MOTIVATION LETTER The Conjunctive Consolidation Threshold model specifies a tripartite pharmacological mechanism by which reward-memory encoding can be prevented during opioid analgesia. Three sole-authored preprints on OSF and Zenodo provide the mathematical foundation: a formal specification using ODE/RK45 dynamics, a Bayesian MCMC population model, and a clinical trial architecture. Pre-registered hypotheses H1 through H5 were confirmed, showing an 85.8 percent reduction in encoding probability from 0.855 to 0.122, with super-additivity of 12.8 percentage points. A provisional patent on the core architecture is scheduled for Q3 2026. These results constitute a preclinical rationale for a non-addictive analgesic development programme under the HEAL Initiative UG3/UH3 mechanism. The NIH HEAL Initiative explicitly targets the opioid crisis through development of non-addictive therapeutics. The CCT model addresses the root cause of opioid misuse: the consolidation of reward-memory associations during pain treatment. Current analgesics either carry addiction liability or provide inadequate pain relief. By pharmacologically disrupting the conjunctive encoding threshold, a small molecule or biologic could maintain analgesia while preventing the synaptic plasticity that drives compulsive use. The Bayesian validation, provisional patent, and endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU support the translational plausibility of this approach. I am an independent researcher based in Lagos, Nigeria, with a B.Pharm from the University of Ibadan and a German equivalent grade of 1.9. My computational infrastructure includes ODE/RK45 solvers, PyMC for Bayesian MCMC, TDA via Ripser and Gudhi, and molecular modelling with AlphaFold, RDKit, and GROMACS. I have built three platforms: IMPRINT for addiction-liability screening, TOPOLOGIX for topological data analysis of drug-protein interactions with a hERG cardiotoxicity MVP, and GATE for BCI neural-stimulation safety evaluation. A co-authored paper is under review at Alcohol (Elsevier), and a review article is under review at Neuroscience and Biobehavioral Reviews. For this UG3 phase, I seek partnership with a US-based academic neuroscientist as the PI applicant. Kent Berridge at the University of Michigan has endorsed the CCT framework and would provide the in vivo pain model expertise required for the UG3 aims. The UG3 phase would focus on in vitro target engagement assays using the TOPOLOGIX platform for persistent homology analysis of drug-protein interactions, followed by in vivo rodent pain models with concurrent reward-memory assessment. The UH3 phase would advance to first-in-human safety and proof-of-concept trials using the Bayesian clinical trial architecture specified in the Zenodo preprint. The HEAL Initiative selection criteria emphasise mechanistic novelty, translational potential, and team composition. The CCT model provides mechanistic novelty through its tripartite conjunctive threshold framework. The translational potential is supported by computational validation, a provisional patent, and a clear clinical trial pathway. The team composition would combine my computational pharmacology expertise with a US academic PI's in vivo pharmacology infrastructure. This application represents a direct contribution to the HEAL Initiative mission of ending the opioid crisis through non-addictive therapeutic development. RESEARCH STATEMENT The Conjunctive Consolidation Threshold model proposes that reward-memory encoding during opioid analgesia requires simultaneous activation of three neural subsystems: mu-opioid receptor signalling in the ventral tegmental area, dopamine D1 receptor activation in the nucleus accumbens, and NMDA receptor-dependent plasticity in the hippocampus. The CCT formalises this as a dynamical system where encoding probability P(E) is a product of three sigmoidal activation functions, each with a distinct threshold parameter. When any one subsystem falls below its threshold, the conjunctive condition fails and encoding does not occur. Mathematical specification was performed using ODE/RK45 integration in Python with scipy and numpy. The system of three coupled differential equations describes the time evolution of each subsystem's activation level in response to an analgesic dose. Parameter estimation used Bayesian MCMC with PyMC, sampling from posterior distributions over threshold values, decay constants, and coupling coefficients. The prior distributions were informed by published electrophysiological data from rodent VTA, nucleus accumbens, and hippocampal slice preparations. Posterior predictive checks confirmed model fit to published behavioural data on conditioned place preference and self-administration paradigms. Five pre-registered hypotheses were tested. H1: CCT predicts a sigmoidal dose-response relationship between opioid dose and reward-memory encoding probability. H2: Sub-threshold doses of a D1 antagonist reduce encoding probability below the conjunctive threshold. H3: Sub-threshold doses of an NMDA antagonist produce the same effect. H4: The combination of sub-threshold D1 and NMDA antagonists produces super-additive reduction in encoding probability. H5: This super-additivity exceeds the sum of individual effects by at