Tools for integrated sequence-structure analysis with UCSF Chimera

Abstract

Background: Comparing related structures and viewing the structures in the context of sequence alignments are important tasks in protein structure-function research. While many programs exist for individual aspects of such work, there is a need for interactive visualization tools that: (a) provide a deep integration of sequence and structure, far beyond mapping where a sequence region falls in the structure and vice versa; (b) facilitate changing data of one type based on the other (for example, using only sequence-conserved residues to match structures, or adjusting a sequence alignment based on spatial fit); (c) can be used with a researcher's own data, including arbitrary sequence alignments and annotations, closely or distantly related sets of proteins, etc.; and (d) interoperate with each other and with a full complement of molecular graphics features. We describe enhancements to UCSF Chimera to achieve these goals.

Results: The molecular graphics program UCSF Chimera includes a suite of tools for interactive analyses of sequences and structures. Structures automatically associate with sequences in imported alignments, allowing many kinds of crosstalk. A novel method is provided to superimpose structures in the absence of a pre-existing sequence alignment. The method uses both sequence and secondary structure, and can match even structures with very low sequence identity. Another tool constructs structure-based sequence alignments from superpositions of two or more proteins. Chimera is designed to be extensible, and mechanisms for incorporating user-specific data without Chimera code development are also provided.

Conclusion: The tools described here apply to many problems involving comparison and analysis of protein structures and their sequences. Chimera includes complete documentation and is intended for use by a wide range of scientists, not just those in the computational disciplines. UCSF Chimera is free for non-commercial use and is available for Microsoft Windows, Apple Mac OS X, Linux, and other platforms from http://www.cgl.ucsf.edu/chimera.

Figure 1

Protein structure and associated sequence alignment. The structure of proclavaminate amidino hydrolase [PDB:1gq6] and the seed alignment from Pfam [33] for its family. Only a portion of the alignment is depicted. One chain of the trimer, shown as a ribbon, is associated with the first sequence in the alignment. The other two chains are shown as light blue backbone traces. The ribbon is colored from blue to pink to yellow with increasing conservation. Gray ribbon segments represent residues in columns for which conservation was not calculated due to a high proportion of gaps (0.5 or higher). Conservation values, also shown as histogram bars above the alignment, were obtained with the entropy-based measure in AL2CO [8] and "independent counts" weighting. Active site ions are shown in cyan. Residues in the structure that are completely conserved in the alignment have their side chains displayed and are indicated with blue boxes on the alignment. The part of the alignment that is shown includes conserved (boxed) ion-binding residues in a loop to the lower right of the ions. Highly conserved residues are primarily involved in the active site or inter-subunit interactions. Residues in the protein core are moderately conserved. In the sequence alignment, Clustal X coloring is used for the residue one-letter codes. The alignment part of the figure was exported directly from Multalign Viewer as Encapsulated PostScript.

Figure 2

Comparison of matched orientations (pair 8). Comparison of matched orientations from MatchMaker (default settings) and CE [26] (web server [34] with default settings). Each matched structure [PDB:4fgf] is rainbow-colored from blue at the N-terminus to red at the C-terminus. The reference structure [PDB:1tie] is shown in tan on the right.