This paper introduces two deep learning methods, TM-Vec and DeepBLAST, for protein remote homology detection and structural alignment. TM-Vec efficiently searches for structure-structure similarities in large sequence databases by predicting TM-scores from sequence pairs. DeepBLAST structurally aligns proteins using only sequence information, outperforming traditional methods. The authors demonstrate improved remote homology identification compared to state-of-the-art methods.
Publisher
Nature Biotechnology
Published On
Jun 01, 2024
Authors
Tymor Hamamsy, James T. Morton, Robert Blackwell, Daniel Berenberg, Nicholas Carriero, Vladimir Gligorijevic, Charlie E. M. Strauss, Julia Koehler Leman, Kyunghyun Cho, Richard Bonneau
Tags
deep learning
protein homology
structural alignment
TM-Vec
DeepBLAST
sequence databases
protein structures
Related Publications
Explore these studies to deepen your understanding of the subject.