MOTIVATION LETTER The Mozilla Foundation’s 2026 Fellowship programme asks for people who build technology that serves the public interest. I am a computational researcher and pharmacist from Nigeria, and my independent research programme has produced five pre-registered studies, three sole-authored preprints, and one co-authored paper under review at Alcohol (Elsevier). Each project was designed to answer a question that matters for human welfare, not for a publication count. My CCT model addresses addiction by simulating the pharmacological conditions under which reward-memory encoding can be prevented. The model uses a three-axis ODE system with Bayesian MCMC calibration on 14 free parameters drawn from a literature screen of 1,847 records. All five pre-registered hypotheses were confirmed, with posterior super-additivity of 13 to 22 percentage points across model versions. This work is open science: the code, data, and preprints are on OSF and Zenodo. It is also directly relevant to the opioid crisis and substance-use disorders that disproportionately affect low- and middle-income countries, including Nigeria. My cardiotoxicity topology study tested whether bipartite persistent homology could predict hERG cardiotoxicity from protein-ligand interface geometry. The pre-registered, powered replication found that topological features do not beat a plain descriptor baseline (AUROC 0.8426 versus 0.8782). This result settles a comparison the published literature had never actually run. I reported the negative finding directly rather than reframing it. That is how open, ethical research should work. My current project, TOPOLOGIX, uses ESM-2 protein-language-model delta-embeddings and Morgan fingerprints with a Random Forest classifier to predict drug-resistance mutations from sequence alone. It achieves AUROC 0.804 on the Platinum benchmark, covering 100 percent of mutations versus roughly 18 percent for structure-limited tools. This matters for antimicrobial resistance, a global health crisis that hits Africa hardest. I am enrolled in the M.Sc. Digital Health at the Hasso Plattner Institute in Germany, starting winter 2026. I am 29 years old, Nigerian, and an independent researcher with no institutional faculty position. My collaborators include Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. I have endorsements from Gershman for arXiv submissions. The Mozilla Fellowship would allow me to continue this work in the open, to build a public-interest technology practice rooted in African realities, and to become a trusted voice on how computational methods can serve health equity rather than extractive models. I meet the Global Majority emphasis. I work in the open. I produce concrete, falsifiable results. I am ready. RESEARCH STATEMENT My research programme spans addiction neuroscience, protein-machine-learning, and dynamical-systems methods. The unifying thread is that I build computational models that make testable predictions about biological systems, then test those predictions with pre-registered experiments. I do not chase benchmarks. I chase mechanistic understanding. The CCT model is the centrepiece of my addiction work. It couples three axes: dopaminergic reward-prediction error, NMDAR-dependent long-term potentiation, and affective contrast. The ODE system is solved with RK45 and calibrated with PyMC’s DEMetropolisZ sampler. The priors come from a systematic screen of 1,847 records from the addiction neuroscience literature. All five pre-registered hypotheses were confirmed. The model predicts that conjunctive consolidation threshold crossing requires simultaneous activation of all three axes, which suggests a pharmacological strategy for preventing reward-memory encoding without ablating reward processing entirely. A co-authored paper is under review at Alcohol. The cardiotoxicity topology study was a direct test of a claim in the computational chemistry literature: that topological data analysis of protein-ligand interfaces can predict hERG cardiotoxicity. I pre-registered a powered replication, built the opposition-distance metric, ran Ripser and GUDHI, and found that topological features yield AUROC 0.8426 against a plain Morgan fingerprint baseline of 0.8782. The difference is not significant. The paper reports this honestly. The field needs more such studies. TOPOLOGIX is my current main project. It uses ESM-2 embeddings from the protein language model, Morgan fingerprints for the drug, and a Random Forest classifier. On the Platinum benchmark of 553 mutations, it achieves AUROC 0.804 with a standard deviation of 0.025. On SKEMPI 2.0, it achieves 0.634. It beats mCSM-lig, which scores around 0.70, while covering every mutation in the benchmark. Structure-based tools cover only 18 percent because they require a crystal structure. TOPOLOGIX requires only sequence. The neurocascade engine couples pharmacokinetics to receptor binding to Wilson-Cowan circuit dynamics to behavioural readout. Three receptor systems are calibrated: mu-opioid, D2 dopamine, and GABA-A. All 62 tests pass. The circuit-layer parameters are labelled illustrative pending real behavioural data. The ergofluids project applies Koopman-operator methods with a Mori-Zwanzig memory kernel to model drug transport through tumour tissue. The synthetic-data gates passed. The first real-data gate, tested