Evaluating Protein Language Model Embeddingsfor Structural Similarity in the Protein-Sequence Twilight Zone
Shivaram Danwada, Priscilla Udomprasert, Nilanjana Raychawdhary, Cheryl Seals, Lei Wu, Sutanu Bhattacharya
This study evaluates protein language model embeddings for estimating structural similarity in the protein-sequence twilight zone and finds that ProstT5 performs best.
bioRxiv · Jul 31, 2026 · PreprintBioinformatics