Abstract

The functional properties of a protein primarily depend on its three-dimensional (3D) structure. These properties have classically been assigned, visualized and analysed on the basis of protein secondary structures. The β-turn is the third most important secondary structure after helices and β-strands. β-turns have been classified according to the values of the dihedral angles φ and ψ of the central residue. Conventionally, eight different types of β-turns have been defined, whereas those that cannot be defined are classified as type IV β-turns. This classification remains the most widely used. Nonetheless, the miscellaneous type IV β-turns represent 1/3rd of β-turn residues. An unsupervised specific clustering approach was designed to search for recurrent new turns in the type IV category. The classical rules of β-turn type assignment were central to the approach. The four most frequently occurring clusters defined the new β-turn types. Unexpectedly, these types, designated IV1, IV2, IV3 and IV4, represent half of the type IV β-turns and occur more frequently than many of the previously established types. These types show convincing particularities, in terms of both structures and sequences that allow for the classical β-turn classification to be extended for the first time in 25 years.

Highlights

  • The functional properties of a protein primarily depend on its three-dimensional (3D) structure

  • The β-turn is the third most important secondary structure after helices and β-strands. β-turns have been classified according to the values of the dihedral angles φ and ψ of the central residue

  • An unsupervised specific clustering approach was designed to search for recurrent new turns in the type IV category

Read more

Summary

Methods

Protein Blocks (PBs62,63) corresponded to a set of 16 local prototypes, labelled from a to p, of 5 residue length that were described on the basis of dihedral angles (φ, ψ). A specific clustering approach was designed to cluster type IV β-turns by using the classical rule, allowing +/− 30° for all angles, with the exception of one at +/− 45° for the defined values. The training was carried out in 2 successive parts; the first one limited the potential bias of initialization, and the second refined the clustering by using the specific rules for β-turn types. 2. One of the T type IV β-turns was randomly selected from the dataset D (denoted V2) and compared with each of the k clusters. Most of the quantitative analysis was performed using in-house Python scripts, and statistics and visualization were performed with R software (version 3.2.2)[70]

Results and Discussion
AL BCDEFghiKMNoP bCFK ADIjLMnoP
Ivlar iVLmAfc ivLay vla
Additional Information

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.