Home › Hard problems › Science and maths

Science and maths

Predicting what a protein does or binds to

Solo: possibleOne person can test a narrow slice of it.

SEQUENCEa chain of amino acidsfoldSTRUCTURElargely solvedSTILL OPENWhat does it do?its job in the cell?What does it bind to?which molecules fit?Will a design work?in a real lab?
Predicting a protein's shape is largely solved; predicting what it does, what it binds to, and which designs work in a lab is not.

Why it is still open

Structure prediction (AlphaFold) is largely solved; function, binding and design that works in a lab are not.

An experiment one person can run in 8 weeks

Use open protein models (ESM) to rank variants on the public ProteinGym benchmark and test one new idea.

Verdict: Solo: possible

Fine on public benchmarks; lab proof needs partners.

The solo test behind the verdict

  1. Can it be measured? There is a number that shows you did better, such as an error rate, a catch rate or a cost per answer.
  2. Is it cheap to try? It runs on a laptop, free open models and open data, or under about ₹50,000 of rented GPU time.
  3. Is there a public yardstick? A benchmark or dataset others use, so your result can be compared and believed.
  4. Would a small win get noticed? A lab, a funder or a customer would care about a 10% improvement, not only a full solution.

Four yeses means Solo: strong. Two or three means Solo: possible. Fewer means Follow, don't build.

PreviousWeather beyond two weeksNextPredicting drug toxicity early

All 79 hard problems · Working on this, or know of a result we missed? basant@cleandataengineering.ai