41
1,500+
8,436
R² 0.849
🥗 Nutrition
😰 Stress
😴 Sleep
🌿 Environmental exposures
🏃 Exercise
⏳ Ageing
💡 Lifestyle habits
💊 Medication
🧠 Psyche
💰 Social & economic status
R² 0.849
R² 0.775
Training set
1,521 women
Male test cohort
2,097 men
Female error (MAE)
~3.54 years
Male error (MAE)
~4.46 years
CpG sites
41 selected
Array
Illumina EPIC 850k
Predicted vs Actual Age (41 CpG Clock), Female (orange x) vs Male (blue x) vs Ideal Fit (red dashed). Female data points cluster tighter around the ideal line.
Team Ummu are specialists in saliva DNA methylation.
5 Biological Age Tests
The Epi-Vitality Test
Post-Menopause Cohort — High Scorers Showed (all p<0.001):
Lower subjective hot flushes
p<0.001
Thicker hair (subjective)
p<0.001
Superior memory (tested & subjective)
p<0.001
Stronger bones (BMD)
p<0.001
Superior cholesterol levels
p<0.001
Superior muscle power / handgrip
p<0.001
Injury Prediction Before the Injury
the data
Why Single-Gene Tests Are Misleading — and What Ummu Does Instead
1. Most traits are polygenic
2. Aggregates small effects
3. Single SNPs are misleading
4. Better prediction accuracy
5. Enables stratification
6. Captures hidden heritability
SIMPLE ANALOGY · PREDICTING EXAM RESULTS BY...
Single SNP
One homework score
Single gene
One subject
Polygenic score
Every assignment, test and project combined, the full picture.
APPROACH
VARIANCE EXPLAINED
Single SNP
<0.1%
Single gene
1–5% (rare cases)
Polygenic score
10–40%+ depending on trait
TRAITS UMMU SCORES POLYGENICALLY:
| Clock / Method | Type | Female-Trained | Epigenetic Biological Ageing | Predicts Age from DNA | Hormone / Ageing Loci | Pace of Ageing |
|---|---|---|---|---|---|---|
| UmmuAge (Collins) | Saliva DNAm | ✓ Yes | ✓ Yes | ✓ Yes | ✓ Yes | ✓ Yes |
| 10-CpG Saliva Clock (Collins et al.) | Saliva DNAm | ✗ No | ✓ Yes | ✓ Yes | Limited | ✗ No |
| Horvath Pan-Tissue Clock | Multi-tissue DNAm | ✗ No | ✓ Yes | ✓ Yes | Mixed | ✗ No |
| Pace-of-Ageing Clocks | Blood DNAm | ✗ No | Rate only | ✗ No | Mixed | ✓ Yes |
| Wearable "Biological Age" | Device metrics | ✗ No | ✗ No | ✗ No | ✗ No | ✗ No |
| MuhdoAge | Saliva DNAm | ✗ No | ✓ Yes | ✓ Yes | ✓ Yes | ✗ No |
| GlycanAge | Blood glycomics | ✗ No | Immune ageing | Weak | Limited | ✗ No |
2025
A Cost-Effective Saliva-Based Human Epigenetic Clock Using 10 CpG Sites Identified with the Illumina EPIC 850k Array
View Research ↗
2025
Creatine Monohydrate Use Is Associated with Performance Enhancing DNA Methylation Patterns
View Research ↗
2025
High Intensity Resistance Training Is Associated with Epigenetic Reprogramming & Distinct Salivary DNA Methylation Patterns
View Research ↗
2025
Genetic Predisposition vs. Performance Enhancing Drug Use and Fat Free Mass Index in Strength Trained Men
View Research ↗
2025
DNA Methylation Signatures Diverge Between Endurance and Resistance Training Modalities
View Research ↗
2026
High Intensity Exercise and Sport Type Are Associated with Lower Epigenetic Biological Age in Middle Aged Men
View Research ↗
2024
A Randomised, Double Blind, Placebo Controlled, Cross Over Clinical Trial to Evaluate the Biological Effects and Safety of a Polyphenol Supplement on Healthy Ageing
View Research ↗
2023
Responsiveness to Endurance Training Can Be Partly Explained by the Number of Favourable Single Nucleotide Polymorphisms an Individual Possesses
View Research ↗
2022
Born Equal: Can Genetics Make the Perfect Athlete
View Research ↗
2018
Can Genetics Predict Sports Injury? The Association of the Genes GDF5, AMPD1, COL5A1 and IGF2 on Soccer Player Injury Occurrence
View Research ↗
2018
The Impact of a Reduced Calorie, Macronutrient Diet Change on Caucasian Males in Relation to Genotypes Associated with Obesity, Increased BMI and Dietary Response
View Research ↗