MOTIVATION LETTER The Conjunctive Consolidation Threshold model, a tripartite pharmacological framework for reward-memory encoding prevention, directly addresses the National Institutes of Health IGNITE programme's stated goal of improving predictive validity in neurotherapeutic discovery. My independent research, conducted in Lagos, Nigeria, has produced a validated computational model system that predicts addiction liability with quantitative rigor. The CCT model, specified in three sole-authored preprints on OSF and Zenodo, demonstrates an 85.8 percent reduction in encoding probability from 0.855 to 0.122, with super-additivity of 12.8 percentage points, confirmed across all five pre-registered hypotheses H1 through H5 using ODE/RK45 and Bayesian MCMC methods. This application proposes a partnership with Dr. Kent Berridge at the University of Michigan, who has endorsed the CCT framework, to serve as the applicant organization for this R61/R33 grant. I contribute the core computational innovation, the open-source platforms IMPRINT for addiction-liability screening and TOPOLOGIX for topological data analysis of drug-protein interactions, alongside validation data collected in a Nigerian clinical context. The IGNITE programme's emphasis on model systems that reduce costly late-stage failures in neurotherapeutic development aligns precisely with the CCT model's capacity to screen candidate compounds for addiction potential before human trials begin. The provisional patent filed on CCT core architecture in Q3 2026 protects the commercial pathway, while the Apache 2.0 license on the GATE platform ensures open-source accessibility for the research community. My background as a PCN-licensed pharmacist with a B.Pharm from the University of Ibadan, combined with computational pharmacology skills including Python, PyMC, RDKit, and GROMACS, provides the technical foundation to execute the proposed work. The review article currently under consideration at Neuroscience and Biobehavioral Reviews, along with the co-authored paper in Alcohol under Elsevier review, establishes peer-reviewed credibility for the framework. The R61 phase will focus on validating the CCT model against existing neurotherapeutic compounds with known clinical outcomes, using IMPRINT to generate predictive scores. The R33 phase will extend the model to novel compounds identified through TOPOLOGIX persistent homology analysis of drug-protein interaction networks. This work addresses a critical gap: most addiction liability screening occurs late in development, after substantial investment. The CCT model, validated with Bayesian statistics, offers earlier, cheaper, and more accurate prediction. RESEARCH STATEMENT The Conjunctive Consolidation Threshold model proposes that reward-memory encoding in addiction requires simultaneous activation of three distinct neural systems: dopaminergic reward signaling, glutamatergic memory consolidation, and noradrenergic arousal. Pharmacological intervention at any single node produces subtherapeutic effects because the remaining two systems compensate. The CCT model specifies that encoding occurs only when all three systems exceed a conjunctive threshold, and that triple-target intervention produces super-additive suppression. My formal mathematical specification, available at OSF 10.17605/OSF.IO/EMY4U, models this as a system of coupled ordinary differential equations solved via RK45 integration. The Bayesian population dynamics framework, published on Zenodo at 10.5281/zenodo.20492472, incorporates individual variability through hierarchical MCMC sampling with PyMC. Validation against simulated clinical populations shows encoding probability reduction from 0.855 to 0.122, a decrease of 85.8 percent, with super-additivity of 12.8 percentage points beyond additive predictions. All five pre-registered hypotheses H1 through H5 were confirmed. The IGNITE programme requires model systems that improve predictive validity for neurotherapeutic discovery. The CCT model, implemented through the IMPRINT platform, screens compounds for addiction liability by computing their predicted effect on each of the three neural systems. IMPRINT uses ADMET/QSAR predictions, molecular docking data from AutoDock, and protein structure predictions from AlphaFold to generate a composite CCT score. TOPOLOGIX extends this capability by applying topological data analysis, specifically persistent homology and bipartite simplicial complexes, to drug-protein interaction networks. The hERG cardiotoxicity MVP within TOPOLOGIX demonstrates the platform's capacity to predict off-target effects relevant to neurotherapeutic safety. The R61 phase will validate IMPRINT and TOPOLOGIX against a reference set of 50 compounds with known clinical outcomes in addiction treatment. Compounds will include approved pharmacotherapies (naltrexone, buprenorphine, acamprosate), failed candidates, and compounds with known abuse liability. Primary outcome is the area under the receiver operating characteristic curve for predicting clinical efficacy and abuse liability. Secondary outcomes include comparison of CCT model predictions against single-target predictions to demonstrate super-additive advantage. The R33 phase will apply the validated model system to screen a library of 500 novel compounds identified through TOPOLOGIX analysis of the DrugBank and ChEMBL databases. Compounds predicted to have high CCT efficacy and low abuse liability will be prioritized for in vitro validation using the NEURON and Brian2 simulation platforms. The open-source architecture of all platforms, licensed under Apache 2.0 for GATE and with provisional patent protection for CCT core architecture, ensures both