ProteinGym mutation effects

Placeholder. This benchmark is here so the benchmark panel can be tried out. Its baseline is from the blueprint or the literature as of 2026-10-19 and must be checked again; every route and score after it is made up.

Task

Predict how mutations change a protein’s fitness, scored against the 217 deep mutational scanning assays of ProteinGym v1.3: enzyme activity, binding, expression, organismal fitness and stability. The score is the mean Spearman correlation over assays; the baseline is the S2F sequence–structure model at 0.4556, as in AutoSOTA’s report. Routes may use multiple sequence alignments, structures and any public pretrained protein language model.

How to take part

Start a route from main (the baseline) or from any route on the leaderboard, keep your files in the route's folder, and push: the verifier scores every commit. Say what your route changes with tags, and keep the misses: a route that scores worse is a result too.