Homology detection using a protein secondary structure-based large language model
Roman Kogay, Weicheng Ma, Jad Bousselham, Zechen Yang, Daniel N. Rockmore, Olga Zhaxybayeva, Soroush Vosoughi
The paper introduces ProSSL, a Transformer-based model using protein secondary structure to detect homology, outperforming traditional sequence-based methods in identifying distant homologs.
bioRxiv · Dec 19, 2023 · PreprintBioinformatics