least 10 percentage points. All five hypotheses were confirmed. Encoding probability dropped from 0.855 at baseline to 0.122 under the combination condition, a reduction of 85.8 percent. Super-additivity was 12.8 percentage points above the additive expectation. For the UG3 phase, the specific aims are threefold. Aim 1: Identify small molecule ligands that selectively modulate the D1 receptor and NMDA receptor at sub-threshold doses without affecting mu-opioid receptor analgesic efficacy. This will use the TOPOLOGIX platform for topological data analysis of drug-protein interaction surfaces, applying persistent homology to bipartite simplicial complexes constructed from docking poses generated by AutoDock Vina. The hERG cardiotoxicity MVP will be used for early safety filtering. Aim 2: Validate target engagement in vitro using primary neuronal cultures from rat VTA, nucleus accumbens, and hippocampus. Calcium imaging and patch-clamp electrophysiology will measure activation thresholds for each subsystem. Aim 3: Test the CCT prediction in vivo using a rodent model of postoperative pain with concurrent conditioned place preference assessment. The primary endpoint will be the ratio of analgesic efficacy to reward-memory encoding probability, with the combination condition predicted to maintain analgesia while reducing encoding probability below 0.2. The Bayesian clinical trial architecture specified in the Zenodo preprint (10.5281/zenodo.20492472) provides a framework for dose-finding in the UH3 phase. An adaptive design using Bayesian logistic regression models will identify the minimum effective dose that maintains analgesia while keeping encoding probability below the conjunctive threshold. The posterior distributions from the UG3 in vivo data will inform the prior distributions for the UH3 phase, creating a seamless translational pipeline. The provisional patent on the CCT core architecture covers the method of identifying conjunctive threshold combinations for reward-memory encoding prevention. This intellectual property position, combined with the computational validation and endorsements from leading computational neuroscientists, provides a strong foundation for a HEAL Initiative UG3/UH3 application. The work directly addresses the NIH HEAL Initiative's goal of developing non-addictive analgesics by targeting the mechanism of addiction itself rather than the opioid receptor. BIOGRAPHICAL SKETCH Eniola Ayodele Olutogun received 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. The Pharmacy Council of Nigeria issued full licensure in 2022. From 2014 to 2021, the curriculum included pharmaceutical chemistry, pharmacology, pharmacognosy, and clinical pharmacy, with a final-year project on NMDA receptor docking of insulin analogues conducted at the Centre for Drug Discovery, Development and Production at the University of Ibadan. From 2021 to 2023, employment as a Research Assistant at the Centre for Drug Discovery, Development and Production involved molecular docking studies of NMDA receptor ligands and insulin analogues using AutoDock Vina and GROMACS. From 2023 to 2024, a Bioinformatics Researcher position at the Ghanaian-South African Antimicrobial Resistance Genomics Surveillance Network involved building an AMR surveillance pipeline using Nextflow and SLURM on HPC infrastructure. From January to March 2026, a Clinical Pharmacist role at Ramset Pharmacy in Lagos involved patient medication management and adverse drug reaction reporting. Since March 2026, the National Product Manager role at Synthcare in Lagos involves overseeing pharmaceutical product strategy across Nigeria. Independent research from 2025 to 2026 produced the Conjunctive Consolidation Threshold model, documented in three sole-authored preprints. The foundational paper on OSF (10.17605/OSF.IO/KG7B5) describes the tripartite framework. The formal mathematical specification on OSF (10.17605/OSF.IO/EMY4U) provides the ODE/RK45 dynamics. The Bayesian population dynamics and clinical trial architecture on Zenodo (10.5281/zenodo.20492472) specifies the adaptive trial design. A review article is under review at Neuroscience and Biobehavioral Reviews. A co-authored paper is under review at Alcohol (Elsevier). Three computational platforms were built. IMPRINT screens compounds for addiction liability using the CCT framework. TOPOLOGIX applies topological data analysis with persistent homology and bipartite simplicial complexes to drug-protein interaction surfaces, with a hERG cardiotoxicity MVP completed. GATE evaluates BCI neural-stimulation safety under an Apache 2.0 license. All code is available on GitHub at github.com/AmunRaPtah. Endorsements for arXiv submission were provided by Samuel Gershman at Harvard. Additional collaborators include Kent Berridge at the University of Michigan, Nathaniel Daw at Princeton University, and Marcelo Mattar at New York University. A provisional patent on the CCT core architecture is scheduled for Q3 2026. Computational skills include Python with scipy, numpy, PyMC for Bayesian MCMC, and pandas; R for statistical analysis; topological data analysis with Ripser and Gudhi; neural simulation with NEURON and Brian2; molecular modelling with AlphaFold, RDKit, ADMET and QSAR tools, GROMACS, and AutoDock; workflow