Digital Twins in Systems Biology Modeling

Abstract

Digital twins, virtual replicas of biological systems that integrate multi-scale data with mechanistic and data-driven models to simulate, predict, and optimise system behaviour, are emerging as a transformative paradigm in systems biology, yet the translation of digital twin concepts from engineering into biological applications varies widely across modelling scales and application domains. We evaluated 214 digital twin programmes for systems biology across centres in Italy, Estonia, and Spain between 2018 and 2022, spanning five modelling categories: whole-cell mechanistic simulations, organ-level physiological models, patient-specific disease twins, metabolic flux digital twins, and multi-scale integrative frameworks. A Digital Twin Systems Biology Index (DTSBI) was constructed from five sub-scores -- predictive accuracy for system dynamics, personalisation fidelity from individual-level data, multi-scale integration coherence, experimental validation concordance, and clinical or industrial translatability -- with weights from regression against sustained adoption into research or clinical pipelines. DTSBI correlated with adoption at r = +0.84 and discriminated adopted from non-adopted twins with an AUC of 0.882. Metabolic flux digital twins scored highest (mean DTSBI 0.824), while whole-cell simulations trailed at 0.598. Only 35.5 percent exceeded the 0.75 threshold. Predictive accuracy carried the largest weight (beta = +0.278), followed by personalisation fidelity (beta = +0.230).