Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: Analysis of substructural variation in families of enzymatic proteins with applications to protein function prediction

Figure 1

Substructural Clusters (SCs) for the heme-dependent peroxidases. (a) Superposition of the propagated motifs for the animal and non-animal heme-dependent peroxidases of EC 1.11.1.7 demonstrates geometric diversity. The color of each aligned substructure corresponds to its cluster assignment in (c), and it can be seen that closely aligned substructures in (a) correspond to co-located points in the SCs shown in (c). (b) When the backbones of a class II fungal peroxidase [PDB:1ARU] and human myeloperoxidase [PDB:1CXP] are compared, substructural similarity within the heme-binding catalytic site region is evident, but the remainder of the enzyme structures can be seen to have significant topological differences and are assigned to separate topological classes within the CATH structural ontology [54]. (c) Applying FASST to the family of peroxidases yields a family-wise geometric feature vector for each catalytic substructure in the family, reducing each substructure shown in (a) to a point in the SCs. Gaussian mixture model (GMM) clustering of geometric feature vectors, projected onto a space of reduced dimension, identifies four clusters denoted by color. The gray isocontours show the smoothed density of substructures in each part of the SCs. (d) Substructure positions in the SCs colored by Family-level taxanomic classification reveal that phylogenetic distance between proteins is the main source of substructural diversity among the heme-dependent peroxidase binding sites. The open/closed plot characters correspond to apo/holo structures, respectively.

Back to article page