic causes of disease — connecting symptom patterns to root causes across genomic, clinical and epidemiological data. Shorter diagnostic odysseys. Clearer causal understanding.

Disease causation mapping Symptom-to-diagnosis matching Genomic variant analysis Rare disease focus In development
// 02
Coming Soon
MNTU
MOLECULAR NODE TRACER UNIFIED — CDE ANALYSIS PLATFORM

Molecular network analysis for complex disease environments. MNTU traces nodes and connections in biological molecular networks — protein interactions, metabolic pathways, gene regulatory networks — to identify where disease processes originate and propagate.

Molecular network tracing Protein interaction mapping Metabolic pathway analysis CDE platform Planned
02

The problems we address

People suffering longer than necessary because the biological signal is there and we lack the computational tools to read it.

// Rare Disease
Diagnostic Odyssey

Patients with rare diseases see an average of seven specialists and wait seven years before a correct diagnosis. The information exists. The computational bridge does not.

7,000 rare diseases — most without computational diagnostic tools
// Antimicrobial Resistance
AMR Surveillance

Drug-resistant infections projected to surpass cancer as a cause of death by 2050. Resistance genes spread faster than surveillance tracks them. Computational genomic surveillance could provide early warning.

Projected 10M deaths annually by 2050 without intervention
// Data Privacy
Health Data Silos

Medical AI advances slowly because patient data cannot be shared without privacy violations. Zero-knowledge proofs — built in the security stack — allow hospitals to contribute to global disease models without sharing a single patient record.

ZKP + federated learning — privacy-preserving health research
// Indigenous Health
Health Equity Gap

Indigenous communities in Canada experience dramatically worse health outcomes across almost every measure. Computational tools to map causal chains and prioritize interventions are largely absent.

Canadian reconciliation — health outcomes as measurable indicators
// Energy-Health Nexus
Energy Poverty

Four million people die annually from indoor air pollution caused by cooking fires. No electricity means no refrigeration for vaccines or insulin. Quantifying the health return on energy investment creates the evidence base for life-saving infrastructure.

4M deaths annually — preventable with clean energy access
// Pandemic Preparedness
Early Warning

COVID-19 was spreading globally for weeks before detection. Wastewater epidemiology, genomic surveillance and zoonotic spillover prediction could provide weeks of warning. The computational infrastructure for real-time pandemic early warning remains fragmented.

Weeks of warning vs days — the difference between containment and pandemic
The signal is in the data.

Disease does not occur without cause. Every cause leaves a biological signature somewhere in genomic sequences, clinical records, molecular interaction networks, environmental measurements and epidemiological patterns.

HqA and MNTU are built to close the gap between data that exists and understanding that does not yet. Open source, auditable and designed for the Canadian health research context first.

Privacy is foundational. Zero-knowledge proofs from the security stack mean patient data never leaves its institution — the proof travels, the data does not. PIPEDA compliant by design.

01
Causation Not Correlation
Knowing what happens together is not knowing why. Why is what saves lives. We build causal inference tools, not pattern-matching tools.
02
Privacy By Design
Health data is the most sensitive data that exists. ZKP makes privacy mathematically guaranteed — not just policy.
03
Open Source
Every algorithm auditable. Every model inspectable. In health research, black boxes are unacceptable.
04
Canadian Context First
Indigenous health outcomes, rare disease, pandemic preparedness — Canadian problems first. The tools then apply globally.
05
Equity As Outcome
A health tool that only benefits well-resourced institutions has failed. The design goal is tools that reduce health disparities, not amplify them.
Research collaboration welcome

HqA and MNTU are in active development. We welcome collaboration from Canadian health researchers, genomicists and bioinformaticians.