MOTIVATION LETTER Drug-induced bone marrow suppression remains a clinical blind spot in pharmacological safety assessment. Standard preclinical screens detect acute cytotoxicity but fail to predict the cumulative, threshold-driven collapse of hematopoietic stem cell populations that occurs weeks after drug exposure. My independent research on the Conjunctive Consolidation Threshold (CCT) model, a tripartite pharmacological framework originally designed for reward-memory encoding in addiction, provides a mathematical architecture directly transferable to this problem. The CCT model formalizes how three concurrent signals must cross a conjunctive threshold to produce a consolidated biological effect. In bone marrow, the same logic applies: hematopoietic stem cell exhaustion requires simultaneous disruption of niche signaling, metabolic stress, and replicative demand. My ODE/RK45 simulations and Bayesian MCMC validation, which demonstrated an 85.8 percent reduction in encoding probability and confirmed all five preregistered hypotheses (H1-H5), prove the framework can predict threshold-dependent biological collapse. The Defense Health Agency Contracting Activity funds investigator-initiated research that addresses critical gaps in military and veteran health. Bone marrow failure syndromes, whether from chemotherapy, radiation, or environmental toxin exposure, disproportionately affect service members and veterans. My CCT model, adapted to hematopoietic stem cell dynamics, can identify the precise drug-concentration and exposure-duration thresholds that trigger irreversible aplasia. My TOPOLOGIX platform, which uses persistent homology and bipartite simplicial complexes to analyze drug-protein interaction networks, has already demonstrated a minimum viable product for hERG cardiotoxicity screening. The same topological data analysis pipeline can map drug interactions with marrow niche signaling proteins, including CXCL12, KIT ligand, and thrombopoietin receptors, to predict disruption of stem cell maintenance. I hold a B.Pharm from the University of Ibadan with a German-equivalent grade of 1.9, am a PCN-licensed pharmacist, and have endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. My provisional patent on the CCT core architecture, filed Q3 2026, protects the translational pathway. I am currently an independent researcher affiliated with ZYCO in Lagos, Nigeria, and I am seeking a host institution to sponsor this application. The proposed work will be conducted in collaboration with hematologists at a Nigerian teaching hospital and computational biologists at ZYCO, ensuring both local relevance and global rigor. The DoW Bone Marrow Failure Investigator-Initiated Research Award is the ideal mechanism to validate the CCT model in a new domain, produce open-source prediction tools for drug-induced aplasia, and establish a computational pharmacology research program in sub-Saharan Africa. RESEARCH STATEMENT PROBLEM AND SIGNIFICANCE Drug-induced bone marrow suppression accounts for approximately 30 percent of all hematologic adverse drug reactions and is a leading cause of chemotherapy dose reduction, treatment delay, and mortality. Current risk assessment relies on animal models and in vitro colony-forming assays that do not capture the nonlinear, threshold-driven dynamics of hematopoietic stem cell exhaustion. The Defense Health Agency has identified bone marrow failure as a priority area because service members face unique exposures, including chemotherapeutic agents, radiation, and environmental toxins, that produce delayed, cumulative marrow injury. A computational framework that predicts the precise exposure thresholds for irreversible aplasia would transform clinical decision-making and drug development safety screening. PROPOSED APPROACH I will adapt my Conjunctive Consolidation Threshold (CCT) model to hematopoietic stem cell dynamics. The CCT model, originally validated for reward-memory encoding in addiction, formalizes a conjunctive threshold: three concurrent signals must each exceed a critical value for a consolidated biological effect to occur. In bone marrow, the three signals are: (1) direct cytotoxicity to hematopoietic stem cells, (2) disruption of the marrow niche microenvironment, and (3) replicative stress from increased hematopoietic demand. I will specify these signals as coupled ordinary differential equations, parameterized using published pharmacokinetic and pharmacodynamic data for known marrow-suppressive drugs including carboplatin, busulfan, and chloramphenicol. The model will be solved using RK45 integration and validated against clinical time-series data on neutrophil and platelet counts from published trials. The second component uses my TOPOLOGIX platform to analyze drug-protein interaction networks that govern marrow niche signaling. Persistent homology and bipartite simplicial complexes will map how drugs perturb the topology of protein interaction networks centered on CXCL12, KIT ligand, thrombopoietin, and their receptors. The minimum viable product for hERG cardiotoxicity screening, already built, demonstrates the platform can identify topological signatures of toxicity. For this project, I will train the pipeline on a curated library of 200 drugs with known bone marrow toxicity profiles, using RDKit for molecular featurization and Gudhi for topological computation. The output will be a set of topological features that predict marrow suppression risk, which will be integrated into the CCT model as a prior distribution for Bayesian parameter estimation. The third component is a Bayesian population dynamics model that predicts individual patient risk. Using PyMC for Markov chain Monte Carlo sampling, I will estimate the posterior distribution of the three threshold parameters across a simulated population