Draft — United Nations Volunteer Program 2026

MOTIVATION LETTER

The United Nations Volunteer Program places professionals where institutional capacity is thinnest and technical problems are most acute. My work as a pharmacist and computational researcher has been shaped by the same premise: that drug resistance, digital health infrastructure, and data systems in African health ministries fail most often at the point of implementation, not at the point of discovery. I am applying to serve as a UN Volunteer in a health informatics, data analysis, or antimicrobial resistance surveillance role, preferably with a UN health agency in Africa, because that is where my specific skillset can be applied directly to programmatic outcomes.

My current research, TOPOLOGIX, predicts drug-resistance mutations from protein sequence alone using ESM-2 language-model embeddings and drug fingerprints. On the Platinum benchmark of 553 mutations, it achieves an AUROC of 0.804 plus or minus 0.025, and it covers 100 percent of mutations, whereas structure-based tools like mCSM-lig cover only about 18 percent. This work grew directly from my prior experience building AMR genomics surveillance pipelines at the GHRU-GSAR bioinformatics group in Nigeria, where I processed whole-genome sequencing data to track resistant bacterial lineages. That combination, genomic surveillance plus machine-learning prediction, maps directly onto the UN's antimicrobial resistance priorities under the Global Action Plan on AMR.

I also bring production-grade data infrastructure skills that are often missing in ministry settings. I have built four independent DuckDB-based ingest-to-analyze pipelines across life sciences, tech, and social science domains, and I self-host local LLM serving with llama.cpp on Linux VPS infrastructure with systemd, Caddy TLS, and automated backup and disaster recovery. These are the exact components needed to stand up a functional data unit in a resource-constrained health agency.

My clinical background grounds this work. I am a PCN-licensed pharmacist with a B.Pharm from the University of Ibadan, and I have worked as a clinical pharmacist at Ramset Pharmacy and as National Product Manager at Synthcare. I understand what a clinician needs from a data system because I have been the person at the dispensing counter. I am currently enrolled in the M.Sc. Digital Health program at the Hasso Plattner Institute and the University of Potsdam, which will deepen my capacity to design health information systems that survive contact with real workflows.

UNV assignments require adaptability and cultural awareness. I am a Nigerian national who has worked across clinical, research, and product roles in Nigeria and now studies in Germany. I am prepared to relocate, to serve the full 6 to 12 month assignment duration, and to work in challenging environments. My motivation is the application of validated methods to programs that need them now, not academic advancement.

RESEARCH STATEMENT

The research line I bring to the United Nations Volunteer Program is TOPOLOGIX, a sequence-based machine learning system for predicting drug-resistance mutations. I selected this line over my other active projects because it is the one with direct, near-term applicability to global health programmatic priorities, specifically antimicrobial resistance surveillance and digital health infrastructure in African health systems. My other work, including the CCT model for addiction neuroscience and the neurocascade brain-circuit simulation engine, is methodologically rigorous but does not map onto UNV assignment descriptions with the same clarity.

TOPOLOGIX addresses a concrete operational problem. Structure-based resistance prediction tools require a resolved protein structure, which exists for only a fraction of clinically relevant mutations. On the Platinum benchmark, structure-limited tools like mCSM-lig cover roughly 18 percent of mutations. TOPOLOGIX, using ESM-2 protein language model delta-embeddings combined with Morgan/ECFP drug fingerprints and a Random Forest classifier, covers 100 percent of mutations and achieves an AUROC of 0.804 plus or minus 0.025 on the Platinum benchmark, and 0.634 on SKEMPI 2.0. This means a health agency can screen mutations of interest without waiting for structural biology.

The path to this result included a falsified hypothesis, which I report because it is the honest foundation of the current method. I first tested whether bipartite persistent homology, an opposition-distance metric computed with Ripser and GUDHI, could predict hERG cardiotoxicity from protein-ligand interface geometry. In a pre-registered, powered replication, the topological features did not beat a plain descriptor baseline, AUROC 0.8426 versus 0.8782. I then applied the same topological constructs to drug-resistance prediction and found they carried almost no signal, AUROC 0.425 and 0.485 on the Platinum benchmark. That negative result ruled out interface geometry as the driver and motivated the sequence-representation approach that became TOPOLOGIX. The method is not yet deployed in a live surveillance setting, and I am explicit about that. It is a validated research tool at the preprint stage, not a product.

My prior work in Nigeria is directly relevant to UNV assignments. At the GHRU-GSAR bioinformatics group, I contributed to AMR genomics surveillance, building pipelines to process bacterial whole-genome sequencing data. At the CDDDP, I performed NMDA and insulin docking studies. These roles taught me how genomic data flows through a surveillance system, from sample to sequence to report, and where the bottlenecks are.

