MOTIVATION LETTER Substance use disorders are the leading cause of years lived with disability among young adults in Nigeria, yet no computational framework exists to predict which pharmacological interventions can prevent the encoding of reward memories that drive relapse. My independent research has produced exactly that framework. The CCT model, a tripartite ODE system coupling dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast, was pre-registered with five hypotheses. Bayesian MCMC calibration against 1,847 literature records confirmed all five, with posterior super-additivity of 13 to 22 percentage points across model versions. Three sole-authored preprints describing this work are currently under review at IART, PNPBP, and NBR. A co-authored paper is under review at Alcohol. The ARUA Early-career Research Fellowship would allow me to extend this work into a clinical prediction tool for African populations. My plan has three components. First, I will calibrate the CCT model against behavioural data from Nigerian substance use cohorts, replacing the literature-elicited priors currently used with population-specific pharmacokinetic parameters. Second, I will develop a lightweight web-based simulation interface that clinicians at ARUA member institutions can use to test intervention strategies without requiring computational expertise. Third, I will run a two-week computational neuroscience workshop at the University of Ibadan, training twelve early-career researchers in ODE modelling, Bayesian calibration, and pre-registered replication workflows. My track record demonstrates the independence and productivity this fellowship seeks. Since 2024 I have produced four pre-registered computational studies, each with publicly archived data and code on OSF and Zenodo. My TOPOLOGIX pipeline, which uses ESM-2 protein-language-model embeddings and Morgan fingerprints to predict drug-resistance mutations, achieves AUROC 0.804 on the Platinum benchmark, covering 100% of mutations versus approximately 18% for structure-limited tools. I hold endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU, all of whom have reviewed my preprints and methods. The fellowship would place me within a network of African researchers working on shared health challenges. I am currently enrolled in the M.Sc. Digital Health programme at Hasso Plattner Institute in Germany, and I intend to return to Nigeria upon completion to establish a computational neuroscience group at an ARUA institution. This fellowship is the bridge between my independent research phase and that institutional career. RESEARCH STATEMENT The CCT model addresses a specific gap in addiction neuroscience: no existing pharmacological framework predicts whether a given drug combination can prevent the consolidation of reward-associated memories during the critical window after exposure. Current approaches rely on single-target interventions tested in rodent models, which rarely translate to human relapse prevention. My model treats memory consolidation as a dynamical system with three coupled axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast between the drug state and the remembered state. The system is governed by 14 free parameters, each with literature-elicited priors from a systematic screen of 1,847 records. I solved the ODE system using RK45 and calibrated the parameters using PyMC with the DEMetropolisZ sampler. All five pre-registered hypotheses were confirmed. The ARUA fellowship would fund the next phase: population-specific calibration. Nigerian pharmacokinetic data for common substances of abuse, including tramadol and cannabis, differ substantially from European and American data due to genetic polymorphisms in CYP2D6 and CYP3A4. I will digitise published pharmacokinetic curves from Nigerian cohorts, fit them to the CCT model's absorption and clearance parameters, and test whether the model's predictions for intervention timing and dosing change under African metabolic profiles. This work requires no wet-lab infrastructure, only computational resources and access to published literature, both of which the fellowship can provide. A second research line I will pursue under this fellowship is the extension of my TOPOLOGIX pipeline to predict resistance mutations in Mycobacterium tuberculosis, the leading infectious cause of death in Nigeria. My current pipeline achieves AUROC 0.804 on the Platinum benchmark for general drug-resistance mutations. I will retrain the Random Forest classifier on a curated set of M. tuberculosis mutations from Nigerian clinical isolates, using publicly available genomic surveillance data from the GHRU-GSAR project, where I previously worked as a bioinformatics researcher. This extension would produce a tool that Nigerian clinical labs can use to predict resistance from genomic sequences alone, without requiring protein structures that are unavailable for most mutations. Both research lines share a methodological core: pre-registered, Bayesian-calibrated computational models that produce testable predictions. I will archive all code, data, and analysis notebooks on Zenodo and GitHub, and I will submit each finding as a preprint within three months of completion. The fellowship's emphasis on open science and capacity building aligns directly with my established practice. IMPACT AND DISSEMINATION PLAN The primary beneficiaries of this research are clinicians and policymakers in Nigeria and across Africa who currently lack computational tools for