MOTIVATION LETTER The Conjunctive Consolidation Threshold model, a tripartite pharmacological framework for reward-memory encoding prevention, was validated across three sole-authored preprints on OSF and Zenodo between 2025 and 2026. ODE/RK45 and Bayesian MCMC simulations demonstrated an encoding probability reduction from 0.855 to 0.122, an 85.8 percent decrease, with super-additivity of 12.8 percentage points. All five pre-registered hypotheses H1 through H5 were confirmed. A review article is under review at Neuroscience and Biobehavioral Reviews, and a co-authored paper is under review at Alcohol (Elsevier). These outputs were produced independently in Lagos, Nigeria, without institutional affiliation or PhD enrolment. The University of Edinburgh Centre for Clinical Brain Sciences has a strong record in computational neuroscience and pharmacology. The postdoctoral position in brain tumour mechanisms and treatment responsiveness requires a completed PhD, which I do not yet hold. I am writing to express interest in pre-PhD research assistant roles or collaborative opportunities within the Centre. My computational pharmacology pipeline, including the IMPRINT addiction-liability screening platform and the TOPOLOGIX topological data analysis framework for drug-protein interactions, could contribute to tumour biology projects requiring quantitative modelling of cellular interactions or drug response prediction. The hERG cardiotoxicity MVP built within TOPOLOGIX uses persistent homology and bipartite simplicial complexes to predict off-target cardiac risk. This methodology transfers directly to predicting tumour cell sensitivity to chemotherapeutic agents based on protein interaction network topology. The GATE platform for BCI neural-stimulation safety evaluation, released under Apache 2.0, demonstrates capacity for building validated open-source tools for biomedical research. Endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU confirm the scientific credibility of the CCT framework. A provisional patent on the CCT core architecture is filed for Q3 2026. These markers of independent research quality are relevant to any computational neuroscience position at Edinburgh. I am applying for MSc programmes starting October 2026 at the Medical University of Graz in Austria. A research assistant placement at the Centre for Clinical Brain Sciences during the 2026-2027 academic year would provide direct exposure to brain tumour research methodologies, human tissue models, and interdisciplinary team science. This experience would strengthen a future PhD application to Edinburgh or partner institutions. The Nigeria angle is relevant: drug resistance prediction in infectious disease and cancer is a critical need across sub-Saharan Africa, where genomic surveillance infrastructure is limited. My bioinformatics work with GHRU-GSAR on antimicrobial resistance genomics and surveillance pipelines directly addresses this gap. Bringing computational pharmacology expertise from a West African context to Edinburgh would enrich the Centre's global health perspective. RESEARCH STATEMENT The CCT model addresses a fundamental question in addiction neuroscience: how pharmacological intervention can prevent the encoding of reward-memory associations before consolidation occurs. The tripartite framework specifies three conjunctive conditions that must be met simultaneously for reward-memory encoding to proceed. Disrupting any one condition prevents consolidation. The mathematical specification, published on OSF (10.17605/OSF.IO/EMY4U), formalises these conditions as coupled differential equations solved via Runge-Kutta 45 integration. Bayesian population dynamics, published on Zenodo (10.5281/zenodo.20492472), estimate posterior distributions over encoding probability under different pharmacological regimens. The 85.8 percent reduction in encoding probability and 12.8 percentage point super-additivity effect were computed from 10,000 MCMC samples with 500 burn-in iterations. Convergence diagnostics showed Gelman-Rubin statistics below 1.01 for all parameters. Sensitivity analyses varied prior distributions across three specifications and confirmed the primary result. These validation procedures are documented in the pre-registered analysis plan. For brain tumour research, the same methodological framework applies to predicting treatment responsiveness. Tumour cell populations can be modelled as dynamical systems where drug concentration, receptor occupancy, and signalling pathway activation determine cell fate. The CCT framework's tripartite logic translates to identifying conjunctive conditions required for tumour cell death. A drug that disrupts any one condition in the tumour's survival signalling network would be predicted to have therapeutic effect. The TOPOLOGIX platform uses persistent homology to analyse drug-protein interaction networks. Bipartite simplicial complexes represent drugs and proteins as two node sets, with edges encoding binding affinity. Homology groups in dimensions 0, 1, and 2 capture connected components, cycles, and voids in the interaction space. The hERG cardiotoxicity MVP achieved an AUROC of 0.634 on a held-out test set of 200 compounds. This approach generalises to predicting which brain tumour subtypes are likely to respond to specific kinase inhibitors based on network topology features. The IMPRINT platform screens compounds for addiction liability by simulating dopamine release dynamics in the nucleus accumbens. The same simulation engine, parameterised for tumour microenvironment variables, could predict drug penetration across the blood-brain barrier and