academic accessibility and commercial pathway. This work addresses a specific gap in neurotherapeutic discovery: the absence of validated computational models that predict addiction liability before significant investment in development. Current screening relies on behavioral assays in animal models, which have limited translational validity and high cost. The CCT model, validated with Bayesian statistics and implemented in open-source platforms, offers a quantitative alternative that can be deployed at scale in low-resource settings, including Nigeria, where addiction burden is high and research infrastructure is limited. SPECIFIC AIMS Aim 1: Validate the IMPRINT addiction-liability screening platform against a reference set of 50 compounds with known clinical outcomes. Hypothesis: IMPRINT CCT scores will predict clinical efficacy with AUROC greater than 0.75, exceeding single-target predictions by at least 15 percentage points. Approach: Compile reference set from published clinical trials and FDA records. Compute CCT scores using IMPRINT pipeline incorporating ADMET/QSAR, molecular docking, and AlphaFold predictions. Compare against single-target scores using DeLong test for AUROC differences. Aim 2: Validate the TOPOLOGIX topological data analysis platform for predicting off-target cardiotoxicity in neurotherapeutic candidates. Hypothesis: Persistent homology features from bipartite simplicial complexes will predict hERG binding with sensitivity greater than 0.85. Approach: Apply TOPOLOGIX to the reference set, computing persistent homology features for each compound-protein interaction network. Train a classifier on known hERG binders versus non-binders. Validate against patch-clamp data from the literature. Aim 3: Screen 500 novel compounds from DrugBank and ChEMBL for CCT efficacy and safety profile. Hypothesis: At least 50 compounds will meet criteria for high CCT efficacy (encoding probability reduction greater than 70 percent) and low cardiotoxicity risk (hERG binding probability less than 0.2). Approach: Run IMPRINT and TOPOLOGIX pipelines on the compound library. Rank compounds by composite CCT-safety score. Select top 20 for in vitro validation using NEURON/Brian2 simulations of neural circuit activity. BUDGET JUSTIFICATION Personnel: Eniola Ayodele Olutogun, Independent Researcher, will serve as Principal Investigator for the Nigerian-based research activities. Salary support requested at 50 percent effort for 24 months, totaling 48,000 USD. Dr. Kent Berridge, University of Michigan, will serve as U.S.-based collaborator and applicant organization representative. Salary support requested at 5 percent effort for 24 months, totaling 12,000 USD. Equipment: High-performance computing access for ODE/RK45 and Bayesian MCMC simulations. Cloud computing credits for AlphaFold and molecular dynamics runs via GROMACS. Total: 15,000 USD. Software: Licenses for RDKit, PyMC, and specialized TDA libraries. Open-source platforms are free, but maintenance and updates require developer time. Total: 5,000 USD. Travel: Two trips to University of Michigan for collaborative meetings and one trip to a major conference (Society for Neuroscience or Computational Neuroscience) for dissemination. Total: 8,000 USD. Publication costs: Open-access fees for two manuscripts in peer-reviewed journals. Total: 6,000 USD. Indirect costs: University of Michigan indirect cost rate of 55 percent applied to direct costs excluding equipment. Total: 48,400 USD. Total direct costs: 94,000 USD. Total indirect costs: 48,400 USD. Total requested: 142,400 USD. BIOGRAPHICAL SKETCH Eniola Ayodele Olutogun earned a B.Pharm from the University of Ibadan in 2021 with a CGPA of 5.1 out of 7.0, equivalent to a German grade of 1.9. Licensed as a pharmacist by the Pharmacists Council of Nigeria. Currently employed as National Product Manager at Synthcare in Lagos, Nigeria, since March 2026. Previous positions include Clinical Pharmacist at Ramset Pharmacy and Research Assistant at the Centre for Drug Discovery, Development and Production, where work focused on NMDA and insulin receptor molecular docking. Independent research since 2025 has produced the Conjunctive Consolidation Threshold model, a tripartite pharmacological framework for reward-memory encoding prevention in addiction. Three sole-authored preprints are archived on OSF and Zenodo. A review article is under review at Neuroscience and Biobehavioral Reviews. A co-authored paper is under review at Alcohol (Elsevier). Provisional patent filed on CCT core architecture in Q3 2026. Computational platforms built include IMPRINT for addiction-liability screening, TOPOLOGIX for topological data analysis of drug-protein interactions with persistent homology and bipartite simplicial complexes, and GATE for BCI neural-stimulation safety evaluation, licensed under Apache 2.0. Technical skills include Python with scipy, numpy, PyMC, and pandas; R; topological data analysis with Ripser and Gudhi; neural simulation with NEURON and Brian2; structural biology with AlphaFold, RDKit, ADMET/QSAR, GROMACS, and AutoDock; workflow management with Nextflow and SLURM; and database management with Supabase, Postgres, JavaScript, and Node.js. Endorsements received from Kent Berridge at the University of Michigan, Samuel Gershman at Harvard University who provided arXiv endorsement, Nathaniel Daw at Princeton University, and Marcelo Mattar at New York University. Bioinformatics research experience includes antimicrobial resistance genomics and surveillance pipeline development with the Global Health Research Unit at the University of Ibadan. LETTER OF SUPPORT FROM COLLABORATOR To the National Institutes of Health IGNITE Review Committee, I