MOTIVATION LETTER The National Institutes of Health ancient DNA for Modern Genomics Coordination Center addresses a question central to my research: how did the human reward system evolve, and what does its deep history reveal about modern vulnerability to addiction? My independent work developing the Conjunctive Consolidation Threshold model has demonstrated that reward-memory encoding can be pharmacologically interrupted with 85.8 percent reduction in encoding probability and super-additivity of 12.8 percentage points. That model, however, rests on assumptions about the neurobiological architecture of reward that remain untested across evolutionary timescales. aDMG offers the framework to test them. I am a Nigerian pharmacist and independent computational neuroscientist. My B.Pharm from the University of Ibadan, German equivalent 1.9, grounds me in pharmacology. My preprints on OSF and Zenodo formalize the CCT model mathematically and validate it through ODE/RK45 simulation and Bayesian MCMC. My platforms IMPRINT and TOPOLOGIX apply topological data analysis to drug-protein interaction networks, with persistent homology and bipartite simplicial complexes. These tools were built to predict addiction liability from molecular structure. aDMG can extend that prediction backward in time: ancient alleles in dopamine receptor genes, opioid receptor variants, and transcription factor binding sites in reward circuitry can be analyzed through the same computational lens. Nigeria sits at a crossroads of genomic diversity underrepresented in paleogenomic datasets. My position in Lagos gives me access to clinical populations with high addiction burden and to genetic variation that may carry ancient signatures of selection on reward pathways. I propose to contribute to aDMG by developing computational pipelines that map ancient DNA variants onto modern drug-target interaction networks, using TOPOLOGIX to detect topological signatures of conserved reward-circuit architecture. The CCT model provides a formal framework for predicting how ancient variants modulate encoding probability in modern contexts. Endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU confirm the relevance of this approach. My provisional patent on CCT core architecture, filed Q3 2026, protects the translational pathway. aDMG can connect these computational tools to the paleogenomic community and to the ancient DNA datasets that will validate or falsify the evolutionary hypotheses embedded in my model. I seek to join aDMG as an independent researcher contributing computational pharmacology and addiction neuroscience expertise to the Coordination Center. My goal is to build the bridge between paleogenomics and addiction medicine, using ancient DNA to understand why some brains encode reward memories that lead to compulsive drug seeking. RESEARCH STATEMENT The Conjunctive Consolidation Threshold model proposes that reward-memory encoding requires simultaneous activation of three distinct pharmacological subsystems: dopaminergic salience signaling, glutamatergic plasticity at corticostriatal synapses, and opioidergic hedonic gating. Interruption of any two subsystems produces super-additive reduction in encoding probability. My pre-registered hypotheses H1 through H5 were confirmed through ODE/RK45 simulation and Bayesian MCMC, yielding encoding probability reduction from 0.855 to 0.122. The model is under review at Neuroscience and Biobehavioral Reviews. This framework assumes that the tripartite architecture is conserved across human populations and across evolutionary time. aDMG can test that assumption directly. Ancient DNA from African populations, including those from Nigeria and West Africa, can reveal whether selection acted on genes encoding dopamine D2 receptors, mu-opioid receptors, or NMDA receptor subunits in ways that altered reward encoding thresholds. My computational platforms are designed for this analysis. TOPOLOGIX applies persistent homology to drug-protein interaction networks. Bipartite simplicial complexes capture multi-way interactions between compounds and targets. The hERG cardiotoxicity MVP demonstrated that topological signatures predict off-target binding with higher resolution than sequence-based methods. For aDMG, I propose to construct topological networks of ancient protein variants inferred from ancient DNA, then compare their interaction topologies with modern variants. Differences in persistent homology features may indicate functional divergence in reward circuitry. IMPRINT screens compounds for addiction liability by simulating their effect on CCT parameters. The platform uses PyMC for Bayesian inference of encoding probability from molecular structure. Applied to ancient DNA, IMPRINT can predict how ancient alleles would modulate addiction liability in modern contexts. This creates a direct bridge from paleogenomic data to clinical pharmacology. The Bayesian population dynamics model specified in my third preprint (Zenodo 10.5281/zenodo.20492472) provides a framework for simulating how reward encoding thresholds evolve under selection pressure. I propose to extend this model to incorporate ancient allele frequency