against digitized published figures, did not meet its primary pre-registered criterion. I reported this directly. I have no faculty position. I am an independent researcher. My work is on GitHub, on Zenodo, on OSF. I build everything in the open. SHORT ESSAY: VALUES ALIGNMENT WITH MOZILLA Mozilla’s mission is an internet that puts people first. My research does the same for computational pharmacology. The CCT model is built entirely with open-source tools: Python, PyMC, scipy, OSF for preprints, Zenodo for data. Every line of code is on GitHub. Every result, including negative ones, is published. The cardiotoxicity topology study is a case in point: I ran a pre-registered replication, found that the claimed method does not work, and published that finding. That is public-interest science. My Nigerian background shapes my values. Antimicrobial resistance kills more people in sub-Saharan Africa than in any other region. Drug-resistance prediction tools that require crystal structures are useless in settings where sequencing is available but structural biology is not. TOPOLOGIX works from sequence alone. It is designed for the infrastructure realities of LMICs. I also self-host local LLM serving with llama.cpp and maintain four independent DuckDB-based RAG pipelines across life sciences, tech security, and social science domains. I do not rely on proprietary APIs. I build infrastructure that can run offline, on modest hardware, in any country. That is technological sovereignty. Mozilla’s emphasis on open working and ethical technology matches my practice. I do not file patents. I do not seek venture funding. I produce methods, code, and papers that anyone can use, modify, and build upon. The fellowship would allow me to scale this practice and to advocate for open, public-interest computational research in African health contexts. SHORT ESSAY: LEADERSHIP AND INDEPENDENCE I have built a research programme from scratch with no institutional support. I designed the CCT model, wrote the ODE solver, conducted the literature screen of 1,847 records, ran the Bayesian calibration, and wrote three sole-authored preprints. I did this while working as a clinical pharmacist and later as a national product manager. I did not have a PhD supervisor, a lab, or a grant. The cardiotoxicity topology study required me to learn persistent homology from first principles, implement the opposition-distance metric, run Ripser and GUDHI, and design a pre-registered replication protocol. I did this alone. The result is a paper that corrects a claim in the literature. I have secured endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. These are researchers whose work I have built upon. They know my work because I sent them my preprints and code. I am enrolled in the M.Sc. Digital Health at Hasso Plattner Institute, one of Germany’s top computer science institutions. I will begin in winter 2026. I am 29 years old. I have a B.Pharm from the University of Ibadan with a German-equivalent grade of 1.9. I am licensed by the Pharmacists Council of Nigeria. I do not wait for permission. I identify a problem, learn the methods, build the tool, test it, and publish the result. That is the kind of leadership Mozilla’s fellowship supports. CHECKLIST - [ ] Complete Mozilla Foundation 2026 Fellowship nomination form at the programme URL - [ ] Upload motivation letter as PDF - [ ] Upload research statement as PDF - [ ] Upload short essay on values alignment as PDF - [ ] Upload short essay on leadership and independence as PDF - [ ] Provide ORCID: 0009-0001-9272-6735 - [ ] Provide GitHub: github.com/AmunRaPtah - [ ] Provide personal site: zyco.org - [ ] Confirm eligibility for early-career, pre-PhD, LMIC-track, independent researcher category - [ ] Verify deadline on programme website - [ ] Prepare two reference contacts: Samuel Gershman (Harvard) and Kent Berridge (Michigan) - [ ] Confirm that the fellowship accepts pre-print publications as research output EDITOR NOTES - Eligibility risk: The programme URL points to a third-party aggregator (mediarightsagenda.org), not the official Mozilla Foundation page. Verify the actual fellowship page on mozilla.org and confirm deadlines, award amounts, and required materials before submitting. - Fact to verify: Confirm that the co-authored paper in Alcohol (Elsevier) is still under review and that the journal name is correct. If accepted by submission time, update the status. - Gap to fill: The profile does not include a specific project proposal for the fellowship period. Mozilla typically expects fellows to propose a concrete project or advocacy initiative. Eniola should draft a one-paragraph project description: for example, building an open-source drug-resistance prediction platform for African public-health labs, or creating a curriculum on open-source computational pharmacology for Nigerian universities. - Gap to fill: The profile does not mention any policy or advocacy experience. Mozilla fellows often engage with tech policy, digital rights, or civil society. Eniola should prepare a brief statement on how he would engage with policy audiences, even if his primary work is technical. - Fact to verify: Confirm that the M.Sc. Digital Health at HPI/Potsdam has confirmed admission for winter 2026/27. If the admission is conditional or pending, state that clearly.