ConSole: using modularity of contact maps to locate solenoid domains in protein structures

Abstract

Background: Periodic proteins, characterized by the presence of multiple repeats of short motifs, form an interesting and seldom-studied group. Due to often extreme divergence in sequence, detection and analysis of such motifs is performed more reliably on the structural level. Yet, few algorithms have been developed for the detection and analysis of structures of periodic proteins.

Results: ConSole recognizes modularity in protein contact maps, allowing for precise identification of repeats in solenoid protein structures, an important subgroup of periodic proteins. Tests on benchmarks show that ConSole has higher recognition accuracy as compared to Raphael, the only other publicly available solenoid structure detection tool. As a next step of ConSole analysis, we show how detection of solenoid repeats in structures can be used to improve sequence recognition of these motifs and to detect subtle irregularities of repeat lengths in three solenoid protein families.

Conclusions: The ConSole algorithm provides a fast and accurate tool to recognize solenoid protein structures as a whole and to identify individual solenoid repeat units from a structure. ConSole is available as a web-based, interactive server and is available for download at http://console.sanfordburnham.org.

Figure 1

Contact maps of solenoid protein structures. Contact maps of two solenoid protein structures. A line d₂, parallel to the main diagonal, is clearly visible in both CMs. d₂ is almost fully continuous for the highly regular Ribonuclease Inhibitor (1DFJ—Chain I) and has some gaps in the Clathrin structure (1B89—Chain A), reflecting more variable interaction patterns between helices.

Figure 2

Solenoid contact patterns. (a) The correlation matrix M₁ determined for template matching the contact map of PDB 1DFJ-Chain I determined with the solenoid pattern (b). Bright regions indicate high correlation values [0;1], while dark regions indicate low [−1; 0] correlation values. (b) The solenoid pattern to generate M₁ and (c) the non-solenoid pattern to generate M₂ used for template matching. (d) Plot of the correlation coefficients determined for residue 75 in chain I of 1DFJ. Twenty correlation coefficients were collected around the main diagonal [(75,55); (75,95)] from M₁ and merged with 20 coefficients from M₂ from the same positions. (e) Coefficients determined for residue 81 in chain B of the globular structure from 1QGC. Correlation features determined for solenoid residues have smoother profiles compared to rather noisy features of globular proteins. Peak correlation values also differ for the solenoid/non-solenoid combinations, respectively.

Figure 3

Results of solenoid classification. Four solenoid protein structures taken from the benchmark dataset: (a) 1XKU-A, (b) 1QRL-A, (c) 1M8Z-A, and (d) 1K5C-A. All are colored according to the ConSole-based classification results. Residues correctly assigned to the solenoid class are colored green; residues correctly assigned to the non-solenoid class are colored red. Gray or yellow indicates all residues wrongly assigned to the non-solenoid or solenoid class, respectively. Figures for all results in the benchmark are available online.

Figure 4

Solenoid motif from structure alignment. (a) Automatically detected solenoid units of a Leucine Rich Repeat domain (1DFJ-I) with arbitrary solenoid unit coloring. The middle inset shows all units superimposed after multiple-structure alignment with POSA. The consensus motif determined for the structurally aligned units is displayed on the right. The next two rows display the same results for (b) an Armadillo repeat (1M8Z-A) and (c) an Ankyrin repeat (3B95-A).

Figure 5

An unusually flat LRR structure. (a) An unusually flat LRR (4FD0-A) found within the few solenoid structures that have not yet been mentioned in publications. (b) Structural alignment of individual solenoid units with phenylalanine residues colored in blue. (c) Sequences of the respective solenoid units where all phenylalanines are highlighted in blue and all leucine-like residues are highlighted in green. Below is the sequence motif determined by ConSole.

Figure 6

Unit-length irregularities in Leucine Rich Repeat structures. (a) The structure of the Ribonuclease Inhibitor (1DFJ-I) shows only minimal irregularity. The irregularity profile indicates variations in unit length λ_j (blue curve) and absolute deviation | λ – λ_j | from the mean unit length λ for each solenoid unit j (green curve). The two structural motifs at the bottom represent the two main structural solenoid unit motifs detected for 1DFJ. Possible positions of these structural motifs are indicated above the respective structure. (b) The LRR domain in CARMIL (A) resembles an unusually irregular Ribonuclease Inhibitor structure for which variations in the irregularity profile indicate variability in solenoid units. The structural segment of units 6 – 9 (dashed box) depicts irregularities in units in more detail. Each unit is colored according to the irregularity distribution in Figure 7. Gray segments were classified as insertions by ConSole and do not contribute to the irregularity calculation. (c) The structure of TLR4 (2Z64-A) has a distinctively more irregular region (units 5 – 10, dashed box) when compared to other segments of the structure. Irregularity in this region accrues a significant change in the curvature.

Figure 7

Distribution of unit-length irregularities over several solenoid families. Structures from the Leucine Rich Repeat, Ankyrin Repeat and Armadillo Repeat were sampled for irregularities. Displayed is the length irregularity distribution for each respective family, where the number of residues a solenoid unit length differed from λ.