Topology-Driven Systems Modelling of M.Tuberculosis Host–Pathogen Dynamics Reveals Network Control Nodes Governing Infection Outcome

Understanding the multiscale dynamics of host–pathogen interactions remains a central challenge in systems biology. Here, we develop an integrated computational framework that unifies mechanistic pathway modeling with global network topology to study infection dynamics in Mycobacterium tuberculosis (MTB). We first construct a four-variable ordinary differential equation (ODE) model grounded in curated immune signaling architecture, capturing the population dynamics of extracellular bacilli, macrophages, activated immune cells, and infected host compartments. While this mechanistic formulation preserves biological detail, its dimensional complexity limits analytical tractability. To address this, we incorporate global interactome topology derived from large-scale pathway integration and centrality analysis. Network metrics reveal a structural bottleneck characterized by two dominant opposing regulatory forces: pathogen proliferation driven by metabolic and energetic hubs, and host immune activation governed by highly connected sensing and signaling modules. Guided by this topological structure, we perform a coarse-grained dynamical reduction that aggregates immune and pathogen compartments into effective variables representing collective host defense and pathogen load. The resulting reduced system preserves the emergent interaction laws of the full mechanistic model while yielding a mathematically transparent two-variable dynamical framework. This topology-guided reduction establishes a principled bridge between pathway-level mechanistic detail and systems-level abstraction. More broadly, the study introduces a scalable strategy for translating complex biological networks into analytically tractable dynamical models, providing a generalizable foundation for multiscale modeling of host–pathogen systems.