MOTIVATION LETTER The failure of neuroinflammation models to translate from bench to bedside is primarily a quantitative problem, not a biological one. Across the field, in-vitro systems capture isolated molecular events while clinical trials test whole-organism outcomes, and no formal framework connects the two scales. My research program addresses exactly this gap. I develop mechanistic, dynamical-systems models that simulate how molecular-level pharmacological interventions propagate through neural circuits to produce behavioral and clinical outcomes. The Conjunctive Consolidation Threshold (CCT) model, my primary framework, is a tripartite ODE system coupling dopaminergic reward-prediction error, NMDAR-dependent long-term potentiation, and affective contrast to predict whether reward-memory encoding is prevented. I calibrated all 14 free parameters using Bayesian MCMC (PyMC DEMetropolisZ) against priors elicited from a systematic screen of 1,847 published records. All five pre-registered hypotheses were confirmed, with posterior super-additivity of 13 to 22 percentage points across model versions. Three sole-authored preprints are under review at peer-reviewed journals, and a co-authored paper is under review at Alcohol (Elsevier). The EMD Group Research Grants programme funds early-career researchers developing predictive in-vitro models for human neuroinflammatory diseases. My CCT framework is directly adaptable to this call. The same mathematical architecture that simulates reward-memory encoding can model the cytokine-mediated synaptic dysregulation, microglial activation dynamics, and dopaminergic neurodegeneration central to Parkinson's disease. The model's Bayesian calibration pipeline is host-agnostic; it can be parameterized from published human post-mortem transcriptomic data, CSF cytokine measurements, and longitudinal imaging cohorts. This addresses the selection criterion requiring models that recapitulate human disease pathology, including heterogeneity. My approach requires rigorous quantitative integration of existing human data, which is my documented strength, and does not require a wet lab. I am an independent researcher based in Nigeria, with a B.Pharm from the University of Ibadan and enrollment in the M.Sc. Digital Health programme at the Hasso Plattner Institute, University of Potsdam, beginning Winter Semester 2026/27. I have endorsements from Kent Berridge (University of Michigan), Samuel Gershman (Harvard), Nathaniel Daw (Princeton), and Marcelo Mattar (NYU). My computational toolkit spans ODE/RK45 simulation, PyMC Bayesian inference, topological data analysis, and protein language models. The EMD programme's emphasis on scientific excellence, innovation, and collaboration with Merck scientists aligns with my goal of translating mechanistic models into clinically actionable predictions. I seek the research grant to support the adaptation and validation of the CCT framework for neuroinflammatory disease modeling. RESEARCH STATEMENT The Conjunctive Consolidation Threshold (CCT) model was built to answer a specific question: under what pharmacological conditions does the brain fail to consolidate reward-associated memories, and can that failure be predicted before it occurs? The model couples three axes: dopaminergic reward-prediction error, NMDAR-dependent long-term potentiation, and affective contrast. These are represented as a system of coupled ordinary differential equations solved with RK45. The model was calibrated using Bayesian MCMC with 14 free parameters and literature-elicited priors from a 1,847-record screen. All five pre-registered hypotheses (H1-H5) were confirmed. Posterior analysis showed super-additive effects of 13 to 22 percentage points across model versions, meaning the combined pharmacological intervention outperformed any single-axis manipulation. This result has direct implications for combination therapy design in addiction, but the methodological core is transferable. The EMD Research Grants call asks for predictive in-vitro models of human neuroinflammatory diseases, with Parkinson's disease as an explicit example. Parkinson's is characterized by progressive dopaminergic neurodegeneration in the substantia nigra, driven partly by neuroinflammatory processes: microglial activation, elevated pro-inflammatory cytokines (TNF-alpha, IL-1beta, IL-6), and mitochondrial dysfunction. These processes involve feedback loops between neuronal stress signals and glial responses. A static molecular assay cannot capture this. A coupled ODE model can. My proposed project, CCT-Neuro, adapts the CCT architecture to model the neuroinflammatory cascade in Parkinson's disease. The three coupled axes become: (1) dopaminergic neuron integrity and dopamine release dynamics, (2) microglial activation state and cytokine production kinetics, and (3) synaptic plasticity and network-level compensation in the basal ganglia-thalamocortical loop. The model will be parameterized from published human data: post-mortem substantia nigra transcriptomic profiles, longitudinal CSF cytokine measurements from de novo Parkinson's cohorts, and dopamine transporter imaging (DaTSCAN) trajectories. Bayesian calibration will follow the same protocol validated in the CCT model, with priors elicited from a systematic literature screen and posterior predictive checks against held-out clinical data. The model's output will be a quantitative prediction of disease progression trajectory under specified pharmacological interventions, including anti-inflammatory agents, dopaminergic replacement therapies, and combination regimens. This directly addresses the selection criterion that models must recapitulate human disease pathology, including heterogeneity. The model will explicitly represent patient-to-patient variability through hierarchical Bayesian parameter distributions, rather than averaging it away. Feasibility is established. I have already built and validated the core computational infrastructure: neurocascade, a