## Abstract The richness of chemical signalling in the brain supports the flexible neural dynamics required for adaptive behaviour. The NMDA receptor, one of the most ubiquitous receptors in the brain, controls synaptic plasticity and mediates learning and memory formation. Its dysfunction is theorised to be one of the mechanisms of schizophrenia. Profiling the changes and dysfunctions of ion channel receptors can elucidate the biological substrate of functional brain impairments and facilitate individual patients’ personalised treatment strategies. The aims of this project are two-fold: methodological—developing a robust and translational dynamic causal model able to profile NMDA-R dynamics—and clinical—studying NMDA-R dysfunctions and their trajectory within a longitudinal study of a large cohort of schizophrenic patients. The NMDA-R DCM will first be developed and validated on animal data with a high signal-to-noise ratio. The ground-truth dataset comes from a repeated measures study, with one of the measurements involving a pharmacological intervention using an NMDA-R blocker. Next, the model will be translated to human rsEEG data with the same within-subject design. Finally, the validated NMDA-R DCM will be applied to schizophrenic patients. The prospective design of the study will allow for the assessment of the predictive power of generative embedding for individual patients’ clinical trajectories. This approach might help clinicians predict clinical outcomes and tailor treatments to individual patients. We plan to disseminate the project’s results through two main channels: peer-reviewed publications of both methodological and clinical advancements, and by sharing the relevant source code for the analyses, including the envisioned additions to the TAPAS package, thereby fostering replicability, reproducibility, and rapid distribution within the community. Ultimately, this project aims to help establish computational assays as a standard procedure in clinical practice.