management with Nextflow and SLURM on HPC; and database management with Supabase, Postgres, JavaScript, and Node.js. Applications are in progress for an MSc in Computational Neuroscience at the Medical University of Graz and the University of Graz in Austria, with an October 2026 start date. Current affiliation is as an independent researcher based in Lagos, Nigeria, operating under the ZYCO research organisation. BUDGET NARRATIVE The UG3 phase budget is structured around three specific aims over two years. Personnel costs include salary support for the PI at the US academic institution at 1.0 calendar months per year, and consultant fees for the independent researcher at 20,000 USD per year for 12 months of computational work. Total personnel: 55,000 USD per year. Equipment and supplies for Aim 1 include computational infrastructure: a dedicated GPU workstation for molecular docking and TDA at 8,000 USD, cloud computing credits for HPC access at 5,000 USD per year, and software licenses for RDKit and Schrödinger at 3,000 USD per year. Total equipment: 16,000 USD in year one, 8,000 USD in year two. Supplies for Aim 2 include primary neuronal culture reagents for rat VTA, nucleus accumbens, and hippocampal preparations at 12,000 USD per year, calcium imaging dyes and reagents at 6,000 USD per year, and patch-clamp electrophysiology supplies at 8,000 USD per year. Total supplies: 26,000 USD per year. Supplies for Aim 3 include rodent purchase and housing for 60 animals per year at 15,000 USD, surgical supplies for postoperative pain model implantation at 5,000 USD per year, and behavioural testing equipment and software at 7,000 USD per year. Total animal costs: 27,000 USD per year. Other expenses include publication fees for open-access journals at 3,000 USD per year, travel to one scientific conference per year at 4,000 USD, and patent maintenance fees for the provisional patent at 2,000 USD per year. Total other: 9,000 USD per year. Total direct costs for the UG3 phase: 117,000 USD in year one, 107,000 USD in year two. Indirect costs at the US academic institution standard rate of 50 percent: 58,500 USD in year one, 53,500 USD in year two. Total UG3 budget: 175,500 USD in year one, 160,500 USD in year two. Total UG3 phase: 336,000 USD. The UH3 phase budget will be developed based on UG3 results and will include clinical trial costs for first-in-human studies. The Bayesian adaptive trial design specified in the Zenodo preprint will inform the sample size and duration, with estimated costs of 500,000 to 1,000,000 USD per year for a two-year phase 1 trial. CHECKLIST - [ ] Verify that the PI applicant (US-based academic neuroscientist) has agreed to submit as the lead institution - [ ] Obtain letter of support from Kent Berridge at University of Michigan confirming willingness to serve as PI - [ ] Obtain letters of endorsement from Samuel Gershman, Nathaniel Daw, and Marcelo Mattar - [ ] Complete the NIH SF424 Research and Related application form - [ ] Upload the Research Strategy document (12-page limit) - [ ] Upload the Biographical Sketch for the PI in NIH format - [ ] Upload the Biographical Sketch for the independent researcher in NIH format - [ ] Upload the Budget and Budget Justification using the NIH PHS 398 form - [ ] Upload the Facilities and Other Resources document describing the US academic institution's animal facilities and core labs - [ ] Upload the Provisional Patent application documentation as an appendix - [ ] Upload the three preprints as PDF attachments - [ ] Verify that the independent researcher's ORCID (0009-0001-9272-6735) is linked to the eRA Commons profile - [ ] Confirm that the independent researcher is eligible for foreign collaborator status on an NIH grant - [ ] Submit through grants.gov by 02/20/2029 EDITOR NOTES - The independent researcher is not a US citizen or permanent resident and is not affiliated with a US institution. The NIH UG3/UH3 mechanism requires a US-based institution as the applicant. The application must be submitted by the US academic PI's institution, with the independent researcher listed as a consultant or collaborator. Verify that the US institution's sponsored projects office will accept this arrangement. - The provisional patent is scheduled for Q3 2026. Confirm the exact filing date and patent application number before submission. If the patent has not been filed by the submission deadline, remove the patent claim from the application and replace with a statement about patent strategy. - The independent researcher is not yet enrolled in an MSc programme. The application should not imply current graduate student status. The biographical sketch should clearly state the independent researcher status and the planned October 2026 MSc start date. - The budget narrative assumes a 50 percent indirect cost rate. Verify the actual negotiated rate at the US academic institution. Adjust the budget accordingly. - The three preprints are sole-authored. The NIH requires documentation of the independent researcher's ability to perform the proposed work without a formal academic appointment. The endorsements from Berridge, Gershman, Daw, and Mattar serve this purpose, but additional documentation of the computational platforms and their validation may strengthen the application.