For a UNV assignment, I would apply TOPOLOGIX and my surveillance pipeline experience in one of two ways. First, as a data analyst or health informatics officer supporting a UN health agency in Africa, I would build or maintain the data infrastructure for AMR surveillance, including the ingest-to-analyze pipelines I have built four times independently using DuckDB. Second, I would use TOPOLOGIX to screen resistance mutations in locally collected genomic data, providing an additional layer of evidence for treatment guideline decisions. Both applications require the same skills I have demonstrated: Python, PyMC, RDKit, Nextflow, SLURM, and production Linux systems operations.

I am applying to put these methods to work in a programmatic context, where the metric of success is not a publication but a functioning surveillance report that informs a clinical or policy decision. I am not applying to conduct academic research under UNV auspices.

ESSAY: MOTIVATION AND COMMITMENT TO VOLUNTEERISM

My commitment to volunteer service began in a professional context, not a charitable one. As a clinical pharmacist at Ramset Pharmacy and earlier as a pharmacy student at the University of Ibadan, I saw how often patients in Nigeria received treatment decisions made without access to local resistance data. Clinicians were making calls with incomplete information, not because they lacked skill, but because the surveillance infrastructure did not exist. That gap is a systems problem, and I have spent the years since building the skills to address it.

The UNV model appeals to me because it places technical expertise inside the institutions that need it, rather than outside them. A consultant can deliver a report and leave. A volunteer embedded in a ministry or agency unit can build the pipeline, train the staff who will run it, and troubleshoot the first six months of real data. I have done this kind of embedded work before. At the GHRU-GSAR bioinformatics group, I helped build AMR genomics surveillance pipelines that had to function with real samples, real sequencing runs, and real reporting deadlines. At Synthcare, as National Product Manager, I learned how to translate technical capacity into operational plans that non-technical colleagues could execute.

I am prepared for the conditions of UNV service. I am a Nigerian national, 29 years old, and I have worked in clinical, research, and product roles across Nigeria. I am comfortable with relocation, with long assignment durations, and with environments where infrastructure is intermittent. I speak English fluently and I am learning German for my M.Sc. program at the Hasso Plattner Institute. I understand the UN's values because I have worked in systems where those values, particularly accountability and leaving no one behind, are not abstractions but operational requirements.

My motivation is specific. I want to serve in a role where my TOPOLOGIX method and my surveillance pipeline experience can strengthen a UN health program's ability to detect and respond to drug resistance. That is the work I know how to do, and it is the work that matters most in the current global health landscape.

CHECKLIST

- [ ] Create account and complete candidate profile at https://www.unv.org/become-volunteer
- [ ] Upload CV in UNV standard format, including all employment history and education
- [ ] Verify ORCID (0009-0001-9272-6735) and GitHub (github.com/AmunRaPtah) are linked in profile
- [ ] Search UNV assignments for health informatics, data analysis, or AMR surveillance roles in Africa
- [ ] Tailor the motivation letter above to the specific assignment description, referencing the duty station and agency
- [ ] Confirm eligibility for the specific assignment, including any degree and years-of-experience requirements
- [ ] Prepare two references, one from a supervisor at GHRU-GSAR or CDDDP and one from a clinical or product role
- [ ] Confirm availability for the assignment duration (6 to 12 months) and willingness to relocate
- [ ] Submit application before the specific assignment deadline, not the rolling general deadline
- [ ] Save a copy of the submitted application and assignment description for follow-up

EDITOR NOTES

- Eligibility risk: UNV assignments often require a university degree and two or more years of relevant professional experience. The applicant has the degree and the experience, but the M.Sc. enrollment at HPI/Potsdam begins Winter Semester 2026/27, which may conflict with a 6 to 12 month UNV assignment. Confirm the assignment timeline against the academic calendar before applying.
- Fact verification: The AUROC values for TOPOLOGIX (0.804 plus or minus 0.025 on Platinum, 0.634 on SKEMPI 2.0) and the coverage comparison (100 percent versus 18 percent) must be verified against the current preprint before submission. The hERG baseline comparison (0.8426 versus 0.8782) and the Platinum resistance topology results (0.425 and 0.485) must also be confirmed as reported.
- Gap to fill: The applicant must insert the specific UNV assignment title and duty station into the motivation letter and essay. The current draft is written for a general health informatics or AMR surveillance role; it must be tailored to the actual assignment description once selected.
- Honest framing: The research statement correctly presents TOPOLOGIX as a validated research tool at preprint stage, not a deployed product. Do not upgrade this claim in the application. The ergofluids project is not mentioned because it is behind a real-data validation gate that did not meet its primary pre-registered criterion; do not include it in any UNV materials.
- Language and tone: The motivation letter opens with the UNV program, not with the applicant, which is correct. The essay opens with the applicant's professional context, which is acceptable for a short-answer response. No banned phrases are used. Maintain this standard in any additional materials.