receptor-to-behavior brain-circuit simulation engine coupling pharmacokinetics to receptor binding to Wilson-Cowan circuit dynamics, with 62 of 62 tests passing. The Bayesian calibration pipeline is battle-tested. The data sources are public and accessible. The primary risk is model misspecification, which I mitigate through pre-registered validation gates, a practice I have followed consistently. In my ergofluids project, when the first real-data gate did not meet its pre-registered criterion, I reported the negative result directly rather than reframing it. That rigor will apply here. The project aligns with Merck KGaA's strategic interest in neuroinflammation and translational neuroscience. A mechanistic, quantitative model that predicts disease trajectory from molecular interventions is a tool that can accelerate drug development decisions, prioritize compound selection, and reduce reliance on animal models that frequently fail to predict human outcomes. I am available for collaboration with Merck scientists throughout the project, including the deep-dive workshop for finalists. The requested funding will support computational infrastructure, data acquisition, and travel for collaboration. SHORT-ANSWER ESSAY: RELEVANCE TO EMD RESEARCH GRANTS MISSION The EMD Research Grants programme funds early-career researchers developing predictive in-vitro models for human neuroinflammatory diseases. My CCT-Neuro project is a direct response to this call. The programme's selection criteria emphasize scientific excellence, innovation, relevance to human disease pathology, feasibility, and collaboration potential. My track record demonstrates each. Scientific excellence: the CCT model's five confirmed pre-registered hypotheses and three sole-authored preprints under peer review. Innovation: the application of dynamical-systems modeling with Bayesian calibration to neuroinflammation, a domain dominated by static molecular assays. Relevance: the model is parameterized from human post-mortem and longitudinal clinical data, not animal models. Feasibility: the computational infrastructure exists and is tested. Collaboration: I have endorsements from leading computational neuroscientists and am enrolled in the Hasso Plattner Institute's Digital Health programme, which has direct ties to German research institutions. The project also addresses the programme's emphasis on models that recapitulate human disease heterogeneity through hierarchical Bayesian parameter distributions. My independent researcher status, based in Nigeria, brings a perspective that is underrepresented in neuropharmacology, and my work demonstrates that rigorous computational research does not require a wet lab. CHECKLIST - [ ] Verify EMD Group Research Grants 2026 application portal access and account creation - [ ] Confirm deadline of 2026-08-31 and timezone for submission - [ ] Prepare CV in required format, including ORCID (0009-0001-9272-6735), GitHub, and personal site - [ ] Upload MOTIVATION LETTER (300-500 words, final word count verified) - [ ] Upload RESEARCH STATEMENT (400-600 words, final word count verified) - [ ] Upload SHORT-ANSWER ESSAY (200-350 words, final word count verified) - [ ] Compile list of three preprints (OSF/Zenodo DOIs) and co-authored Alcohol paper as supporting evidence - [ ] Prepare one-page summary of CCT model architecture and validation results for reviewer reference - [ ] Draft letter of endorsement from Kent Berridge or Samuel Gershman (confirm willingness) - [ ] Confirm enrollment status at Hasso Plattner Institute and obtain proof of enrollment letter - [ ] Verify all claims regarding pre-registration (OSF links) and Bayesian calibration details - [ ] Review programme terms for any undisclosed eligibility restrictions on independent researchers or non-EU applicants - [ ] Prepare non-confidential project summary for initial application (per programme instructions) - [ ] Identify potential Merck KGaA collaborators in neuroinflammation or computational neuroscience for the deep-dive workshop stage EDITOR NOTES - Eligibility risk: The programme page lists no explicit geography or career-stage restrictions, but Merck KGaA is a German company; confirm whether non-EU independent researchers are eligible before investing further time. The applicant's enrollment at HPI/Potsdam strengthens the EU connection but does not guarantee eligibility. - Verification needed: The profile states three preprints are under review at IART, PNPBP, and NBR, and a co-authored paper at Alcohol (Elsevier). Confirm these journals exist and the submission status is current. Also verify the 1,847-record screen number and the 13-22pp super-additivity range against the actual preprints. - Gap: The profile does not specify the exact funding amount requested. The programme lists no amount, but the applicant should insert a specific budget request (e.g., 20,000 to 50,000 EUR) in the motivation letter or research statement if the application form requires it. - Gap: The applicant's employment history includes a role as National Product Manager at Synthcare starting March 2026. This is a commercial role; clarify how it coexists with the independent researcher track and whether it creates any conflict of interest with Merck KGaA, a pharmaceutical company. - Gap: The profile lists psyche-twin as a knowledge-graph architecture for self-modeling. This is not mentioned in the application materials, which is correct, but the applicant should be prepared to explain it if asked during the deep-dive workshop, as it may appear on their personal site. - The chosen research line is CCT, adapted as CCT-Neuro for neuroinflammation. This is consistent with the recommended framing angle. The cardiotoxicity topology study and TOPOLOGIX are not mentioned in the application materials because they address drug resistance and toxicity, not neuroinflammation; they are relevant only as evidence of methodological breadth, which is covered by the neurocascade mention in the research statement.