# Bioweave — Longevity Protocol Engine
## Technical Brief for Wil | April 2026

---

## WHAT WE'RE BUILDING

A longevity-focused protocol engine that designs the most effective supplement stack for each individual based on their genetics, biomarkers, and health goals. This is the core product differentiator — IM8 sells the same 90+ compounds to everyone. We analyse which longevity compounds YOUR biology will actually benefit from.

## THE FRAMEWORK: Hallmarks of Aging

Modern aging research identifies 12 hallmarks. Each person's biomarkers and genetics reveal which hallmarks are most active in THEIR biology. The engine should:

1. **Profile** which hallmarks are priorities for this user (from their blood work + DNA)
2. **Select** compounds with evidence for those specific hallmarks
3. **Dose** based on their genetics (e.g., CYP metaboliser status affects optimal dose)
4. **Stack safely** using the pathway load model (see companion document)
5. **Evolve** each cycle as biomarkers change

## HALLMARK → BIOMARKER → COMPOUND MAPPING

### 1. Genomic Instability
- **Biomarkers**: DNA damage markers, 8-OHdG
- **Compounds**: NMN/NR (NAD+ for DNA repair), selenium, zinc
- **Genetics**: PARP1 variants affect DNA repair capacity

### 2. Telomere Attrition
- **Biomarkers**: telomere length tests (if available)
- **Compounds**: Vitamin D, omega-3, astragalus (astragaloside IV)
- **Genetics**: TERT variants

### 3. Epigenetic Alterations
- **Biomarkers**: homocysteine (methylation proxy), folate, B12
- **Compounds**: methylfolate (if genetics allow), TMG, B vitamins
- **Genetics**: MTHFR, COMT, MTR, MTRR — the methylation panel

### 4. Loss of Proteostasis
- **Biomarkers**: protein markers, autophagy indicators
- **Compounds**: spermidine (autophagy inducer), sulforaphane
- **Genetics**: ATG variants

### 5. Deregulated Nutrient Sensing
- **Biomarkers**: HbA1c, fasting glucose, insulin, IGF-1
- **Compounds**: NMN/NR (NAD+), berberine, alpha-lipoic acid
- **Genetics**: FTO, AMPK pathway variants

### 6. Mitochondrial Dysfunction
- **Biomarkers**: CoQ10 levels, lactate, mitochondrial markers
- **Compounds**: CoQ10 (ubiquinol), PQQ, urolithin A, D-ribose
- **Genetics**: mitochondrial DNA variants, SOD2

### 7. Cellular Senescence
- **Biomarkers**: inflammatory markers (CRP, IL-6), p16INK4a (emerging)
- **Compounds**: fisetin (senolytic), quercetin + dasatinib protocol, luteolin
- **Genetics**: CDKN2A/p16 variants

### 8. Stem Cell Exhaustion
- **Biomarkers**: blood cell counts, recovery markers
- **Compounds**: vitamin D, NAD+ precursors, nicotinamide
- **Genetics**: stem cell pathway variants

### 9. Altered Intercellular Communication
- **Biomarkers**: CRP, homocysteine, inflammatory panel
- **Compounds**: omega-3 (SPMs), curcumin, resveratrol
- **Genetics**: IL-6, TNF-alpha variants

### 10. Disabled Macroautophagy
- **Biomarkers**: autophagy markers (emerging)
- **Compounds**: spermidine, lithium (low-dose), EGCG
- **Genetics**: ATG pathway

### 11. Chronic Inflammation
- **Biomarkers**: hs-CRP, IL-6, ESR, ferritin
- **Compounds**: omega-3, curcumin, SPMs (specialised pro-resolving mediators)
- **Genetics**: inflammatory pathway variants

### 12. Dysbiosis
- **Biomarkers**: microbiome diversity (if tested), GI symptoms
- **Compounds**: precision psychobiotics (see Psychobiotics Proposal), prebiotics, digestive enzymes
- **Genetics**: FUT2 (secretor status), lactase persistence

## HOW THIS WORKS IN THE ENGINE

### Step 1: Hallmark Scoring
From the user's profile, calculate a 0-10 priority score for each hallmark:
```
hallmark_score = f(relevant_biomarkers, genetic_variants, age, goals)
```

Example for DT:
- Genomic instability: 4 (moderate — SOD2 variant)
- Epigenetic alterations: 8 (high — MTHFR + COMT + elevated homocysteine)
- Mitochondrial dysfunction: 6 (moderate — SOD2, declining eGFR)
- Chronic inflammation: 3 (low — CRP is good)
- Cellular senescence: 5 (moderate — age-related)

### Step 2: Compound Selection
For each hallmark scoring above threshold (e.g., >4), select the top evidence-based compounds. Cross-reference with genetics for safety and efficacy.

### Step 3: Pass Through Safety Layers
1. Gene-compound gate
2. Drug interaction check
3. Confidence gating
4. Evidence grounding
5. Pathway load model (cumulative stacking check)

### Step 4: Output Protocol
A ranked list of compounds with:
- Compound name + form + dose
- Which hallmarks it targets
- Confidence level
- Evidence sources
- Safety clearance status

## GOAL-BASED MODIFICATION

The same engine handles different user goals by weighting hallmarks differently:

| Goal | Priority Hallmarks |
|------|-------------------|
| Longevity/healthspan | All 12, weighted by biomarkers |
| Cardiovascular | Inflammation, epigenetic, nutrient sensing |
| Fertility | Epigenetic, mitochondrial, hormonal markers |
| Sleep | Intercellular communication, inflammation, GABA pathways |
| Anxiety | Inflammation, epigenetic, gut-brain axis |
| Perimenopause | Hormonal markers, bone density, inflammation |
| Athletic performance | Mitochondrial, stem cell, inflammation |

The user's stated goal adjusts the hallmark weights. The biology still drives the compound selection.

## PRIORITY FOR BUILD

1. **Hallmark scoring function** — takes profile data, outputs 12 scores
2. **Compound-hallmark mapping database** — which compounds target which hallmarks, with evidence grades
3. **Integration with existing AI Medical Board** — hallmark scores become an input to specialist agents
4. **Goal-weighting system** — user's stated goal adjusts hallmark priorities

## DATA SOURCES FOR COMPOUND EVIDENCE

- Examine.com systematic reviews
- PubMed meta-analyses (filter for human RCTs)
- DrugBank for interaction data
- Natural Medicines Database
- Bryan Johnson's published protocols (for compound identification, not dosing)
- Sinclair lab publications (NAD+, sirtuins)
- Mayo Clinic aging research