Summary
Structure-derived representations (embeddings) outperform sequence-derived representations on protein alignment (1). They also outperform structure tokenization methods such as alignment using Foldseek alphabets.
See also
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