Draft — Emergent Ventures

MOTIVATION LETTER

Emergent Ventures funds people who build things that do not fit inside existing boxes. I am one of those people. I am a 29-year-old Nigerian pharmacist who, without a PhD or an institutional lab, has executed five pre-registered, independently funded computational research projects across addiction neuroscience, protein machine learning, and dynamical systems. My CCT model, a tripartite pharmacological framework for reward-memory encoding prevention in addiction, confirmed all five pre-registered hypotheses using Bayesian MCMC calibration on 14 free parameters with literature-elicited priors from an 1,847-record screen. Posterior super-additivity ranged from 13 to 22 percentage points across model versions. Three sole-authored preprints are under review at peer-reviewed journals. My TOPOLOGIX tool predicts drug-resistance mutations from protein sequence alone, achieving an AUROC of 0.804 on the Platinum benchmark covering 553 mutations. It covers 100 percent of mutations versus roughly 18 percent for structure-based tools like mCSM-lig, which tops out at 0.70. These projects address global health challenges that hit Africa hardest: addiction treatment access is near zero across most of the continent, and antimicrobial resistance is projected to cause 4.1 million deaths in Africa by 2050. I built these tools on a laptop in Nigeria using open-source software and self-hosted infrastructure. That is the kind of resourcefulness and independence Emergent Ventures exists to amplify. I am applying for the OISI track because my work sits at the intersection of open science, independent inquiry, and computational methods that bypass expensive wet-lab infrastructure. I do not need a building. I need time, compute, and the freedom to keep asking questions that established funding mechanisms ignore. A grant from Emergent Ventures would allow me to complete the clinical validation of the CCT model through behavioral-data fits, extend TOPOLOGIX to cover all WHO priority pathogens, and build the full neurocascade simulation engine into a publicly usable tool for researchers in low-resource settings. I have already done the hard part: proving the ideas work. Now I need the runway to ship them.

SHORT ESSAY: RESEARCH VISION

My research vision is to build computational infrastructure for neuroscience and drug discovery that works from anywhere, including Nigeria. The dominant model in these fields requires expensive labs, large teams, and institutional affiliation. That model excludes most of the world. I am building an alternative: a stack of open-source, simulation-driven tools that run on consumer hardware and produce results that match or beat institutional baselines. The CCT model is a dynamical-systems framework for addiction that replaces animal experiments with ODE simulations calibrated to human literature. TOPOLOGIX replaces crystallography with protein-language-model embeddings. neurocascade replaces multi-site electrophysiology with coupled pharmacokinetic-circuit-behavioral ODE layers. Each tool is pre-registered, Bayesian-calibrated, and tested against published data. Each tool is designed to be used by a single researcher with a laptop. The long-term goal is a unified platform where a researcher in Lagos can simulate a drug-receptor interaction, predict resistance mutations, and model population-level treatment outcomes without leaving their desk. That platform does not exist today. I intend to build it.

SHORT ESSAY: WHY EMERGENT VENTURES

I am applying to Emergent Ventures because it funds people, not projects, and it funds ideas that established systems reject. I have applied to traditional fellowships and grants. The feedback is consistent: you need a PhD, you need a PI, you need a lab. I have none of those. What I have is five completed research projects, three preprints under review, a co-authored paper under review at Alcohol, endorsements from Kent Berridge, Samuel Gershman, Nathaniel Daw, and Marcelo Mattar, and a track record of shipping code and papers without supervision. Emergent Ventures is one of the few programmes that sees that as a feature, not a bug. The rolling application and rapid decision process also matter. I am enrolled in an M.Sc. at HPI starting winter 2026, and I need funding that moves at the speed of my work, not the speed of an academic calendar. A grant from Emergent Ventures would let me focus on research full-time during the critical transition from independent work to formal graduate study. I am asking for the resources to keep doing what I have already proven I can do, at a larger scale.

RESEARCH STATEMENT

I am a computational researcher working at the intersection of addiction neuroscience, protein machine learning, and dynamical-systems methods. My work is defined by three principles: pre-registration, Bayesian calibration, and open-source release. Every project I list here has a pre-registration, a public code repository, and a preprint on OSF or Zenodo. I do not publish results I cannot reproduce.

