A research project at the Technical University of Munich

From whether
to why.

Traceable Rejuvenation Analytics & Implications for Society We develop causal, interpretable AI to help explain what drives biological aging and rejuvenation, and we study what it takes for people to trust it.

Our approach Get in touch
Duration
3 years
Start
2026
Campuses
Heilbronn & Garching
Program
HDSC-MDSI Tandem

The challenge

Longevity research can measure change. It cannot yet explain it.

Over the past decade, two approaches have moved to the center of longevity research. Both have shown promising results, yet they share the same limitation: they establish whether an intervention changes biological age, but not why.

Missing mechanistic insight constrains trust and slows clinical translation.

Epigenetic aging clocks

Use molecular biomarkers to estimate biological age and to gauge how an intervention shifts it.

Partial cellular reprogramming

Can restore youthful molecular profiles in animal models, a promising route to reversing aspects of biological aging.

  • CorrelativeEstimates rest largely on DNA methylation and do not separate drivers of aging from mere markers.
  • OpaqueMechanisms across molecular, cellular, and tissue levels remain poorly understood.
  • UnresolvedLimited transparency and open ethical questions around sensitive genetic data constrain trust.

Our approach

Two coupled strands, one goal.

Rigor on one strand is meant to earn justified trust on the other.

Technical & data strandHDSC, Heilbronn

What does the number mean, and what actually moves it?

Separate
causal drivers of aging from mere markers
Integrate
multi-omics data across biological scales
Explain
results faithfully, for every audience
Behavioral & societal strandMDSI, Garching

What does acting on it do to people and to society?

Understand
what people need from an AI-derived age
Design
explanations that earn warranted trust
Anticipate
societal implications and shape governance

Built to operate at the scale of the field

An agentic AI research assistant

An AI system that reads the research literature into a living knowledge graph and helps plan experiments, while the researchers make the final scientific call.

  1. Mine the literature
  2. Grow the knowledge graph
  3. Support experiments
  4. Researchers decide

Team

A tandem across two TUM campuses.

Expertise in data science, machine learning, bioinformatics, and behavioral trust research.

  • Prof. Dr. Ali Sunyaev

    Principal Investigator, technical & data strand
    Chair of Information Infrastructures, TUM Campus Heilbronn

  • Prof. Jens Grossklags, Ph.D.

    Principal Investigator, behavioral & societal strand
    Chair of Cyber Trust, TUM Campus Garching

  • Felix Pietsch

    Doctoral researcher
    TUM Campus Heilbronn

  • Philipp A. Toussaint

    Postdoctoral researcher
    Technical University of Munich

Heilbronn

TUM Heilbronn Data Science Center

HDSC

Garching

Munich Data Science Institute

MDSI

Funded through the HDSC-MDSI Tandem Projects program of the TUM Heilbronn Data Science Center and the Munich Data Science Institute.

Background

Selected prior work by the team.

Published research that TRAIS builds on: explainable AI for omics data, trustworthy and high-risk AI, and interdependent privacy, including genetic data sharing.

  • 2025
    High-risk artificial intelligenceSunyaev A, Benlian A, Pfeiffer J, Jussupow E, Thiebes S, Maedche A, Gawlitza J. Business & Information Systems Engineering 67(6), 981-994.
    DOI ↗
  • 2024
    Explainable artificial intelligence for omics data: a systematic mapping studyToussaint PA, Leiser F, Thiebes S, Schlesner M, Brors B, Sunyaev A. Briefings in Bioinformatics 25(1), bbad453.
    DOI ↗
  • 2021
    Trustworthy artificial intelligenceThiebes S, Lins S, Sunyaev A. Electronic Markets 31(2), 447-464.
    DOI ↗
  • 2019
    On sharing intentions, and personal and interdependent privacy considerations for genetic data: a vignette studyWeidman J, Aurite W, Grossklags J. IEEE/ACM Transactions on Computational Biology and Bioinformatics 16(4), 1349-1361.
    DOI ↗
  • 2017
    Valuating friends’ privacy: does anonymity of sharing personal data matter?Pu Y, Grossklags J. Proceedings of the Thirteenth Symposium on Usable Privacy and Security (SOUPS 2017), 339-355.
    USENIX ↗

Contact

Interested in collaborating?

We welcome conversations with researchers, clinicians, and partners working on aging, interpretable AI, and trust.

Contact persons

Prof. Dr. Ali Sunyaev
Felix Pietsch

felix.pietsch@tum.de
Technical University of Munich
Chair of Information Infrastructures
Campus Heilbronn, Bildungscampus 2
74076 Heilbronn, Germany
Project page at TUM →