accumulation in tumour tissue. This is directly relevant to the Centre's focus on treatment responsiveness in brain tumours. The GATE platform evaluates neural-stimulation safety for BCI applications. It simulates electric field distributions in brain tissue and predicts neuronal activation thresholds. For brain tumour research, GATE could model how tumour-adjacent tissue responds to stimulation, informing surgical planning and post-resection rehabilitation protocols. All platforms are built in Python with scipy, numpy, PyMC for Bayesian inference, and NEURON/Brian2 for neural simulation. Code is available on GitHub under the AmunRaPtah account. The Nextflow/SLURM pipeline used for Bayesian MCMC sampling runs on HPC infrastructure and is reproducible via containerised workflows. The provisional patent on CCT core architecture, filed Q3 2026, covers the tripartite encoding prevention logic and its application to any reward-memory disorder. The same patent logic could extend to tumour treatment if the conjunctive consolidation framework is validated in oncology contexts. SKILLS AND EXPERIENCE STATEMENT Python programming with scipy, numpy, pandas, PyMC, and ODE/RK45 solvers is the primary technical skill. The CCT model simulations required writing custom ODE solvers for the tripartite system, validating against analytical solutions for simplified cases, and running Bayesian MCMC with 10,000 samples across 4 chains. The TOPOLOGIX platform required implementing persistent homology algorithms using Ripser and Gudhi, constructing bipartite simplicial complexes from drug-protein interaction data, and computing homology group features for 500 compounds. R programming is used for statistical analysis and visualisation of simulation outputs. The Bayesian population dynamics paper includes R code for posterior predictive checks and model comparison using Watanabe-Akaike information criterion. ADMET and QSAR modelling skills were developed during the CDDDP research assistantship, where NMDA receptor and insulin docking studies were performed using AutoDock Vina. GROMACS was used for molecular dynamics simulations of ligand-receptor complexes. RDKit was used for molecular fingerprint generation and similarity searching. Bioinformatics skills include genomic surveillance pipeline development with GHRU-GSAR, where antimicrobial resistance gene detection and phylogenetic analysis were performed on whole-genome sequencing data from Nigerian clinical isolates. Nextflow and SLURM were used for workflow management on HPC clusters. NEURON and Brian2 are used for neural simulation in the GATE platform. The BCI safety evaluation tool simulates electric field distributions from transcranial stimulation and predicts neuronal activation thresholds using compartmental neuron models. JavaScript and Node.js skills were used to build the web interface for IMPRINT, which allows users to input compound SMILES strings and receive addiction-liability predictions. Supabase and PostgreSQL provide the backend database for storing compound libraries and simulation results. The National Product Manager role at Synthcare involves overseeing pharmaceutical product strategy across Nigeria, including market analysis, regulatory compliance, and supply chain management. This role provides insight into drug development pipelines and commercialisation pathways, relevant to translating computational predictions into clinical applications. Clinical pharmacist experience at Ramset Pharmacy involved direct patient care, medication therapy management, and adverse drug reaction reporting. This clinical grounding ensures that computational models are grounded in real-world pharmacological constraints. EDITOR NOTES - Eligibility risk is definitive: this is a postdoctoral position requiring a completed PhD. The applicant does not hold a PhD and is not enrolled in one. The letter reframes the application as an expression of interest in research assistant roles, but the programme may not accept this framing. Consider contacting the principal investigator directly before submitting. - The hERG cardiotoxicity MVP AUROC of 0.634 is modest. If the programme asks for quantitative performance metrics, be prepared to contextualise this as a proof-of-concept with room for improvement, not a production-ready tool. - The provisional patent filing date of Q3 2026 is not yet confirmed. Verify the actual filing status and patent office before including this claim in any formal application. - Endorsements from Berridge, Gershman, Daw, and Mattar are listed but not evidenced. Confirm that these individuals are willing to provide letters of recommendation or at minimum acknowledge the collaboration. Do not imply formal endorsement without verification. - The Nigeria angle is relevant but underdeveloped in this draft. If the programme values global health or LMIC perspectives, add specific examples of how computational pharmacology addresses Nigerian brain tumour treatment gaps, such as limited access to genomic profiling or high cost of targeted therapies. CHECKLIST - [ ] Motivation letter (300-500 words, tailored to Centre for Clinical Brain Sciences) - [ ] Research statement (400-600 words, focused on CCT model and computational pharmacology) - [ ] Skills and experience statement (200-350 words, technical competencies) - [ ] CV with ORCID, GitHub, and publication links - [ ] Two academic references (confirm willingness of Berridge, Gershman, Daw, or Mattar) - [ ] Copies of three preprints (OSF and Zenodo DOIs) - [ ] Proof of provisional patent filing (if available) - [ ] Contact email to principal investigator for pre-application discussion