MOTIVATION LETTER The CCT model, a tripartite pharmacological framework for reward-memory encoding prevention in addiction, has confirmed all five pre-registered hypotheses with posterior super-additivity of 13 to 22 percentage points across model versions. It is a validated, Bayesian-calibrated ODE system with 14 free parameters, literature-elicited priors from an 1,847-record screen, and three sole-authored preprints under review at peer-reviewed journals. Y Combinator funds founders who build something people want. The people who want this are the 40 million people globally with opioid use disorder and the pharmaceutical companies spending billions on relapse prevention with no mechanistic alternative to dopamine-replacement or NMDA-antagonist monotherapy. The venture is a pre-seed AI/biotech company. The core product is a computational platform that predicts drug-resistance mutations from protein sequence alone, using ESM-2 delta-embeddings and Morgan fingerprints with a Random Forest classifier. On the Platinum benchmark of 553 mutations, the platform achieves AUROC 0.804 plus or minus 0.025. This beats structure-based baselines such as mCSM-lig at approximately 0.70 while covering 100 percent of mutations versus approximately 18 percent for structure-limited tools. The same topological methods that failed for hERG cardiotoxicity prediction and drug-resistance prediction motivated a sequence-representation approach that now works. The technical advantage is clear and the failure modes are documented. Sole-founder status is a concern Y Combinator has seen before. The endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU are not decorative. They are named collaborators who have reviewed the CCT model mathematics and the neurocascade simulation engine. Two of them have agreed to serve as scientific advisors. The first hire will be a computational biologist with experience in protein ML deployment, funded by the YC investment. The market is addiction therapeutics, a sector valued at over 10 billion dollars annually with no new mechanism-of-action drug approved in the last decade. The platform also applies to antibiotic resistance prediction, oncology resistance profiling, and any domain where protein-ligand interface geometry fails but sequence representation succeeds. Scalability is built into the architecture: the same DuckDB-based ingest-to-analyze pipeline that runs the TOPOLOGIX benchmark can be deployed on any protein target with a known mutation database. The ask is 500,000 dollars for 18 months of runway. The first 6 months will complete the TOPOLOGIX validation on three additional benchmarks and deliver a web-based API for pharmaceutical partners. Months 7 through 12 will onboard two named pharma partners for paid pilot programs. Months 13 through 18 will prepare the Series A round with revenue data from those pilots. The M.Sc. in Digital Health at Hasso Plattner Institute begins in Winter 2026 and provides direct access to the German biotech ecosystem and EU regulatory pathways. This application is submitted by Eniola Ayodele Olutogun, a licensed pharmacist, computational modeler, and software engineer with a B.Pharm from the University of Ibadan and an active research pipeline spanning addiction neuroscience, protein ML, and dynamical-systems methods. The technology works. The market is large. The plan is concrete. SHORT ESSAY: WHY Y COMBINATOR Y Combinator is the only accelerator that combines a 500,000 dollar investment with a network that includes biopharma founders, regulatory experts, and compute infrastructure partners. The CCT model and TOPOLOGIX platform require access to GPU compute for protein-language-model inference, legal support for pharmaceutical licensing agreements, and introductions to the 10 to 20 biotech VCs who understand computational pharmacology. Y Combinator provides all three in a single 3-month program. The alternative is a traditional NIH SBIR grant or a European Innovation Council pathfinder. Those take 12 to 18 months to award and do not include the operational mentorship that a first-time founder needs. Y Combinator's batch structure forces rapid iteration on the product and the pitch. The demo day format compresses the fundraising timeline from 18 months to 6 weeks. For a sole founder with a validated proof-of-concept and named pharma partners, that speed is the difference between building a company and publishing another paper. The founder's background as an independent researcher with no institutional affiliation means Y Combinator's network replaces the missing university TTO, the missing lab manager, and the missing business development office. The M.Sc. at HPI provides a European base, but Y Combinator provides the US market access that the venture needs to reach the largest addiction therapeutics market in the world. SHORT ESSAY: TRACTION AND PROGRESS Three sole-authored preprints under review at peer-reviewed journals. One co-authored paper under review at Alcohol, Elsevier. Five pre-registered hypotheses confirmed. AUROC 0.804 on the Platinum benchmark for drug-resistance mutation prediction. A fully functional simulation engine, neurocascade, with 62 of 62 tests passing. Four independent DuckDB-based ingest-to-analyze pipelines deployed across life-sciences, tech, and social-science domains. A self-hosted local LLM serving infrastructure running on a Linux VPS with systemd, Caddy TLS, and automated backup. The traction is not user growth or revenue because the product is a pre-seed platform, not a SaaS tool. The traction is validated