AutSim: Principled, data-driven model development and abstraction for synaptic translation in Fragile X Syndrome (FXS) and healthy control.
Data provenance and model complexity are recurring challenges when simulating neural function, particularly signaling, at the level of detail needed to address disease conditions. Here we report a data-driven modeling and model abstraction framework applied to ~40 signaling pathways to study translation in FXS. We used a detailed biochemical (Mass action+ODE) model with signaling pathways spanning from receptors such as mGluR1/5, TrKB, EGFR and b2AR, to intervening kinases, to protein synthesis. We used over 300 published experiments to validate and parameterize the model. Since the model has a large parameter space, we developed a pipeline to hierarchically optimize the model to fit the experimental database and score the model based on how closely the model outcome matches experiments. To anchor this enormous parameter fitting process, we built abstract models with all the major nodes of the detailed model using the HillTau formulation (Bhalla US, PLoS CompBiol., 2021). The abstract models were smaller with a much reduced parameter space, 2-3 orders faster than the detailed model and easier to optimize. These abstract models were then used to synthesize experiments to provide a scaffold of input-output properties that the detailed model subsets must fit. In summary, we have developed a principled, hierarchical methodology to use experimental data to generate both detailed and abstract models of complex cellular signaling, which are all valuable resources for the field.