Prediction of nearest neighbor effects on backbone torsion angles and NMR scalar coupling constants in disordered proteins

Protein Sci. 2018 Jan;27(1):146-158. doi: 10.1002/pro.3292. Epub 2017 Oct 25.

Abstract

Using fine-tuned hydrogen bonding criteria, a library of coiled peptide fragments has been generated from a large set of high-resolution protein X-ray structures. This library is shown to be an improved representation of ϕ/ψ torsion angles seen in intrinsically disordered proteins (IDPs). The ϕ/ψ torsion angle distribution of the library, on average, provides good agreement with experimentally observed chemical shifts and 3 JHN-Hα coupling constants for a set of five disordered proteins. Inspection of the coil library confirms that nearest-neighbor effects significantly impact the ϕ/ψ distribution of residues in the coil state. Importantly, 3 JHN-Hα coupling constants derived from the nearest-neighbor modulated backbone ϕ distribution in the coil library show improved agreement to experimental values, thereby providing a better way to predict 3 JHN-Hα coupling constants for IDPs, and for identifying locations that deviate from fully random behavior.

Keywords: 3JHN-Hα; IDP; NMR; Ramachandran map; coil library; scalar coupling.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Intrinsically Disordered Proteins / chemistry*
  • Intrinsically Disordered Proteins / genetics
  • Nuclear Magnetic Resonance, Biomolecular*
  • Peptide Library*
  • Protein Structure, Secondary

Substances

  • Intrinsically Disordered Proteins
  • Peptide Library