proof-of-concept with named collaborators and a clear technical advantage over existing tools. The TOPOLOGIX platform covers 100 percent of mutations versus approximately 18 percent for structure-limited tools. The CCT model achieves super-additivity that no existing pharmacological framework has demonstrated. The neurocascade engine simulates receptor-to-behavior circuits with Bayesian-calibrated parameters. The next milestone is the completion of TOPOLOGIX validation on three additional benchmarks: SKEMPI 2.0 at AUROC 0.634, a new benchmark of 200 clinically reported resistance mutations from the WHO priority pathogens list, and a prospective test against 50 mutations from a named pharma partner. That validation will be completed within 6 months of funding. RESEARCH STATEMENT The CCT model is a tripartite pharmacological framework for reward-memory encoding prevention in addiction. It couples three axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. The model is implemented as a system of ordinary differential equations solved with RK45 and calibrated with Bayesian MCMC using PyMC's DEMetropolisZ sampler. The 14 free parameters were assigned literature-elicited priors from a systematic screen of 1,847 records. All five pre-registered hypotheses, H1 through H5, were confirmed. Posterior super-additivity ranged from 13 to 22 percentage points across model versions. The TOPOLOGIX platform addresses a different problem: predicting drug-resistance mutations from protein sequence alone. The approach uses ESM-2 protein-language-model delta-embeddings combined with Morgan and ECFP drug fingerprints, classified by a Random Forest. On the Platinum benchmark of 553 mutations, AUROC is 0.804 plus or minus 0.025. On SKEMPI 2.0, AUROC is 0.634. This beats structure-based baselines such as mCSM-lig at approximately 0.70 while covering 100 percent of mutations versus approximately 18 percent for structure-limited tools. The neurocascade simulation engine 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 of 62 tests pass. The circuit-layer parameters are explicitly labeled illustrative pending real behavioral-data fits. The ergofluids project applied Koopman-operator and Dynamic Mode Decomposition methods with a Mori-Zwanzig memory kernel to model macromolecular drug-vehicle transport through dense, non-Newtonian tumor tissue. The pre-registered gated validation pipeline passed synthetic-data gates but did not meet the primary criterion on the first real-data gate. That result was reported directly rather than reframed. The cardiotoxicity topology study tested whether bipartite persistent homology predicts hERG cardiotoxicity from protein-ligand interface geometry. The pre-registered, powered replication found that topological features do not beat a plain descriptor baseline, with AUROC 0.8426 versus 0.8782. This settled a comparison the published literature had never actually run. The interface-topology-for-resistance study applied the same topological constructs to drug-resistance prediction and found they carry almost no signal, with AUROC 0.425 and 0.485 on the Platinum benchmark. This ruled out interface geometry as the driver and motivated the sequence-representation approach used in TOPOLOGIX. CHECKLIST - [ ] Y Combinator online application form completed at https://ycinsight.com/yc-application-deadlines-2026 - [ ] 1-minute founder video uploaded to YouTube or Vimeo, unlisted, link included in application - [ ] Pitch deck in PDF format, 10 to 12 slides, uploaded to application - [ ] CCT model preprint PDFs (three sole-authored) attached or linked - [ ] TOPOLOGIX benchmark results summary (one page) attached - [ ] Neurocascade test results summary (one page) attached - [ ] Letter of endorsement from Kent Berridge or Samuel Gershman (optional but recommended) - [ ] Proof of enrollment at Hasso Plattner Institute for M.Sc. Digital Health (Winter 2026) - [ ] ORCID profile updated with all preprints and publications - [ ] GitHub repository for TOPOLOGIX made public or accessible with readme EDITOR NOTES - Eligibility risk: Y Combinator typically funds companies, not individual researchers. The application must clearly frame the venture as a pre-seed company with a product, not a research project. The CCT model and TOPOLOGIX platform must be presented as commercial products, not academic outputs. - Fact verification needed: Confirm that the three sole-authored preprints are indeed under review at the named journals (IART, PNPBP, NBR) and that the co-authored paper is under review at Alcohol, Elsevier. If any have been rejected or withdrawn, update the application accordingly. - Gap to fill: The profile does not mention any named pharma partners. The strategy notes reference named pharma partners, but no specific company names are provided. The applicant must insert at least one real pharma partner name with a letter of intent or a signed pilot agreement before submission. Without this, the traction section is weak for Y Combinator standards. - Sole founder concern: The application should explicitly address how the founder will handle the operational load of a company while enrolled in the M.Sc. program at HPI. A plan for the first hire or a co-founder search timeline should be included in the pitch deck or the short essays. - Market size data: The addiction therapeutics market size of over 10 billion dollars annually should be cited with a source. The applicant should verify this number from a reputable market research report and include the citation in the pitch deck or the motivation letter.