trajectories, using aDMG datasets to parameterize the selection coefficients acting on reward-related genes. The model predicts that populations with higher frequencies of alleles that lower encoding thresholds should show higher addiction prevalence. This hypothesis can be tested against modern epidemiological data from Nigeria and other African populations. My collaborators provide domain expertise. Kent Berridge studies hedonic hotspots in the nucleus accumbens. Samuel Gershman models reinforcement learning and Bayesian inference in the brain. Nathaniel Daw works on dopamine and decision-making. Marcelo Mattar studies memory consolidation and replay. Each has endorsed my CCT framework. For aDMG, I would seek collaboration with paleogenomicists to access ancient DNA datasets and with population geneticists to model selection. The provisional patent on CCT core architecture, filed Q3 2026, covers the method for predicting addiction liability from genetic and pharmacological data. This intellectual property can be licensed to aDMG for non-commercial research use, ensuring that any tools developed under this collaboration remain open to the Coordination Center. Nigeria offers unique advantages for this work. The country has high genetic diversity, high addiction burden with limited treatment access, and growing computational infrastructure. My position as National Product Manager at Synthcare and my prior work with GHRU-GSAR on antimicrobial resistance genomics demonstrate my ability to execute genomic surveillance pipelines in resource-limited settings. The same skills apply to ancient DNA analysis. I propose three specific deliverables for aDMG: first, a topological analysis pipeline for ancient protein variants, implemented in Python with Ripser and Gudhi, deposited on GitHub under Apache 2.0 license; second, a Bayesian model of reward encoding threshold evolution, parameterized with aDMG allele frequency data; third, a report mapping ancient alleles onto modern addiction liability predictions, using IMPRINT. All code and data will be open access. BIOGRAPHICAL SKETCH Eniola Ayodele Olutogun. Independent researcher based in Lagos, Nigeria. B.Pharm, University of Ibadan, 2021, CGPA 5.1 of 7.0, German equivalent 1.9. Licensed pharmacist, Pharmacists Council of Nigeria. Research focuses on computational pharmacology of addiction. Developed the Conjunctive Consolidation Threshold model, a tripartite framework for reward-memory encoding prevention. Three sole-authored preprints on OSF and Zenodo. Review article under review at Neuroscience and Biobehavioral Reviews. Co-authored paper in Alcohol, under review at Elsevier. Built three computational platforms. IMPRINT screens compounds for addiction liability using Bayesian inference. TOPOLOGIX applies topological data analysis to drug-protein interactions, with persistent homology and bipartite simplicial complexes. GATE evaluates brain-computer interface neural stimulation safety. All platforms are open source under Apache 2.0. Computational skills include Python with scipy, numpy, PyMC, and pandas; R for statistical analysis; topological data analysis with Ripser and Gudhi; neural simulation with NEURON and Brian2; structural biology with AlphaFold, RDKit, ADMET and QSAR tools; molecular dynamics with GROMACS; docking with AutoDock; workflow management with Nextflow and SLURM; database management with Supabase and Postgres; web development with JavaScript and Node.js. Employment history includes National Product Manager at Synthcare from March 2026 to present; Clinical Pharmacist at Ramset Pharmacy from January to March 2026; Research Assistant at the Centre for Drug Discovery, Development and Production, where I performed NMDA and insulin docking studies; Bioinformatics Researcher at the Genomic Surveillance of Antimicrobial Resistance unit of the Global Health Research Unit, where I built AMR surveillance pipelines. Endorsements from Kent Berridge at University of Michigan, Samuel Gershman at Harvard University, Nathaniel Daw at Princeton University, and Marcelo Mattar at New York University. arXiv endorsement from Samuel Gershman. Provisional patent on CCT core architecture filed Q3 2026. Applying for MSc in Computational Neuroscience at Medical University of Graz, Austria, starting October 2026. Seeking aDMG affiliation as independent researcher for the interim period. PROJECT SUMMARY Title: Topological and Bayesian Analysis of Ancient Reward Pathway Variants for Addiction Liability Prediction Specific Aim 1: Construct topological networks of ancient protein variants in dopamine, opioid, and glutamate receptor genes using persistent homology and bipartite simplicial complexes, comparing interaction topologies with modern variants to identify functional divergence. Specific Aim 2: Parameterize a Bayesian population dynamics model of reward encoding threshold evolution using ancient allele frequency trajectories from aDMG datasets, predicting selection coefficients acting on reward-related genes. Specific Aim 3: Apply the IMPRINT platform to predict how ancient alleles modulate modern addiction liability, generating a map of paleogenomic risk variants for validation