Summary
Discrete structural tokens are worse than continuous ML-derived representations for calculating structural alignments (1). In this case, the tokens were from Foldseek and a related method, while the continuous representations were from a retrained ProteinMPNN encoder.
1.
Trinquier J, Petti S, Park S, Herath K, van Kempen M, Feng S, et al. SoftAlign: End-to-end protein structures alignment. openRxiv; 2025. Available from: https://doi.org/10.1101/2025.05.09.653096