A Three-Layered Framework Integrating Classical, Multivariate, and Machine-Learning Methods for Systemic Treatment Effect Detection in High-Dimensional Biomarker Data
Articles
Monika Ošmianskienė
Vilnius University image/svg+xml
Published 2026-05-08
https://doi.org/10.15388/LMITT.2026.21
PDF

Keywords

biomarker analysis
framework
systems biology
machine learning
PCA

Abstract

Longevity supplement trials often rely on single-biomarker tests, which can miss distributed systemic effects. This paper proposes a three-layer analytical framework: (i) classical biostatistics, (ii) multivariate systems biology, and (iii) machine learning with responder analysis. Applied to a 99-participant trial with 20 biomarkers, Layer 1 found few isolated effects, Layer 2 detected significant multivariate separation, and Layer 3 supported reliable directional effects for nicotinamide adenine dinucleotide (NAD+) and low-density lipoprotein cholesterol (LDL-C). Together, the layers provide complementary evidence beyond any single method.

PDF

References

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Downloads

Download data is not yet available.