MOTIVATION LETTER

The United Nations Volunteer Program places professionals where institutional capacity is thinnest and technical problems are most acute. My work as a pharmacist and computational researcher has been shaped by the same premise: that drug resistance, digital health infrastructure, and data systems in African health ministries fail most often at the point of implementation, not at the point of discovery. I am applying to serve as a UN Volunteer in a health informatics, data analysis, or antimicrobial resistance surveillance role, preferably with a UN health agency in Africa, because that is where my specific skillset can be applied directly to programmatic outcomes.

My current research, TOPOLOGIX, predicts drug-resistance mutations from protein sequence alone using ESM-2 language-model embeddings and drug fingerprints. On the Platinum benchmark of 553 mutations, it achieves an AUROC of 0.804 plus or minus 0.025, and it covers 100 percent of mutations, whereas structure-based tools like mCSM-lig cover only about 18 percent. This work grew directly from my prior experience building AMR genomics surveillance pipelines at the GHRU-GSAR bioinformatics group in Nigeria, where I processed whole-genome sequencing data to track resistant bacterial lineages. That combination, genomic surveillance plus machine-learning prediction, maps directly onto the UN's antimicrobial resistance priorities under the Global Action Plan on AMR.

I also bring production-grade data infrastructure skills that are often missing in ministry settings. I have built four independent DuckDB-based ingest-to-analyze pipelines across life sciences, tech, and social science domains, and I self-host local LLM serving with llama.cpp on Linux VPS infrastructure with systemd, Caddy TLS, and automated backup and disaster recovery. These are the exact components needed to stand up a functional data unit in a resource-constrained health agency.

My clinical background grounds this work. I am a PCN-licensed pharmacist with a B.Pharm from the University of Ibadan, and I have worked as a clinical pharmacist at Ramset Pharmacy and as National Product Manager at Synthcare. I understand what a clinician needs from a data system because I have been the person at the dispensing counter. I am currently enrolled in the M.Sc. Digital Health program at the Hasso Plattner Institute and the University of Potsdam, which will deepen my capacity to design health information systems that survive contact with real workflows.

UNV assignments require adaptability and cultural awareness. I am a Nigerian national who has worked across clinical, research, and product roles in Nigeria and now studies in Germany. I am prepared to relocate, to serve the full 6 to 12 month assignment duration, and to work in challenging environments. My motivation is the application of validated methods to programs that need them now, not academic advancement.

RESEARCH STATEMENT

The research line I bring to the United Nations Volunteer Program is TOPOLOGIX, a sequence-based machine learning system for predicting drug-resistance mutations. I selected this line over my other active projects because it is the one with direct, near-term applicability to global health programmatic priorities, specifically antimicrobial resistance surveillance and digital health infrastructure in African health systems. My other work, including the CCT model for addiction neuroscience and the neurocascade brain-circuit simulation engine, is methodologically rigorous but does not map onto UNV assignment descriptions with the same clarity.

TOPOLOGIX addresses a concrete operational problem. Structure-based resistance prediction tools require a resolved protein structure, which exists for only a fraction of clinically relevant mutations. On the Platinum benchmark, structure-limited tools like mCSM-lig cover roughly 18 percent of mutations. TOPOLOGIX, using ESM-2 protein language model delta-embeddings combined with Morgan/ECFP drug fingerprints and a Random Forest classifier, covers 100 percent of mutations and achieves an AUROC of 0.804 plus or minus 0.025 on the Platinum benchmark, and 0.634 on SKEMPI 2.0. This means a health agency can screen mutations of interest without waiting for structural biology.

The path to this result included a falsified hypothesis, which I report because it is the honest foundation of the current method. I first tested whether bipartite persistent homology, an opposition-distance metric computed with Ripser and GUDHI, could predict hERG cardiotoxicity from protein-ligand interface geometry. In a pre-registered, powered replication, the topological features did not beat a plain descriptor baseline, AUROC 0.8426 versus 0.8782. I then applied the same topological constructs to drug-resistance prediction and found they carried almost no signal, AUROC 0.425 and 0.485 on the Platinum benchmark. That negative result ruled out interface geometry as the driver and motivated the sequence-representation approach that became TOPOLOGIX. The method is not yet deployed in a live surveillance setting, and I am explicit about that. It is a validated research tool at the preprint stage, not a product.

My prior work in Nigeria is directly relevant to UNV assignments. At the GHRU-GSAR bioinformatics group, I contributed to AMR genomics surveillance, building pipelines to process bacterial whole-genome sequencing data. At the CDDDP, I performed NMDA and insulin docking studies. These roles taught me how genomic data flows through a surveillance system, from sample to sequence to report, and where the bottlenecks are.