The CCT model is my primary contribution to addiction neuroscience. It is a tripartite pharmacological framework that models reward-memory encoding prevention through three coupled axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. The model is implemented as a system of ODEs solved with RK45 and calibrated with Bayesian MCMC using PyMC's DEMetropolisZ sampler. I screened 1,847 records from the literature to elicit priors for 14 free parameters. All five pre-registered hypotheses were confirmed. Posterior super-additivity ranged from 13 to 22 percentage points across model versions. Three sole-authored preprints are under review at IART, PNPBP, and NBR. A co-authored paper is under review at Alcohol. The next step is to fit the model to real behavioral data, which requires compute time and access to published datasets.

In protein machine learning, I built TOPOLOGIX, a tool that predicts drug-resistance mutations from sequence alone. It uses ESM-2 protein-language-model delta-embeddings combined with Morgan/ECFP drug fingerprints and a Random Forest classifier. On the Platinum benchmark of 553 mutations, it achieves an AUROC of 0.804 plus or minus 0.025. On SKEMPI 2.0, it achieves 0.634. It beats structure-based baselines like mCSM-lig, which scores roughly 0.70, while covering 100 percent of mutations versus roughly 18 percent for structure-limited tools. This matters because most clinically relevant mutations occur in proteins with no solved structure. TOPOLOGIX works on any sequence. I am currently extending it to cover all WHO priority pathogens.

I also completed a pre-registered replication study on topological data analysis for hERG cardiotoxicity prediction. The published literature claimed that bipartite persistent homology could predict cardiotoxicity from protein-ligand interface geometry. I ran a powered replication using Ripser and GUDHI and found that topological features do not beat a plain descriptor baseline: AUROC 0.8426 versus 0.8782. I published the result directly, without reframing. A follow-up study applied the same methods to drug-resistance prediction and found they carry almost no signal: AUROC 0.425 and 0.485 on the Platinum benchmark. These studies settled questions the field had never actually tested.

My dynamical-systems work includes neurocascade, a receptor-to-behavior brain-circuit simulation engine that couples pharmacokinetics to receptor-binding to Wilson-Cowan circuit dynamics to behavioral-readout ODE layers. Three literature-calibrated receptor and circuit systems are implemented: mu-opioid, D2 dopamine, and GABA-A. All 62 tests pass. The circuit-layer parameters are explicitly labeled as illustrative pending real behavioral-data fits. I also built ergofluids, which extends Koopman-operator and Dynamic Mode Decomposition methods with a Mori-Zwanzig memory kernel to model macromolecular drug-vehicle transport through dense tumor tissue. The project is pre-registered with a gated validation pipeline. 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 that result directly rather than reframing it. Methods-validation research does not require positive results.

Finally, I built psyche-twin, a typed multi-scale knowledge-graph architecture for self-modeling. Multiple independent evidence streams, including LLM-derived, assessment-derived, behavioral, and document-derived, fuse into one append-only event log. Disagreement between streams becomes an explicit graph edge rather than being averaged away. A first-person conversational interface sits on top of the graph, generating responses grounded in the graph's current state. This project is at an earlier stage but represents a long-term interest in computational models of cognition and selfhood.

My technical stack includes Python with scipy, numpy, PyMC, and pandas; R; topological data analysis with Ripser and GUDHI; neural simulation with NEURON and Brian2; protein modeling with AlphaFold and RDKit; molecular dynamics with GROMACS; docking with AutoDock; and HPC workflow management with Nextflow and SLURM. I also maintain four independent DuckDB-based ingest-to-analyze corpus and RAG pipelines across life sciences, tech and AI security, and social science domains. I self-host local LLM serving with llama.cpp and manage production systems on Linux VPS with systemd, Caddy TLS, CI/CD, and automated backup and disaster-recovery.

I hold a B.Pharm from the University of Ibadan with a CGPA of 5.1 out of 7.0, a German equivalent of 1.9, and I am a PCN-licensed pharmacist. I am enrolled in the M.Sc. Digital Health programme at the Hasso Plattner Institute and the University of Potsdam starting winter semester 2026. I currently work as National Product Manager at Synthcare. I have previously worked as a clinical pharmacist and as a research assistant in computational drug design and antimicrobial resistance genomics.

