BriX: a database of protein building blocks for structural analysis, modeling and design

Abstract

High-resolution structures of proteins remain the most valuable source for understanding their function in the cell and provide leads for drug design. Since the availability of sufficient protein structures to tackle complex problems such as modeling backbone moves or docking remains a problem, alternative approaches using small, recurrent protein fragments have been employed. Here we present two databases that provide a vast resource for implementing such fragment-based strategies. The BriX database contains fragments from over 7000 non-homologous proteins from the Astral collection, segmented in lengths from 4 to 14 residues and clustered according to structural similarity, summing up to a content of 2 million fragments per length. To overcome the lack of loops classified in BriX, we constructed the Loop BriX database of non-regular structure elements, clustered according to end-to-end distance between the regular residues flanking the loop. Both databases are available online (http://brix.crg.es) and can be accessed through a user-friendly web-interface. For high-throughput queries a web-based API is provided, as well as full database downloads. In addition, two exciting applications are provided as online services: (i) user-submitted structures can be covered on the fly with BriX classes, representing putative structural variation throughout the protein and (ii) gaps or low-confidence regions in these structures can be bridged with matching fragments.

Figure 1.

The BriX database. (A) Number of BriX classes for lowest class thresholds per length. A peak in the number of classes can be observed at fragment length 7 and class threshold 0.5 Å. (B) Increase in the number of classified fragments from the first version of BriX (22) to the current version. (C) BriX classes with class thresholds varying from 0.5 to 1.0 Å RMSD for fragments of length 7. The class threshold indicates the compactness and structural homogeneity of the class, with lower thresholds causing classes to be more compact than higher thresholds.

Figure 2.

The Loop BriX database. (A) Example of a superclass containing three subclasses. The superclass contains fragments with end-to-end distance around 11.78 Å RMSD and two β-strand anchor residues. At the subclass level, fragments with similar length and backbone are grouped (length 7 for subclass 2219 and length 13 for subclass 2220 and 2221, superposition threshold of 1 Å). (B) Number of superclasses (blue) and subclasses (red) per class size, distributed in bins. In general, classes from Loop BriX are less populated than classes from BriX.

Figure 3.

The BriX website (http://brix.crg.es) (A) An overview of the class level with secondary structure content and sequence and structure logos per entry. (B) A panel on the class level where a user can filter on length, threshold, sequence and secondary structure content. Similar panels are implemented at every level of the class hierarchy. (C) BriX contains two levels: the class level and the fragment level. (D) Loop BriX contains three levels: the superclass, the subclass and the fragment level.

Figure 4.

BriX applications: ‘covering’ and ‘bridging’. (A) Covering: an input PDZ structure (PDB: 2WL7) is shown for which the algorithm finds matching structural fragments for the β-strand starting at residue 112 in chain A (red). The algorithm returns a set of protein fragment structures (green) superposed on the β-strand, together with structure and sequence logos. (B) Bridging: the same PDB structure (PDB: 2WL7), now with a missing loop. The algorithm finds loop fragments that match the regular anchor residues 104 and 112 spanning the loop with the same end-to-end distance (green).