of 10,000 virtual patients, incorporating variability in drug clearance, baseline marrow reserve, and genetic polymorphisms in drug-metabolizing enzymes. The model will output a probability of grade 3 or higher neutropenia for a given drug regimen, enabling personalized dose adjustment. TRANSLATIONAL PATHWAY AND COLLABORATION The immediate output will be an open-source Python package, MarrowCCT, that takes drug molecular structure and dosing regimen as input and returns a predicted bone marrow failure risk score. The package will be validated against retrospective clinical data from the FDA Adverse Event Reporting System and the Veterans Health Administration oncology database. Within 18 months, I will produce a prospective validation protocol for a Nigerian teaching hospital cohort of patients receiving myelosuppressive chemotherapy, with hematologist collaborators already identified through my network. The provisional patent on the CCT core architecture, filed Q3 2026, covers the mathematical framework and its application to any biological threshold system, including hematopoiesis. PERSONAL STATEMENT My trajectory from a B.Pharm at the University of Ibadan to independent computational pharmacology research in Lagos reflects a deliberate choice to build research capacity where it is most needed. Nigeria has 0.3 pharmacists per 1,000 population and virtually no computational pharmacology infrastructure. I built my entire research pipeline, from ODE/RK45 simulations to Bayesian MCMC validation to topological data analysis platforms, without a formal graduate programme, using open-source tools and remote collaboration. My five preregistered hypotheses for the CCT model were all confirmed, and the model achieved an 85.8 percent reduction in encoding probability with super-additivity of 12.8 percentage points. The review article is under review at Neuroscience and Biobehavioral Reviews, and a co-authored paper is under review at Alcohol. The DoW Bone Marrow Failure Award would fund the computational development and clinical validation of MarrowCCT, support my travel to collaborate with hematologists at the University of Ibadan Teaching Hospital, and cover publication fees for open-access dissemination. I am applying for MSc programmes at the Medical University of Graz and the University of Graz, Austria, starting October 2026, and this award would bridge my independent research to formal graduate training. The Defense Health Agency mission to protect service members from preventable medical harm aligns directly with my goal of building predictive tools that prevent drug-induced bone marrow failure in vulnerable populations. BUDGET JUSTIFICATION The requested funds will support three activities over 12 months. First, computational infrastructure: cloud computing credits for HPC-based Bayesian MCMC sampling and topological data analysis on a library of 200 drugs, estimated at 8,000 USD. Second, personnel: a part-time research assistant in Lagos to curate clinical data and manage the drug library, estimated at 12,000 USD. Third, dissemination: open-access publication fees for two manuscripts and travel to one international conference to present results, estimated at 10,000 USD. Total request: 30,000 USD. No equipment or indirect costs are requested, as all computational work will be performed on existing cloud infrastructure and the research assistant will work remotely. CHECKLIST - [ ] Confirm host institution sponsorship (ZYCO or University of Ibadan) and obtain signed letter of institutional support - [ ] Complete Grants.gov registration for Eniola Ayodele Olutogun as principal investigator - [ ] Obtain DUNS number or UEI for host institution - [ ] Upload research statement (this document, 600 words) - [ ] Upload motivation letter (this document, 500 words) - [ ] Upload budget justification (this document, 150 words) - [ ] Upload biosketch or CV including ORCID 0009-0001-9272-6735 and GitHub github.com/AmunRaPtah - [ ] Upload letters of collaboration from Kent Berridge, Samuel Gershman, Nathaniel Daw, or Marcelo Mattar - [ ] Upload provisional patent filing documentation for CCT core architecture - [ ] Upload preprints: OSF 10.17605/OSF.IO/KG7B5, OSF 10.17605/OSF.IO/EMY4U, Zenodo 10.5281/zenodo.20492472 - [ ] Verify eligibility: applicant must be affiliated with an eligible institution; confirm ZYCO or University of Ibadan qualifies - [ ] Confirm deadline: 11/04/2026, submit by 5:00 PM Eastern Time EDITOR NOTES - Eligibility risk: The grant requires an institutional sponsor. Eniola is currently an independent researcher. He must secure a letter of institutional support from ZYCO or a Nigerian university before submission. Verify that ZYCO is a registered entity eligible to receive federal funds. - Fact verification: The provisional patent on CCT core architecture is listed as Q3 2026. Confirm the filing date and patent application number. If not yet filed, remove this claim or mark as pending. - Collaboration letters: Eniola has endorsements from Berridge, Gershman, Daw, and Mattar, but these are endorsements for arXiv and general support, not formal letters of collaboration for this specific grant. He must request and obtain signed letters that explicitly reference the DoW Bone Marrow Failure project. - Clinical data access: The proposal references retrospective data from the FDA Adverse Event Reporting System and the Veterans Health Administration. Confirm that Eniola can access these databases as a non-US researcher. The VA database may require a US-based collaborator or data use agreement. - Host institution gap: The budget justification lists a research assistant in Lagos but does not name the institution that will employ them. Clarify whether ZYCO or University of Ibadan will serve as the host and whether they can administer the grant funds.