For a UNV assignment, I would apply TOPOLOGIX and my surveillance pipeline experience in one of two ways. First, as a data analyst or health informatics officer supporting a UN health agency in Africa, I would build or maintain the data infrastructure for AMR surveillance, including the ingest-to-analyze pipelines I have built four times independently using DuckDB. Second, I would use TOPOLOGIX to screen resistance mutations in locally collected genomic data, providing an additional layer of evidence for treatment guideline decisions. Both applications require the same skills I have demonstrated: Python, PyMC, RDKit, Nextflow, SLURM, and production Linux systems operations.

I am applying to put these methods to work in a programmatic context, where the metric of success is not a publication but a functioning surveillance report that informs a clinical or policy decision. I am not applying to conduct academic research under UNV auspices.

ESSAY: MOTIVATION AND COMMITMENT TO VOLUNTEERISM

My commitment to volunteer service began in a professional context, not a charitable one. As a clinical pharmacist at Ramset Pharmacy and earlier as a pharmacy student at the University of Ibadan, I saw how often patients in Nigeria received treatment decisions made without access to local resistance data. Clinicians were making calls with incomplete information, not because they lacked skill, but because the surveillance infrastructure did not exist. That gap is a systems problem, and I have spent the years since building the skills to address it.

The UNV model appeals to me because it places technical expertise inside the institutions that need it, rather than outside them. A consultant can deliver a report and leave. A volunteer embedded in a ministry or agency unit can build the pipeline, train the staff who will run it, and troubleshoot the first six months of real data. I have done this kind of embedded work before. At the GHRU-GSAR bioinformatics group, I helped build AMR genomics surveillance pipelines that had to function with real samples, real sequencing runs, and real reporting deadlines. At Synthcare, as National Product Manager, I learned how to translate technical capacity into operational plans that non-technical colleagues could execute.

I am prepared for the conditions of UNV service. I am a Nigerian national, 29 years old, and I have worked in clinical, research, and product roles across Nigeria. I am comfortable with relocation, with long assignment durations, and with environments where infrastructure is intermittent. I speak English fluently and I am learning German for my M.Sc. program at the Hasso Plattner Institute. I understand the UN's values because I have worked in systems where those values, particularly accountability and leaving no one behind, are not abstractions but operational requirements.

My motivation is specific. I want to serve in a role where my TOPOLOGIX method and my surveillance pipeline experience can strengthen a UN health program's ability to detect and respond to drug resistance. That is the work I know how to do, and it is the work that matters most in the current global health landscape.

CHECKLIST

- [ ] Create account and complete candidate profile at https://www.unv.org/become-volunteer
- [ ] Upload CV in UNV standard format, including all employment history and education
- [ ] Verify ORCID (0009-0001-9272-6735) and GitHub (github.com/AmunRaPtah) are linked in profile
- [ ] Search UNV assignments for health informatics, data analysis, or AMR surveillance roles in Africa
- [ ] Tailor the motivation letter above to the specific assignment description, referencing the duty station and agency
- [ ] Confirm eligibility for the specific assignment, including any degree and years-of-experience requirements
- [ ] Prepare two references, one from a supervisor at GHRU-GSAR or CDDDP and one from a clinical or product role
- [ ] Confirm availability for the assignment duration (6 to 12 months) and willingness to relocate
- [ ] Submit application before the specific assignment deadline, not the rolling general deadline
- [ ] Save a copy of the submitted application and assignment description for follow-up

EDITOR NOTES

- Eligibility risk: UNV assignments often require a university degree and two or more years of relevant professional experience. The applicant has the degree and the experience, but the M.Sc. enrollment at HPI/Potsdam begins Winter Semester 2026/27, which may conflict with a 6 to 12 month UNV assignment. Confirm the assignment timeline against the academic calendar before applying.
- Fact verification: The AUROC values for TOPOLOGIX (0.804 plus or minus 0.025 on Platinum, 0.634 on SKEMPI 2.0) and the coverage comparison (100 percent versus 18 percent) must be verified against the current preprint before submission. The hERG baseline comparison (0.8426 versus 0.8782) and the Platinum resistance topology results (0.425 and 0.485) must also be confirmed as reported.
- Gap to fill: The applicant must insert the specific UNV assignment title and duty station into the motivation letter and essay. The current draft is written for a general health informatics or AMR surveillance role; it must be tailored to the actual assignment description once selected.
- Honest framing: The research statement correctly presents TOPOLOGIX as a validated research tool at preprint stage, not a deployed product. Do not upgrade this claim in the application. The ergofluids project is not mentioned because it is behind a real-data validation gate that did not meet its primary pre-registered criterion; do not include it in any UNV materials.
- Language and tone: The motivation letter opens with the UNV program, not with the applicant, which is correct. The essay opens with the applicant's professional context, which is acceptable for a short-answer response. No banned phrases are used. Maintain this standard in any additional materials.

AI Draft — United Nations Volunteer Program 2026