Smotifs as structural local descriptors of supersecondary elements: classification, completeness and applications

Abstract

Protein structures are made up of periodic and aperiodic structural elements (i.e., α-helices, β-strands and loops). Despite the apparent lack of regular structure, loops have specific conformations and play a central role in the folding, dynamics, and function of proteins. In this article, we reviewed our previous works in the study of protein loops as local supersecondary structural motifs or Smotifs. We reexamined our works about the structural classification of loops (ArchDB) and its application to loop structure prediction (ArchPRED), including the assessment of the limits of knowledge-based loop structure prediction methods. We finalized this article by focusing on the modular nature of proteins and how the concept of Smotifs provides a convenient and practical approach to decompose proteins into strings of concatenated Smotifs and how can this be used in computational protein design and protein structure prediction.

Tracing the birth of structural domains from loops during protein evolution

“…The library of single domain protein structures is essentially complete 70 , and so is the library of loop prototypes, which has all geometries sampled 71 , and is considered saturated even for the case of long loops 72 . These properties not only guarantee an exhaustive phylogenomic exploration of the history of domains and loops but also enable the sequence-to-structure mapping of deep learning methods needed to solve the fold recognition problem in ab initio explorations.…”

“…Convergence towards the formation of the 'binding' cassette of the primordial ABC transporter required a P-loop-centered nucleation of only three loop structures (213, 80 and 186), which are relatively far away from each other in the extant sequence and structure. To test if convergence was resilient, we conducted reshuffling experiments based on the 34|80|186|213 loop sequence where we systematically replaced loop 80 in the second position by all possible loops (60,68,72,85,110,126 and 151) that would maintain sequence order, and separately loop 213 in the fourth position by the only option, loop 222 (note that loop 186 could not be replaced). We then modeled structures from the reshuffled sequences and compared them to the reference structure using pruned and total RMSD measurements of structural overlaps (Supplementary Fig.…”

Tracing the birth of structural domains from loops during protein evolution

“…The library of single domain protein structures is essentially complete 75 , and so is the library of loop prototypes, which has all geometries sampled 76 , and is considered saturated even for the case of long loops 77 . These properties not only guarantee an exhaustive phylogenomic exploration of the history of domains and loops but also enable the sequence-to-structure mapping of deep learning methods needed to solve the fold recognition problem in ab initio explorations.…”

Some insights into the transition state ensemble of the folding of globular proteins

“…In a similar vein, Bonet et al [3] propose a classification of fragments composed of two adjacent regular secondary structures and the loop connecting these stems. This analysis at the level of subdomain is somehow related to the notion of super secondary structures, and they call them Smotifs.…”