Please login first
The Early detection of frailty syndrome using a model that employs a combination of omic data
* 1 , 2 , 2 , 2 , 2 , 3, 4, 5 , 3 , 3 , 1 , 1 , 1 , 1, 4, 6, 7 , 1, 4, 6, 7
1  FISABIO, Valencia, Spain
2  Sabartech, Valencia, Spain
3  Universidad Autónoma de Madrid, Madrid, Spain
4  CIBEResp, Madrid, Spain
5  IMDEA-Nutrition, Madrid, Spain
6  Institute for Integrative Systems Biology (I2SysBio), Valencia, Spain
7  Universidad de Valencia, Valencia, Spain
Academic Editor: Antonio Carvajal-Rodríguez

Published: 05 February 2026 by MDPI in The 1st International Online Conference on Biology session Evolutionary Biology
Abstract:

Frailty syndrome (FS) is an age-related condition characterised by a loss of physiological reserves across multiple organs and systems, resulting in high vulnerability to even mild stressors [1]. This state of physiological deterioration and generalized loss of homeostasis has been shown to increase the risk of premature mortality, falls, fractures, hospitalization and institutionalization among the elderly [2]. An early and accurate diagnosis of FS is therefore critical for improving patient quality of life and guiding clinical decision-making.

FS is a complex phenotype influenced by multiple factors, with approximately 40% of its development attributable to genetic determinants. Genome-wide association studies have identified significant variants in genes involved in inflammation, neurotransmission, and aging pathways [3]. Concurrently, more evidence has emerged indicating a correlation between gut microbiota dysbiosis and the progression of FS [4].

In this study, a cohort comprising 936 genomic samples (whole-genome DNA microarrays) and 199 microbiome profiles obtained thro­ugh 16S rRNA sequencing was analysed. The cohort included both frail and healthy individuals aged 65 years. Supplementary clinical data provided additional context on participant health status. Predictive models were generated for each type of data: genomic, microbiome and clinical. Subsequently, an ensemble learning approach was implemented for the purpose of integrating all three model predictions, with a view to enhancing predictive accuracy.

The findings suggest that the combined ensemble model demonstrates superior performance in comparison to single-source predictors. The conclusions of the present study demonstrate the potential of omic data fusion and advanced machine learning techniques for FS diagnosis.

References:

  1. Kim DH, Rockwood K. Frailty in Older Adults. N Engl J Med. 2024;391(6):538-548.

  2. Khan KT, Hemati K, Donovan AL. Geriatric Physiology and the Frailty Syndrome. Anesthesiology clinics. 2019;37:453-474.

  3. Weiss CO. Frailty and chronic diseases in older adults. Clinics in geriatric medicine. 2011;27:39-52.

  4. Tongeren SP, Slaets JPJ, Harmsen HJM, Welling GW. Fecal microbiota composition and frailty. Applied and environmental microbiology. 2005;71:6438-6442.

Keywords: Frailty Syndrome, omics, machine learning, mixed model

 
 
Top