Structome: a tool for the rapid assembly of datasets for structural phylogenetics

Area of Science:

• Structural biology
• Bioinformatics
• Evolutionary biology

Background:

• Protein structures provide insights into evolutionary relationships.
• Reconstructing evolutionary histories often relies on structural information.
• Existing methods can be time-consuming for structure-informed phylogenetic analysis.

Purpose of the Study:

• To develop a web server, Structome, for rapid identification of similar protein structures.
• To facilitate the assembly of datasets for structure-based phylogenetics.
• To expedite the process of structure-informed evolutionary inference.

Main Methods:

• Clustering protein structures from the RCSB PDB database based on 90% sequence identity.
• Representing each cluster by a centroid structure.
• Calculating structure similarity between centroid proteins and integrating PDB, SCOP, and CATH annotations.

Main Results:

• Structome successfully identifies diverse protein structures, including those with sequence and structural variations.
• The tool provides pre-computed distance matrices for analyzing evolutionary relationships.
• Analysis using an H3 histone query demonstrated Structome's ability to capture evolutionary diversity.

Conclusions:

• Structome enables rapid generation of structural neighbor datasets from a single query structure.
• The web server aids in studying the deep evolutionary history of proteins.
• Structome enhances structure-based phylogenetic analysis by overcoming limitations of sequence similarity searches.