A graphic and command line protocol for quick and accurate comparisons of protein and nucleic acid structures with US-align.

Abstract

With the success of structural biology and the advancements in deep-learning-based structure predictions, rapid and accurate structural comparisons among macromolecular structures have become increasingly important in structural bioinformatics. US-align is a highly efficient, versatile, open-source program for sequential and nonsequential structure comparisons of proteins, RNAs and DNAs in pairwise and multiple alignment forms and applicable to both monomeric and multimeric complex structures. The core algorithm of US-align is built on a highly optimized, iterative superimposition and dynamic programming alignment process, guided with a unified and sequence length-independent scoring function, TM-score. The unique design of US-align not only ensures its high accuracy and speed compared with other state-of-the-art methods designed for specific alignment tasks but also makes it the only protocol that can be applied to multiple alignment tasks and allow a structural comparison across different molecular types, the latter of which is critical for template-based heteromolecular structure prediction and function annotations. Here we describe how to install and effectively utilize US-align as a command line tool, as an online web server, and as a plugin to commonly used molecular graphic systems such as PyMOL. US-align installation takes a few minutes to setup, while the actual alignment implementation can be completed typically within 1 s.