EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Key Piece of the Whole-Cell Puzzle: Mechanistic signaling for interpretable whole-cell models

David Liaskos, Vassily Hatzimanikatis

Current Opinion in Systems Biology · 2026-11 · 100613

DOI: 10.1016/j.coisb.2026.100613 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Current Opinion in Systems Biology, 2026-11. Verfasst von David Liaskos und 1 weiteren Autor:innen. Die Arbeit nennt 74 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Swiss National Science Foundation, Swiss Federal Institute of Technology Lausanne.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cellular signaling orchestrates how cells sense and respond to their environment by coupling extracellular cues to gene regulation and metabolism. As efforts toward human whole-cell models accelerate, representing signaling at mechanistic resolution is increasingly important for integrating cellular processes into coherent models that explain and predict cellular behavior. At the same time, AI virtual cells are emerging as a promising direction, supported by progress in single-cell foundation models, perturbation-response prediction, and multimodal learning. Although these approaches can predict cell states and responses, they often do not explain which molecular mechanisms drive cellular phenotypes, hindering hypothesis prioritization for experimental testing and therapeutic targeting. Unifying learning-based approaches with causal models grounded in prior knowledge could bridge this interpretability gap by encoding established cellular mechanisms through molecular states, causal interactions, and biological constraints. A key aspect of such unification is representing signaling beyond pathway-level activity as mechanistic cascades that capture biochemical dependencies, crosstalk, alternative routing, and intervention points. In this Opinion, we discuss computational frameworks that reconstruct biochemical signaling networks using omics data and offer a basis for mechanistically linking extracellular stimuli to transcriptional regulation and metabolic control. Such frameworks could serve as scaffolds for future whole-cell models that also account for the dynamic, stochastic, spatial, physical, and multiscale realities of the living cell. Together, causal models of signaling and learning-based advances could support interpretable whole-cell models for mechanistic hypothesis generation and intervention design.

Förderung

  • Swiss National Science Foundation
  • Swiss Federal Institute of Technology Lausanne

Zitieren

APA

Liaskos, D., & Hatzimanikatis, V. (2026). A Key Piece of the Whole-Cell Puzzle: Mechanistic signaling for interpretable whole-cell models. Current Opinion in Systems Biology, 100613. https://doi.org/10.1016/j.coisb.2026.100613

Vancouver

Liaskos D, Hatzimanikatis V. A Key Piece of the Whole-Cell Puzzle: Mechanistic signaling for interpretable whole-cell models. Current Opinion in Systems Biology. 2026:100613. doi:10.1016/j.coisb.2026.100613

BibTeX
@article{Liaskos2026,
  author = {Liaskos, David and Hatzimanikatis, Vassily},
  title = {A Key Piece of the Whole-Cell Puzzle: Mechanistic signaling for interpretable whole-cell models},
  journal = {Current Opinion in Systems Biology},
  year = {2026},
  pages = {100613},
  doi = {10.1016/j.coisb.2026.100613}
}
Kostenlos starten →