Review of in vitro sarcopenia models and in silico benchmarking
This review compares in vitro models of selected sarcopenia-related phenotypes, reductionist tools for exercise-associated adaptations, and in silico methods for benchmarking, noting limits such as batch effects and overfitting. [1]
Automatic summary · not editorially reviewed
The key findings
The review compares in vitro models that reproduce selected sarcopenia-related pathological phenotypes. [1]
Electrical pulse stimulation, mechanical stretch and pathway-targeted pharmacological probes are assessed as reductionist tools rather than reproductions of whole-body exercise. [1]
In silico benchmarking must account for batch effects, dataset heterogeneity, cell-composition confounding, overfitting, interpretability and external validation. [1]
The source at a glance
- Study type / format
- Review (abstract)
- Population / context
- In vitro muscle models, exercise-adaptation tools and in silico methods discussed in a review
- Summary based on
- Study abstract
- Original publication
- September 26, 2026
Sarcopenia phenotypes in cell and tissue models
The review describes sarcopenia as a progressive, generalized skeletal muscle disorder marked mainly by reduced muscle strength; low muscle quantity or quality confirms the diagnosis, and poor physical performance indicates greater severity. It critically evaluates in vitro models that reproduce selected sarcopenia-related pathological phenotypes. Myogenic cell cultures, three-dimensional engineered muscle tissues, senescence-inducing systems and microphysiological platforms are compared by the phenotypes modeled, principal endpoints, strengths and limitations.
Supporting source [1]
Reductionist tools for exercise-associated adaptations
Electrical pulse stimulation, mechanical stretch and pathway-targeted pharmacological probes are assessed as reductionist tools. According to the review, they examine contraction-associated, load-sensitive, humoral and metabolic responses rather than reproducing whole-body exercise. Human studies and animal models remain essential, but their use is constrained by ethical considerations, inter-individual variability, species-specific differences, cost and the difficulty of isolating individual mechanisms within whole-body systems.
Supporting source [1]
In silico benchmarking and its stated pitfalls
Multi-omics profiling, machine learning, deep learning, public transcriptomic datasets and cell-type-aware analyses are discussed as ways to benchmark in vitro signatures against in vivo data. The review states that these approaches must account for batch effects, dataset heterogeneity, cell-composition confounding, overfitting, interpretability and external validation. It suggests that integrating experimentally validated findings with clearly identified hypothesis-generating observations may improve model selection, reduce reliance on animal experiments where appropriate and accelerate development of interventions to preserve skeletal muscle function during aging.
Supporting source [1]
How to interpret this
Abstract only; the full text was not reviewed. This is a review of models and methods, not a new experiment, and it reports no sample sizes, follow-up or outcome results. Statements about improving model selection or accelerating interventions are proposals, not demonstrated benefits.
Based on the study abstract; the full paper, tables and supplementary data were not reviewed.
Source & supporting passages
Baek KW, Jeong YR, Kim JS, Lee AY. · europepmc.org · September 26, 2026