Bookmarks
Deep characterisation of common and rare diseases
Safe People
Organisation name
Pheiron GmbH
Organisation sector
Commercial
Applicant name(s)
Jakob Steinfeldt
Funders/ Sponsors
Safe Projects
Project ID
OFHS240110
Lay summary
We want to make new medicines faster by understanding diseases better. Right now, our disease definitions are too basic and miss early signs and details. We aim to improve these definitions by looking at lots of health data, like lifestyle and medical histories. By doing this, we can spot early signs of diseases in healthy people and see how diseases progress in those who are already sick. We will also study genes to find out which ones affect how diseases develop. Our main goals are: 1. Better understand diseases by looking at data patterns. 2. Find new ways to make drugs for challenging diseases. 3. Figure out who would benefit most from these new drugs. This work could help us diagnose diseases more accurately, prevent them better, and treat them more effectively. To find out what disease someone has, we often ask questions or check their health records. These records use medical codes to describe illnesses, but they don't catch all the details. This is a problem because many diseases have similar symptoms but different causes. Without these details, it's hard to create new treatments. Our research uses deep learning, a type of AI, to better define diseases. We will study lots of health data, including questionnaires and medical records, to find patterns that show more about diseases. We will then link these patterns to genetic markers to see how these genes affect health. By improving how we define diseases, our research will help fill gaps in our knowledge. It will improve our understanding of diseases and help us find genetic markers that point to the best drug targets. This will help us choose the right people for clinical trials, leading to better and more personalized treatments.
Public benefit statement
This research project aims to generate significant value in the way we understand complex diseases and develop new medicines. We will enable a more comprehensive understanding of disease etiology and progression by utilizing advanced deep-learning models to transform complex biomedical data into disease signatures. This will support our effort in the identification of therapeutic targets and the development of effective treatments as well as the identification of the individuals most likely to benefit. Beyond its general value in identifying new therapies, this work will also serve the public interest by driving progress toward personalized interventions and precision medicine
Request category type
Public Health Research
Other approval committees
Project start date
20/06/2025
Latest approval date
04/11/2024
Safe Data
Dataset(s) name
Safe Setting
Access type
TRE