Loading
Loading
About the Project
What would it mean to sequence the DNA of the world's unsolved problems? You'd find that a snakebite researcher in India and a WHO epidemiologist looking at the same crisis see different diseases. You'd find that a sensor failing in a Brazilian pasture shares structural DNA with a sensor failing on an ocean buoy. You'd find that the same regulatory mismatch blocks progress in digital therapeutics and traditional medicine.
The Problem Genome Project is a living collection of 500 unsolved problems across 19 domains, each tagged with eight structural genes. The patterns that emerge reveal something no individual expert could see: the deep structural repetitions that connect fields that have never heard of each other.
Method
Like the Art Genome Project (Artsy), we tag problems with structured attributes — “genes” — that reveal deep cross-domain connections no individual expert would notice.
The collection begins with expert sources: AI agents process research papers, NSF solicitations, patent filings, and standards roadmaps, extracting unsolved problems along with their constraints and context. Each problem is then tagged across eight structural dimensions — constraint type, domain, scale, failure mode, breakthrough needed, stakeholder complexity, temporal dynamics, and tractability — creating a multidimensional fingerprint.
Once tagged, problems in unrelated fields that share structural DNA become visible. An energy-sector materials crisis and a global health equity challenge turn out to carry the same regulatory mismatch, the same stakeholder complexity, the same worsening trajectory. These cross-domain patterns are exactly the kind of structural repetition where creative solutions originate.
Classification
Eight categories of structured attributes classify each problem, creating a multidimensional map of the problem space.
How to read a specimen
Each specimen is a portrait of a problem's structural DNA. Eight concentric layers encode the genome — from the domains at the core to the temporal dynamics at the halo. No two specimens look alike, but problems that share structural DNA look like relatives. Here are two specimens read side by side.
| Layer | What you see | Tags |
|---|---|---|
| Center | 3 filled circles — a larger cluster | energy, environment, manufacturing |
| Inner ring | 5 dots in a band | national (5 = national scale) |
| Inner ring | 1 short arc | multi-institution |
| Mid spines | 3 radial spines, each with a different end marker | regulatory, technical, economic |
| Mid arcs | 2 dashed arcs | regulatory-mismatch, not-attempted |
| Outer petals | 3 petal ellipses | process, materials, policy |
| Outer markers | 1 marker | proof-of-concept |
| Halo | 1 dashed ring | worsening |
| Layer | What you see | Tags |
|---|---|---|
| Center | 1 single dot — compact center | health |
| Inner ring | Dense band — 8 + 3 dots in two clusters | global (8) + community (3) |
| Inner ring | 1 short arc | multi-institution |
| Mid spines | 4 radial spines — visibly more "spiny" | behavioral, equity, infrastructure, economic |
| Mid arcs | 3 dashed arcs, more coverage around the ring | wrong-stakeholder, ignored-context, adoption-barrier |
| Outer petals | 4 petal ellipses — a fuller corona | design, systems-redesign, cost-reduction, behavior-change |
| Outer markers | 1 marker | design-proposal |
| Halo | 1 dashed ring | worsening |
Wind blade has a 3-dot cluster (multi-domain problem spanning energy / environment / manufacturing); cervical cancer has a single tight dot (pure health domain). Multi-domain problems look more “nucleated.”
The cervical cancer brief shows a denser blue dot band because it operates at two scales (global + community = 11 dots vs. 5 for national alone). More dots = more scale complexity.
Cervical cancer has 4 spines radiating out vs. 3 — visually “spikier,” reflecting that it faces more categories of constraint (including behavioral and equity, which the engineering problem doesn’t have).
The health brief has more amber arcs and more violet petals, making the mid-to-outer rings denser and more layered. This reflects a problem with more documented failure modes and more types of breakthroughs needed — a hallmark of complex socio-technical problems vs. primarily technical ones.
Both have identical stakeholder arcs (multi-institution), temporal halos (worsening), and single tractability markers. These shared structural genes are exactly the kind of cross-domain pattern PGP is designed to surface — an energy-sector materials problem and a global health equity problem that share the same stakeholder complexity, temporal urgency, and tractability level.
Origins
Problem Genome Project was created by Beth Altringer Eagle and originally funded by the Harvard Initiative for Learning and Teaching (HILT) in 2013–2014. It was designed to collect and make accessible basic information from top experts on important problems, such as the Grand Challenges in engineering or the Millennium Development Goals. The idea came from seeing aspiring entrepreneurs (often but not exclusively students) repeat the same entry-level research on the problem areas that interested them.
The specific catalyst moment was when a student team in my class at the time that was working on reinventing the syringe to make it safer and cheaper had a half hour call with an expert from the WHO who told them he was currently looking at his bookshelf in his office and it was littered with most of the designs they were in the process of inventing. The syringe is already very cheap and supported by a global system that will not easily transition to a new solution. You need a better problem, and he outlined one for them. They pivoted to a much better project working on the vaccine cold chain.
PGP is about making the equivalent of that call accessible to more people embarking on a significant project so they can build atop what came before. Having this information ready at the outset of their work could help them get on with the more interesting questions much faster. Interviewing experts one at a time, however, for this knowledge was an idea ahead of its time.
Fast-forward to 2025 at Dartmouth, and new technologies and funding from Natural Artificial Labs in the Thayer School of Engineering at Dartmouth College make the original idea possible.