I am seeking funding to complete the clinical validation of the CCT model, extend TOPOLOGIX to all WHO priority pathogens, and build the neurocascade simulation engine into a publicly usable tool. I do not need a lab. I need compute time, dataset access, and the freedom to focus on research. Emergent Ventures is the right programme for that.

CHECKLIST

- [ ] Motivation letter (300-500 words, written above)
- [ ] Short essay: Research vision (200-350 words, written above)
- [ ] Short essay: Why Emergent Ventures (200-350 words, written above)
- [ ] Research statement (400-600 words, written above)
- [ ] CV or resume (applicant must prepare from profile data)
- [ ] Links to preprints: three sole-authored CCT model preprints on OSF/Zenodo
- [ ] Link to TOPOLOGIX repository on GitHub
- [ ] Link to neurocascade repository on GitHub
- [ ] Link to ergofluids repository on GitHub
- [ ] Link to psyche-twin repository on GitHub
- [ ] Link to ORCID profile: 0009-0001-9272-6735
- [ ] Link to personal site: zyco.org
- [ ] Link to GitHub: github.com/AmunRaPtah
- [ ] Confirmation of M.Sc. enrollment at HPI/Potsdam (winter 2026)
- [ ] Confirmation of PCN pharmacist license
- [ ] Endorsement letters or contact information for Kent Berridge, Samuel Gershman, Nathaniel Daw, Marcelo Mattar (optional but recommended)

EDITOR NOTES

- Eligibility risk: Emergent Ventures has no strict eligibility criteria, but the OISI track may have unstated preferences. Confirm that independent researchers with a pending graduate enrollment are explicitly welcome. If the programme prefers full-time founders or researchers without academic commitments, the M.Sc. enrollment could be a complication.
- Fact verification needed: Confirm that all three preprints are still under review at the named journals (IART, PNPBP, NBR) and that the co-authored paper is still under review at Alcohol. Journal review status changes quickly.
- Gap: The profile does not include a specific funding amount request. The applicant should decide whether to ask for the full $100,000 or a smaller amount, and state that clearly in the motivation letter or short essay. The current draft implies a request for compute time and dataset access but does not name a figure.

MOTIVATION LETTER

Emergent Ventures funds people who build things that do not fit inside existing boxes. I am one of those people. I am a 29-year-old Nigerian pharmacist who, without a PhD or an institutional lab, has executed five pre-registered, independently funded computational research projects across addiction neuroscience, protein machine learning, and dynamical systems. My CCT model, a tripartite pharmacological framework for reward-memory encoding prevention in addiction, confirmed all five pre-registered hypotheses using Bayesian MCMC calibration on 14 free parameters with literature-elicited priors from an 1,847-record screen. Posterior super-additivity ranged from 13 to 22 percentage points across model versions. Three sole-authored preprints are under review at peer-reviewed journals. My TOPOLOGIX tool predicts drug-resistance mutations from protein sequence alone, achieving an AUROC of 0.804 on the Platinum benchmark covering 553 mutations. It covers 100 percent of mutations versus roughly 18 percent for structure-based tools like mCSM-lig, which tops out at 0.70. These projects address global health challenges that hit Africa hardest: addiction treatment access is near zero across most of the continent, and antimicrobial resistance is projected to cause 4.1 million deaths in Africa by 2050. I built these tools on a laptop in Nigeria using open-source software and self-hosted infrastructure. That is the kind of resourcefulness and independence Emergent Ventures exists to amplify. I am applying for the OISI track because my work sits at the intersection of open science, independent inquiry, and computational methods that bypass expensive wet-lab infrastructure. I do not need a building. I need time, compute, and the freedom to keep asking questions that established funding mechanisms ignore. A grant from Emergent Ventures would allow me to complete the clinical validation of the CCT model through behavioral-data fits, extend TOPOLOGIX to cover all WHO priority pathogens, and build the full neurocascade simulation engine into a publicly usable tool for researchers in low-resource settings. I have already done the hard part: proving the ideas work. Now I need the runway to ship them.

