No one-size-fits-all best approach to zero-shot or few-shot protein fitness prediction

Summary

There is currently no clear best method for zero-shot or few-shot protein fitness prediction tasks (1). This was tested using ESM, CARP, and the autoregressive Dayhoff model.

Figures

Ref (1)

See also

• No one-size-fits-all best approach to motif scaffolding protein design

1.

Didi K, Alamdari S, Lu AX, Wittmann B, Johnston KE, Amini AP, et al. FLIP2: Expanding Protein Fitness Landscape Benchmarks for Real-World Machine Learning Applications. openRxiv; 2026. Available from: https://doi.org/10.64898/2026.02.23.707496