Climate biotech
Computational biology for decarbonization.
Stoma builds computational tools that help biologists prioritize which genes in unannotated marine genomes are worth testing in the lab — starting with the nitroplast — while building toward a narrow, long-term engineering target in industrial nitrogen fixation.
Approach
Decarbonization is a biology problem, not an energy problem.
Industrial nitrogen fixation, carbon capture in marine systems, and a long list of other decarbonization-relevant processes are bottlenecked by biological knowledge, not engineering throughput. The organisms that already do this work in nature — non-model marine eukaryotes, symbiotic systems, lineages with no close relative in any reference database — are largely unannotated. Most of their genes have no known function. That gap is the actual constraint.
Most computational tools built for this kind of work inherit assumptions from drug-discovery pipelines: well-characterized model organisms, dense reference data, large internal datasets. Non-model marine biology has none of that. Closing the gap means building for the data scarcity directly, not adapting tools built for a different problem.
Stoma is structured around two linked steps: a platform that makes a finite amount of wet-lab time go further, and a narrow engineering commitment that the platform's findings are meant to eventually inform.
Platform — Engine 1
A ranked list of what's worth testing next.
The platform takes an unannotated marine genome and produces a prioritized hypothesis list: which genes are worth a wet-lab biologist's time, and why. Function class assignment, structural homology, interaction-partner prediction, transit peptide targeting, and regulatory inference feed into a single ranked output, rather than sitting as separate, disconnected tools.
- Function class assignment
- Structural homology
- Interaction partner prediction
- Transit peptide targeting
- Regulatory inference
Hero case: the nitroplast
The nitroplast — the nitrogen-fixing, organelle-like structure recently described in the marine alga Braarudosphaera bigelowii, derived from its endosymbiont UCYN-A — is the platform's current proving ground. Thousands of genes in this system carry no assigned function. It's exactly the kind of system the platform is built for: novel, data-poor, and biologically consequential.
What the output actually is
Every output is a ranked hypothesis with a confidence score — not a verified function. This is a method and architecture demonstration, not a calibrated accuracy claim. It tells you where to look first. The wet lab still does the proving.
Engineering — Engine 2
One target, chosen with a co-founder — not before.
The platform's job is to inform a real engineering commitment, not stand in for one. Stoma has deliberately deferred selecting that target until a technical co-founder with structural biology and protein engineering depth is in place — target selection is a scientific decision, and it should be made by the scientist who will own it.
Whichever target is chosen, the commitment is to one target for five to seven years. Re-scoping requires scientific evidence that it's intractable, not a more interesting opportunity next door.
Candidate directions — not yet committed
Transit peptide design
Routing nitroplast-style imports into industrial host organisms.
Oxygen-tolerant nitrogenase
Variants that fix nitrogen without the strict anaerobic constraints of natural systems.
Longer horizon (10–20 years): engineered nitrogen fixation in industrial yeast, or synthetic host-symbiont systems for biomanufacturing. The application is determined by where the science leads, not fixed in advance.
Why both
The coupling is the company.
A platform alone is a force multiplier with no guarantee it ever becomes more. Engineering alone is a slow, capital-intensive bet with no early revenue. Running both, on purpose, from day one, is the actual structure — not a hedge between two strategies.
- 01
Customer data flows back into the platform.
Data agreements that don't allow this are rejected, even at the cost of revenue.
- 02
The engineering target stays narrow.
One target, five to seven years. Re-scoping requires evidence of intractability, not a more interesting adjacent opportunity.
- 03
Engine 2 is staffed from day one.
A permanent line item from the start, not something the platform has to “earn the right” to build later.
The most common failure mode in this structure is what we call eyes-shine drift: as the platform produces visible wins, organizational gravity pulls focus toward platform optimization and away from the engineering program's slower, harder-to-measure work. Five years in, a company can end up running the platform-only strategy without ever deciding to. These three commitments, plus explicit founder discipline, are the defense against that.
Team
Built solo, so far.
Stoma is currently a single founder with an operator background, working on the platform and the relationships the engineering side will need. That's a deliberate, temporary state, not a long-term plan — a structure like this needs a scientific co-founder before the commitments above mean anything.
What's missing is someone to choose and own Engine 2's target, not just advise on it.
What we're looking for
- PhD-level scientific credibility
- Structural biology + protein engineering depth
- Wet-lab fluency or a strong wet-lab network
- Conviction in the dual-engine structure above
Get in touch
Three reasons to write in.
hello@stoma.bio