Network Simulation Models: Integrating Multiscale Information to Generate Actionable Insights into Complex Psychiatric Illnesses

Brisbane Convention & Exhibition Centre 

With a sea of diverse data types ranging from molecular to macroscale emerging across various fields of neuroscience, overarching frameworks are needed to integrate and derive coherent understandings. Network-based models have emerged as a powerful tool for this exact purpose, offering computational approaches that can map complex interactions across biological systems. My talk will demonstrate how network frameworks, such as network-based simulations of disease processes, have matured to the point where we can both generate and test hypotheses that cross spatiotemporal scales. These models, including sophisticated agent-based protein spreading models that track the propagation of molecular changes through biological systems, have been immensely successful in neurology, providing unprecedented insights into atrophy patterns and longitudinal progression across multiple neurodegenerative conditions. Critically, these computational approaches have now opened a path to application in enigmatic psychiatric disorders such as schizophrenia, bridging our understanding of complex neurological phenomena.

I will cover a series of novel studies by multiple different research groups using network simulations that combine macroscale data, such as in vivo diffusion imaging-based tractography, with molecular scales, such as ex vivo gene expression. These integrative approaches aim to comprehensively explain the pathophysiology of complex mental illnesses and uncover previously unknown mechanisms underlying psychiatric medications. By linking brain structural connectivity with molecular-level information, researchers can develop more nuanced models of disease progression and treatment response. I'll conclude by raising several critical future challenges, highlighting the field's trajectory from mechanistic explanation to prediction. This includes developing predictive models to guide treatment selection and transitioning from observational studies in psychiatric neuroscience to more powerful causal experimental designs, such as placebo-controlled randomized controlled trials, ultimately moving towards a more personalized and precise approach to understanding and treating complex mental health conditions.