image: Figure 2. Comparison of biological age predictions between healthy controls and cases of severe COVID-19 infection. (A) Proteomic aging clock shows R2=0.84 in the task of age prediction in CV within the UK Biobank dataset (N = 55,319). (B) Proteomic aging clock’s error depends on the age group and is skewed toward the mean of the total sample. (C) Biological age acceleration (difference between predicted and chronological age) across severity groups. compared to healthy controls. Error bars represent standard error of the mean. (D) Linear regression analysis reveals that patients with severe cases, which are likely to develop lung fibrosis, showed significantly higher biological age predictions (+2.77 years, p=0.026).
Credit: Copyright: © 2025 Galkin et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
“Our findings establish novel connections between aging biology and IPF pathogenesis while demonstrating the potential of AI-guided approaches in therapeutic development for age-related diseases.”
BUFFALO, NY — September 11, 2025 — A new research paper was published in Volume 17, Issue 8 of Aging-US on August 8, 2025, titled “AI-driven toolset for IPF and aging research associates lung fibrosis with accelerated aging.”
In this study, researchers Fedor Galkin, Shan Chen, Alex Aliper, Alex Zhavoronkov, and Feng Ren from Insilico Medicine used artificial intelligence (AI) to investigate the similarities between idiopathic pulmonary fibrosis (IPF), a severe lung disease, and the aging process. Their findings show that IPF is not simply accelerated aging, but a distinct biological condition shaped by age-related dysfunction. This insight may lead to a new approach in how scientists and clinicians treat this complex disease.
IPF mainly affects individuals over the age of 60. It causes scarring of lung tissue, making it harder to breathe and often leading to respiratory failure. Current treatments can slow the disease but rarely stop or reverse its progression. The researchers used AI to identify shared biological features between aging and fibrosis, finding new potential targets for therapy.
The team developed a “proteomic aging clock” based on protein data from more than 55,000 participants in the UK Biobank. This AI-driven tool accurately measured biological age and found that patients with severe COVID-19, who are at increased risk for lung fibrosis, also showed signs of accelerated aging. This suggests that fibrosis leaves a detectable biological trace, supporting the use of aging clocks in studying age-related diseases.
“For aging clock training, we used the UK Biobank collection of 55319 proteomic Olink NPX profiles annotated with age and gender.”
They also developed a custom AI model, ipf-P3GPT, to compare gene activity in aging lungs versus those with IPF. Although some genes were active in both, many showed opposite behavior. In fact, more than half of the shared genes had inverse effects. This means IPF does not just speed up aging but also disrupts the body’s normal aging pathways.
The study identified unique molecular signatures that distinguish IPF from normal aging. While both involve inflammation and tissue remodeling, IPF drives more damaging changes to lung structure and repair systems. This difference could guide the development of drugs that specifically target fibrosis without affecting normal aging.
By combining AI with large-scale biological data, the study also introduces a powerful toolset for examining other age-related conditions such as liver and kidney fibrosis. These models may support personalized treatments and expand understanding of the relationships between aging and disease, opening new directions for therapy development.
DOI: https://doi.org/10.18632/aging.206295
Corresponding author:Alex Zhavoronkov – alex@insilico.com
Keywords: aging, IPF, generative AI, transformer, proteomics
Click here to sign up for free Altmetric alerts about this article.
______
To learn more about the journal, please visit our website at www.Aging-US.com and connect with us on social media at:
- X
- YouTube
- Bluesky
- Spotify, and available wherever you listen to podcasts
Click here to subscribe to Aging publication updates.
For media inquiries, please contact media@impactjournals.com.
Journal
Aging-US
Method of Research
News article
Subject of Research
Not applicable
Article Title
AI-driven toolset for IPF and aging research associates lung fibrosis with accelerated aging
Article Publication Date
8-Aug-2025
COI Statement
All authors are affiliated with Insilico Medicine, a commercial company developing and using generative artificial intelligence and other next-generation AI technologies and robotics for drug discovery, drug development, and aging research. Insilico Medicine has developed a portfolio of multiple therapeutic programs targeting fibrotic diseases, cancer, immunological diseases, and age-related diseases, utilizing its generative AI platform and a range of deep aging clocks and AI life models. Insilico Medicine is a company developing an AI-based end-to-end integrated pipeline for drug discovery and development and is engaged in aging and IPF research. Insilico Medicine holds USPTO patents for transcriptomic and proteomic aging clocks (US10665326B2, US10325673B2).