SHORT ESSAY: RESEARCH VISION

My research vision is to build computational infrastructure for neuroscience and drug discovery that works from anywhere, including Nigeria. The dominant model in these fields requires expensive labs, large teams, and institutional affiliation. That model excludes most of the world. I am building an alternative: a stack of open-source, simulation-driven tools that run on consumer hardware and produce results that match or beat institutional baselines. The CCT model is a dynamical-systems framework for addiction that replaces animal experiments with ODE simulations calibrated to human literature. TOPOLOGIX replaces crystallography with protein-language-model embeddings. neurocascade replaces multi-site electrophysiology with coupled pharmacokinetic-circuit-behavioral ODE layers. Each tool is pre-registered, Bayesian-calibrated, and tested against published data. Each tool is designed to be used by a single researcher with a laptop. The long-term goal is a unified platform where a researcher in Lagos can simulate a drug-receptor interaction, predict resistance mutations, and model population-level treatment outcomes without leaving their desk. That platform does not exist today. I intend to build it.

SHORT ESSAY: WHY EMERGENT VENTURES

I am applying to Emergent Ventures because it funds people, not projects, and it funds ideas that established systems reject. I have applied to traditional fellowships and grants. The feedback is consistent: you need a PhD, you need a PI, you need a lab. I have none of those. What I have is five completed research projects, three preprints under review, a co-authored paper under review at Alcohol, endorsements from Kent Berridge, Samuel Gershman, Nathaniel Daw, and Marcelo Mattar, and a track record of shipping code and papers without supervision. Emergent Ventures is one of the few programmes that sees that as a feature, not a bug. The rolling application and rapid decision process also matter. I am enrolled in an M.Sc. at HPI starting winter 2026, and I need funding that moves at the speed of my work, not the speed of an academic calendar. A grant from Emergent Ventures would let me focus on research full-time during the critical transition from independent work to formal graduate study. I am asking for the resources to keep doing what I have already proven I can do, at a larger scale.

RESEARCH STATEMENT

I am a computational researcher working at the intersection of addiction neuroscience, protein machine learning, and dynamical-systems methods. My work is defined by three principles: pre-registration, Bayesian calibration, and open-source release. Every project I list here has a pre-registration, a public code repository, and a preprint on OSF or Zenodo. I do not publish results I cannot reproduce.

The CCT model is my primary contribution to addiction neuroscience. It is a tripartite pharmacological framework that models reward-memory encoding prevention through three coupled axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. The model is implemented as a system of ODEs solved with RK45 and calibrated with Bayesian MCMC using PyMC's DEMetropolisZ sampler. I screened 1,847 records from the literature to elicit priors for 14 free parameters. All five pre-registered hypotheses were confirmed. Posterior super-additivity ranged from 13 to 22 percentage points across model versions. Three sole-authored preprints are under review at IART, PNPBP, and NBR. A co-authored paper is under review at Alcohol. The next step is to fit the model to real behavioral data, which requires compute time and access to published datasets.

In protein machine learning, I built TOPOLOGIX, a tool that predicts drug-resistance mutations from sequence alone. It uses ESM-2 protein-language-model delta-embeddings combined with Morgan/ECFP drug fingerprints and a Random Forest classifier. On the Platinum benchmark of 553 mutations, it achieves an AUROC of 0.804 plus or minus 0.025. On SKEMPI 2.0, it achieves 0.634. It beats structure-based baselines like mCSM-lig, which scores roughly 0.70, while covering 100 percent of mutations versus roughly 18 percent for structure-limited tools. This matters because most clinically relevant mutations occur in proteins with no solved structure. TOPOLOGIX works on any sequence. I am currently extending it to cover all WHO priority pathogens.

I also completed a pre-registered replication study on topological data analysis for hERG cardiotoxicity prediction. The published literature claimed that bipartite persistent homology could predict cardiotoxicity from protein-ligand interface geometry. I ran a powered replication using Ripser and GUDHI and found that topological features do not beat a plain descriptor baseline: AUROC 0.8426 versus 0.8782. I published the result directly, without reframing. A follow-up study applied the same methods to drug-resistance prediction and found they carry almost no signal: AUROC 0.425 and 0.485 on the Platinum benchmark. These studies settled questions the field had never actually tested.

