A new research paper was published in Aging (listed by MEDLINE/PubMed as "Aging (Albany NY)" and "Aging-US" by Web of Science) Volume 15, Issue 11, entitled, "Precious1GPT: multimodal transformer-based transfer learning for aging clock development and feature importance analysis for aging and age-related disease target discovery."
Aging is a complex and multifactorial process that increases the risk of various age-related diseases and there are many aging clocks that can accurately predict chronological age, mortality, and health status. These clocks are disconnected and are rarely fit for therapeutic target discovery.
In this study, researchers Anatoly Urban, Denis Sidorenko, Diana Zagirova, Ekaterina Kozlova, Aleksandr Kalashnikov, Stefan Pushkov, Vladimir Naumov, Viktoria Sarkisova, Geoffrey Ho Duen Leung, Hoi Wing Leung, Frank W. Pun, Ivan V. Ozerov, Alex Aliper, Feng Ren, and Alex Zhavoronkov from Insilico Medicine propose a novel approach to multimodal aging clock, which they call Precious1GPT, utilizing methylation and transcriptomic data for interpretable age prediction and target discovery developed using a transformer-based model and transfer learning for case-control classification.
"To identify aging biomarkers associated with age-related diseases, in the present work, we combined the ability of aging clocks to predict biological age and thus grasp molecular changes accompanied by senescence and our target ID approach to establish genes that are related to the development of diseases."
While the accuracy of the multimodal transformer is lower within each individual data type, compared to the state of art specialized aging clocks based on methylation or transcriptomic data separately, it may have higher practical utility for target discovery. This method provides the ability to discover novel therapeutic targets that hypothetically may be able to reverse or accelerate biological age providing a pathway for therapeutic drug discovery and validation using the aging clock. In addition, the researchers provided a list of promising targets annotated using the PandaOmics industrial target discovery platform.
"The transformer-based model allowed for the integration of multi-omics data and improved the accuracy of the aging clock, while the transfer learning approach facilitated the identification of disease-related genes in the context of aging."
Source:
Journal reference:
Urban, A., et al. (2023) Precious1GPT: multimodal transformer-based transfer learning for aging clock development and feature importance analysis for aging and age-related disease target discovery. Aging-US. doi.org/10.18632/aging.204788.