against epidemiological data from Nigerian populations. Methodology: Ancient protein sequences will be inferred from aDMG ancient DNA data using sequence reconstruction tools. TOPOLOGIX will compute persistent homology features for each variant. Bayesian MCMC, implemented in PyMC, will estimate encoding probability under the CCT model for each variant. The population dynamics model from Zenodo 10.5281/zenodo.20492472 will be extended to incorporate allele frequency trajectories. Expected Outcomes: Identification of ancient alleles in reward pathway genes that alter topological interaction signatures. Estimates of selection coefficients acting on these genes across human evolutionary history. A ranked list of ancient variants with predicted effects on modern addiction liability. Relevance to aDMG: This project directly addresses the Coordination Center mission to integrate ancient DNA with modern genomics. The computational tools developed will be open source and applicable to other neurobiological traits. The focus on African populations addresses the underrepresentation of African ancient DNA in paleogenomic research. BUDGET NARRATIVE Requested amount: 50,000 USD for 12 months. Personnel: 30,000 USD. Independent researcher salary support for 12 months at 2,500 USD per month, covering living expenses in Lagos, Nigeria. No fringe benefits requested. Equipment: 5,000 USD. One high-performance computing workstation with GPU for molecular dynamics simulations and Bayesian MCMC. Specifications: 64 GB RAM, NVIDIA RTX 4090, 2 TB SSD. Local procurement in Lagos. Travel: 5,000 USD. One trip to aDMG Coordination Center meeting or workshop. Economy airfare from Lagos to United States, 2,000 USD. Accommodation and per diem for 7 days, 3,000 USD. Software and Data: 3,000 USD. Cloud computing credits for Google Cloud Platform or AWS for large-scale topological data analysis and Bayesian inference. Estimated 250 USD per month for 12 months. Publication and Dissemination: 4,000 USD. Open access publication fees for two manuscripts in peer-reviewed journals. 2,000 USD per manuscript. Materials and Supplies: 3,000 USD. Laboratory supplies for validation experiments if required. Stationery, printing, and communication costs. Total: 50,000 USD. CHECKLIST - [ ] Complete Grants.gov registration for Eniola Ayodele Olutogun - [ ] Obtain DUNS number or UEI for independent researcher status - [ ] Upload motivation letter as PDF - [ ] Upload research statement as PDF - [ ] Upload biographical sketch as PDF - [ ] Upload project summary as PDF - [ ] Upload budget narrative as PDF - [ ] Upload ORCID profile printout (0009-0001-9272-6735) - [ ] Upload GitHub profile link (github.com/AmunRaPtah) - [ ] Upload OSF preprint links (10.17605/OSF.IO/KG7B5, 10.17605/OSF.IO/EMY4U) - [ ] Upload Zenodo preprint link (10.5281/zenodo.20492472) - [ ] Upload letter of endorsement from Kent Berridge, University of Michigan - [ ] Upload letter of endorsement from Samuel Gershman, Harvard University - [ ] Upload letter of endorsement from Nathaniel Daw, Princeton University - [ ] Upload letter of endorsement from Marcelo Mattar, New York University - [ ] Upload provisional patent documentation for CCT core architecture - [ ] Upload B.Pharm certificate and transcript from University of Ibadan - [ ] Upload PCN pharmacist license - [ ] Upload proof of Nigerian nationality - [ ] Verify aDMG Coordination Center specific submission portal requirements - [ ] Confirm deadline on aDMG programme website EDITOR NOTES - Eligibility risk: The aDMG programme is an NIH grant. Independent researchers without institutional affiliation may not be eligible as principal investigators. Verify whether aDMG allows independent researchers or requires a sponsoring institution. If required, identify a Nigerian university or research institute willing to serve as host institution. University of Ibadan or Lagos University Teaching Hospital are options. - Fact verification: Confirm that the provisional patent on CCT core architecture was actually filed Q3 2026. The profile states Q3 2026, which is in the future relative to the current date. If the patent has not yet been filed, remove all references to it or change to future tense. - Gap: The profile does not specify any prior experience with ancient DNA, paleogenomics, or population genetics. The application must address this gap directly. Consider adding a sentence about completing online courses in paleogenomics or population genetics, or about collaborating with aDMG researchers who have this expertise. - Gap: The budget requests 50,000 USD but the programme amount is unspecified. Verify the typical award size for aDMG Coordination Center grants. If the programme typically funds larger collaborative projects, this budget may be too small. If it funds smaller pilot projects, the budget may be appropriate. - Gap: The profile mentions applying for MSc at MUG Graz starting October 2026. If the aDMG grant starts before that date, clarify how the independent researcher role will transition to a student role. Consider proposing a 12-month grant period from July 2026 to June 2027, with the MSc starting in October 2026 and continuing in parallel.