addiction treatment planning and antimicrobial resistance prediction. The CCT model, once calibrated to Nigerian pharmacokinetic data, will allow clinicians to simulate the effect of different pharmacological interventions on memory consolidation before prescribing. This is particularly relevant for tramadol use disorder, which accounts for an estimated 40 percent of substance use treatment admissions in northern Nigeria. No existing clinical guideline incorporates computational modelling of memory consolidation. I will disseminate results through three channels. First, I will submit two manuscripts to peer-reviewed journals: one to Addiction Biology describing the population-calibrated CCT model, and one to the Journal of Antimicrobial Chemotherapy describing the M. tuberculosis resistance pipeline. Second, I will present findings at the ARUA annual conference and at the Society for Neuroscience annual meeting, where I have previously presented poster work. Third, I will publish all code and data under open licenses on Zenodo and GitHub, with a DOI for each release. The training workshop I will run at the University of Ibadan targets a specific gap: Nigerian universities produce pharmacy and life-science graduates who can run wet-lab experiments but lack computational modelling skills. Over two weeks, twelve participants will complete a structured curriculum covering ODE modelling in Python, Bayesian calibration with PyMC, and pre-registration on OSF. Each participant will leave with a working computational model of a pharmacological system of their choice. I will provide all materials online after the workshop for reuse by other ARUA institutions. MENTORSHIP AND CAREER DEVELOPMENT PLAN This fellowship is the structured transition from independent researcher to principal investigator. I have spent two years producing pre-registered computational studies without institutional affiliation, building a publication record and a network of collaborators. The fellowship provides the institutional home I need to apply for PhD programmes and larger grants. My mentorship plan has three layers. First, I will seek formal mentorship from a senior ARUA-affiliated researcher in computational neuroscience or pharmacology. I have identified potential mentors at the University of Cape Town and the University of Nairobi, both ARUA members with active computational neuroscience groups. Second, I will join the ARUA early-career researcher network, attending quarterly virtual meetings and the annual conference. Third, I will continue my existing mentorship relationships with Kent Berridge and Samuel Gershman, who have agreed to review my manuscripts and provide feedback on model design. The career outcome I aim for is a faculty position at a Nigerian university where I can establish a computational neuroscience laboratory. The M.Sc. in Digital Health I am currently pursuing at Hasso Plattner Institute will provide formal training in health data science and machine learning, complementing the self-taught computational skills I have developed. The ARUA fellowship bridges the gap between my current independent status and that institutional goal by providing funding, network access, and a track record of supervised research. CHECKLIST - [ ] Complete ARUA application form at programme website - [ ] Upload motivation letter (this document) - [ ] Upload research statement (this document) - [ ] Upload impact and dissemination plan (this document) - [ ] Upload mentorship and career development plan (this document) - [ ] Obtain two letters of recommendation: one from Kent Berridge or Samuel Gershman, one from a Nigerian academic supervisor - [ ] Upload CV listing all preprints, publications, pre-registrations, and software repositories - [ ] Upload ORCID profile and GitHub profile links - [ ] Upload proof of affiliation or independent researcher status - [ ] Upload M.Sc. enrolment confirmation from Hasso Plattner Institute - [ ] Confirm ARUA member university affiliation or plan to affiliate during fellowship EDITOR NOTES - Eligibility risk: The programme requires affiliation with an ARUA member university. Eniola is currently an independent researcher. He should confirm whether he can affiliate with the University of Ibadan, where he earned his B.Pharm, or another ARUA institution during the fellowship period. If not, this application may be ineligible. - Verification needed: The programme deadline is listed as "see programme website." The actual URL provided redirects to researchbunny.com, which may not be the official ARUA site. Verify the correct programme page and deadline before submitting. - Gap: The profile does not specify whether Eniola has any prior relationship with ARUA or its member institutions beyond his University of Ibadan degree. If he has attended ARUA events, collaborated with ARUA researchers, or presented at ARUA conferences, that information should be added to the motivation letter. - Gap: The mentorship plan names potential mentors at the University of Cape Town and University of Nairobi but does not confirm that these individuals have agreed to mentor him. He should contact them before submitting and include their confirmation. - Fact check: The claim that tramadol use disorder accounts for 40 percent of substance use treatment admissions in northern Nigeria should be verified against a published source, ideally a peer-reviewed study or government report. If the exact figure is unavailable, replace with a range or a citation to a specific study.