My dynamical-systems work includes neurocascade, a receptor-to-behavior brain-circuit simulation engine that couples pharmacokinetics to receptor-binding to Wilson-Cowan circuit dynamics to behavioral-readout ODE layers. Three literature-calibrated receptor and circuit systems are implemented: mu-opioid, D2 dopamine, and GABA-A. All 62 tests pass. The circuit-layer parameters are explicitly labeled as illustrative pending real behavioral-data fits. I also built ergofluids, which extends Koopman-operator and Dynamic Mode Decomposition methods with a Mori-Zwanzig memory kernel to model macromolecular drug-vehicle transport through dense tumor tissue. The project is pre-registered with a gated validation pipeline. 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 that result directly rather than reframing it. Methods-validation research does not require positive results.

Finally, I built psyche-twin, a typed multi-scale knowledge-graph architecture for self-modeling. Multiple independent evidence streams, including LLM-derived, assessment-derived, behavioral, and document-derived, fuse into one append-only event log. Disagreement between streams becomes an explicit graph edge rather than being averaged away. A first-person conversational interface sits on top of the graph, generating responses grounded in the graph's current state. This project is at an earlier stage but represents a long-term interest in computational models of cognition and selfhood.

My technical stack includes Python with scipy, numpy, PyMC, and pandas; R; topological data analysis with Ripser and GUDHI; neural simulation with NEURON and Brian2; protein modeling with AlphaFold and RDKit; molecular dynamics with GROMACS; docking with AutoDock; and HPC workflow management with Nextflow and SLURM. I also maintain four independent DuckDB-based ingest-to-analyze corpus and RAG pipelines across life sciences, tech and AI security, and social science domains. I self-host local LLM serving with llama.cpp and manage production systems on Linux VPS with systemd, Caddy TLS, CI/CD, and automated backup and disaster-recovery.

I hold a B.Pharm from the University of Ibadan with a CGPA of 5.1 out of 7.0, a German equivalent of 1.9, and I am a PCN-licensed pharmacist. I am enrolled in the M.Sc. Digital Health programme at the Hasso Plattner Institute and the University of Potsdam starting winter semester 2026. I currently work as National Product Manager at Synthcare. I have previously worked as a clinical pharmacist and as a research assistant in computational drug design and antimicrobial resistance genomics.

I am seeking funding to complete the clinical validation of the CCT model, extend TOPOLOGIX to all WHO priority pathogens, and build the neurocascade simulation engine into a publicly usable tool. I do not need a lab. I need compute time, dataset access, and the freedom to focus on research. Emergent Ventures is the right programme for that.

CHECKLIST

- [ ] Motivation letter (300-500 words, written above)
- [ ] Short essay: Research vision (200-350 words, written above)
- [ ] Short essay: Why Emergent Ventures (200-350 words, written above)
- [ ] Research statement (400-600 words, written above)
- [ ] CV or resume (applicant must prepare from profile data)
- [ ] Links to preprints: three sole-authored CCT model preprints on OSF/Zenodo
- [ ] Link to TOPOLOGIX repository on GitHub
- [ ] Link to neurocascade repository on GitHub
- [ ] Link to ergofluids repository on GitHub
- [ ] Link to psyche-twin repository on GitHub
- [ ] Link to ORCID profile: 0009-0001-9272-6735
- [ ] Link to personal site: zyco.org
- [ ] Link to GitHub: github.com/AmunRaPtah
- [ ] Confirmation of M.Sc. enrollment at HPI/Potsdam (winter 2026)
- [ ] Confirmation of PCN pharmacist license
- [ ] Endorsement letters or contact information for Kent Berridge, Samuel Gershman, Nathaniel Daw, Marcelo Mattar (optional but recommended)

EDITOR NOTES

- Eligibility risk: Emergent Ventures has no strict eligibility criteria, but the OISI track may have unstated preferences. Confirm that independent researchers with a pending graduate enrollment are explicitly welcome. If the programme prefers full-time founders or researchers without academic commitments, the M.Sc. enrollment could be a complication.
- Fact verification needed: Confirm that all three preprints are still under review at the named journals (IART, PNPBP, NBR) and that the co-authored paper is still under review at Alcohol. Journal review status changes quickly.
- Gap: The profile does not include a specific funding amount request. The applicant should decide whether to ask for the full $100,000 or a smaller amount, and state that clearly in the motivation letter or short essay. The current draft implies a request for compute time and dataset access but does not name a figure.

AI Draft — Emergent Ventures