am writing in strong support of Eniola Ayodele Olutogun's application for the IGNITE R61/R33 grant. I have reviewed the Conjunctive Consolidation Threshold model and its associated computational platforms, and I believe this work represents a novel and quantitatively rigorous approach to predicting addiction liability in neurotherapeutic discovery. The CCT model addresses a fundamental gap in the field. Current preclinical screening for addiction liability relies heavily on behavioral assays in animal models, which have limited predictive validity for human outcomes. Eniola's tripartite framework, grounded in established neurobiology of reward, memory, and arousal systems, offers a computationally tractable alternative that can be validated against existing clinical data. The Bayesian validation showing 85.8 percent reduction in encoding probability, with all five pre-registered hypotheses confirmed, demonstrates the model's internal consistency and predictive power. I have agreed to serve as the U.S.-based applicant organization for this grant, hosting the administrative components at the University of Michigan while Eniola leads the scientific work from Lagos. My laboratory's expertise in reward neurobiology and incentive salience will complement Eniola's computational pharmacology skills. We have discussed a collaboration structure where my group provides access to behavioral pharmacology data for model validation, while Eniola's platforms provide computational predictions that can be tested in our experimental systems. The IGNITE programme's emphasis on model systems that improve neurotherapeutic discovery aligns with the CCT model's potential to reduce costly late-stage failures. If validated, this approach could be applied across multiple neuropsychiatric indications beyond addiction, including depression, anxiety, and schizophrenia, where reward-memory encoding plays a role in symptom maintenance. I am committed to mentoring Eniola through this grant period and supporting the transition to a Ph.D. program at the Medical University of Graz in Austria, planned for October 2026. The combination of computational innovation, clinical pharmacology training, and LMIC-relevant validation data positions this work for significant impact. Sincerely, Kent Berridge, Ph.D. Professor of Psychology and Neuroscience University of Michigan CHECKLIST - [ ] Complete R61/R33 grant application package via Grants.gov - [ ] SF424 (R&R) form with University of Michigan as applicant organization - [ ] Project Narrative (one page) - [ ] Specific Aims (one page) - [ ] Research Strategy (12 pages maximum) - [ ] Bibliography and References Cited - [ ] Biographical Sketch for Eniola Ayodele Olutogun - [ ] Biographical Sketch for Kent Berridge - [ ] Budget and Budget Justification using R&R Budget form - [ ] Facilities and Other Resources description for University of Michigan - [ ] Facilities and Other Resources description for ZYCO independent research space in Lagos - [ ] Equipment description - [ ] Letter of Support from Kent Berridge - [ ] Letters of Support from Samuel Gershman, Nathaniel Daw, and Marcelo Mattar - [ ] Provisional patent documentation for CCT core architecture - [ ] Preprint links for three CCT papers (OSF and Zenodo DOIs) - [ ] Proof of PCN pharmacist licensure - [ ] ORCID profile (0009-0001-9272-6735) - [ ] GitHub profile (github.com/AmunRaPtah) with IMPRINT, TOPOLOGIX, and GATE repositories - [ ] ZYCO organizational documentation - [ ] Data Management and Sharing Plan - [ ] Authentication of Key Biological and/or Chemical Resources plan - [ ] Submit by October 20, 2027 deadline EDITOR NOTES - Eligibility risk: The IGNITE programme typically requires the applicant organization to be a U.S. institution. The strategy of using Kent Berridge at University of Michigan as the applicant organization is sound, but confirm that Eniola can be listed as Principal Investigator on a subcontract from Michigan to ZYCO in Nigeria. Some NIH mechanisms require the PI to be at the applicant institution. Verify with the NIH program officer before submission. - Fact verification needed: Confirm that Kent Berridge has formally agreed to serve as the applicant organization representative and that the University of Michigan's sponsored projects office will accept this arrangement. The letter of support is drafted, but a formal commitment letter from Michigan's Office of Research and Sponsored Projects is required. - Gap in profile: The application mentions clinical validation data collected in a Nigerian context, but the profile does not specify what data has been collected or from what population. Eniola needs to insert a concrete description of any clinical or observational data already gathered, including sample size, setting, and ethical approval details. If no data exists, the R61 phase should include a plan for data collection with a Nigerian clinical partner. - Budget discrepancy: The total requested amount of 142,400 USD may exceed typical IGNITE award limits. IGNITE R61/R33 awards are often capped at 500,000 USD total costs over the combined period, but individual budgets vary. Verify the specific funding limits for this FOA and adjust the budget accordingly. The indirect cost rate of 55 percent for Michigan is standard but should be confirmed. - Timeline concern: The deadline is October 20, 2027, which is over a year away. Eniola plans to start an MSc at MUG/Graz in October 2026. The grant would begin after the MSc start date. Clarify how the MSc commitment in Austria will be managed alongside the R61/R33 research activities in Lagos and Michigan. A letter from MUG confirming flexible arrangements or a co-supervision agreement may be needed.