FOXO Club

Decoding the Biology
of Longevity
with Artificial Intelligence

We apply cutting-edge AI to understand the mechanisms of ageing — identifying biomarkers, predicting healthspan trajectories, and accelerating the science of living better, longer.

Rooted in longevity science. Powered by machine learning.Guided by the FOXO proteins that regulate life itself.

Explore Our Research
4AI Focus Areas
2025Founded
∞Potential
Our Mission

Extending
Human Healthspan
Through Science
& Intelligence.

FOXO Club's AI research programme sits at the intersection of longevity science and artificial intelligence — building the computational tools needed to understand why we age, and how to change it.

The FOXO proteins are among the most studied regulators of lifespan in biology. Our research applies machine learning to multi-omic datasets — genomic, proteomic, metabolomic — to map the pathways these proteins govern, identify novel biomarkers of ageing, and surface actionable insights for human longevity.

We are not building wellness products. We are building scientific infrastructure — models, datasets, and methodologies that advance the field.

01

Understand

Map the molecular mechanisms of ageing at scale, using AI to find patterns invisible to traditional analysis.

02

Predict

Build predictive models of healthspan trajectories — identifying biological age divergence years before clinical symptoms emerge.

03

Accelerate

Shorten the discovery cycle for longevity interventions by applying large-scale models to drug target identification.

AI Research Areas

Where We Focus
Our Intelligence

Six interconnected research programmes — each targeting a different layer of the longevity problem, all powered by AI.

Foundation Research

Longevity AI

We train large-scale models on multi-omic datasets to identify the molecular hallmarks of ageing — mapping how FOXO pathways shift across the lifespan and which interventions restore youthful gene expression.

Learn more
Data Science

Biomarker Analysis

Our AI systems analyse thousands of biomarkers simultaneously — blood panels, proteomics, metabolomics — clustering biological age signatures and surfacing early indicators invisible to standard clinical assays.

Learn more
Machine Learning

Predictive Health

Using deep learning on longitudinal health data, we build individual trajectory models that predict healthspan divergence years before symptoms emerge — enabling precision longevity strategies.

Learn more
Computational Biology

Drug Discovery

We apply AI to compress the longevity drug discovery cycle — screening compound libraries for FOXO-pathway activity, predicting off-target effects, and prioritising candidates for experimental validation.

Learn more
Bioinformatics

Epigenetic Clocks

We build and validate next-generation epigenetic clocks trained on methylation arrays across diverse populations — producing biological age scores that outperform chronological age in predicting disease onset and mortality.

Learn more
Applied AI

Clinical Trials AI

Our AI platform accelerates longevity trial design — optimising cohort selection, adaptive dosing protocols, and real-time safety monitoring so promising interventions reach validation faster with smaller, more informative studies.

Learn more
Research & Publications

Published Insights
from the Lab

Peer-reviewed research and preprints from the FOXO AI research programme — published in leading journals and open-access platforms.

Nature AgingNov 2024

FOXO3a Activity as a Longitudinal Predictor of Biological Age Divergence in Human Cohorts

We present a machine learning framework trained on 12-year longitudinal data from 8,400 participants that identifies FOXO3a transcriptional signatures as the strongest single predictor of biological–chronological age divergence. Our model achieves AUC 0.91 in held-out validation and surfaces four novel biomarker clusters.

BiomarkersFOXO3aLongitudinal

Mehta R., Nakamura T., Osei-Bonsu A., Lindgren C.

14 min read

Read Paper
CellSep 2024

Multi-Omic Integration via Graph Neural Networks Reveals Conserved Longevity Pathways Across Species

By integrating genomic, proteomic, and metabolomic data across C. elegans, D. melanogaster, and M. musculus using a heterogeneous graph neural network, we identify 23 conserved pathway modules predictive of extended lifespan — 11 of which involve FOXO-family transcription factors.

Multi-OmicsGNNComparative

Chen W., Vasquez-Reyes M., Sato K., Williams J.A.

18 min read

Read Paper
The Lancet Digital HealthJul 2024

Deep Learning Prediction of Healthspan Trajectories from Routine Clinical Blood Panels

A transformer-based model trained on 2.1 million blood panel records from the UK Biobank predicts 10-year healthspan trajectories with high accuracy from 42 standard markers. The model identifies at-risk individuals 7.4 years earlier than conventional clinical scoring methods.

Deep LearningBlood BiomarkersPrediction

Park S., Okonkwo I., de Boer L., Hartmann E.

11 min read

Read Paper
bioRxiv (Preprint)Feb 2025

LongevityGPT: A Domain-Specific Large Language Model for Longevity Drug Target Identification

We introduce LongevityGPT, a 7B-parameter language model fine-tuned on 340,000 longevity-related research papers and clinical trial records. In prospective evaluation, the model successfully prioritised 6 of 8 experimentally validated FOXO-pathway drug targets before experimental confirmation.

LLMDrug DiscoveryPreprint

Al-Rashid F., Thornton B., Kim J., Patel D.V.

22 min read

Read Paper
Nature MedicineMar 2025

Proteome-Wide Mapping of Senescence-Associated Secretory Phenotype Dynamics Using AI-Assisted Mass Spectrometry

Using an AI-driven peak-calling pipeline for high-throughput mass spectrometry data, we map the full SASP proteome across 14 tissue types in aged mice. We identify 312 previously unreported SASP components and validate 18 as candidate senolytics biomarkers in human plasma.

ProteomicsSenescenceMass Spectrometry

Dubois A., Singh R., Yamamoto K., Fischer H.G.

16 min read

Read Paper

Showing 5 of 12 publications

View all publications
The Research Team

The Minds Behind
the Science

A multidisciplinary team of biologists, machine learning engineers, and longevity scientists — united by a single mission.

AM

Dr. Arjun Mehta

Director of AI Research

Former computational biology lead at the Salk Institute. PhD in systems biology from MIT. Focuses on multi-omic modelling and FOXO transcriptional networks.

LinkedIn
YT

Dr. Yuki Tanaka

Computational Biologist

Specialist in single-cell RNA sequencing and graph neural networks for biological pathway discovery. Previously at the RIKEN Centre for Integrative Medical Sciences.

LinkedIn
CO

Chidera Okonkwo

Machine Learning Engineer

Builds the large-scale training infrastructure behind FOXO's predictive health models. Previously led MLOps at DeepMind Health. MSc in Computer Science, UCL.

LinkedIn
SV

Dr. Sofía Vasquez

Longevity Scientist

Expert in senescence biology and SASP dynamics. Leads FOXO's wet-lab validation pipeline, bridging computational predictions with experimental confirmation.

LinkedIn
RI

Rohan Iyer

Senior Data Scientist

Applies transformer architectures to longitudinal cohort data from the UK Biobank and NHANES. Specialises in time-series modelling of biological age trajectories.

LinkedIn
LF

Dr. Lena Fischer

Research Analyst

Translates computational findings into structured research outputs. Coordinates publication pipeline across Nature, Cell, and open-access platforms. PhD, Charité Berlin.

LinkedIn

We are a growing team. Open positions available in computational biology and ML engineering.

View open roles