Login / Signup

Pseudopotentials for coarse-grained cross-link-assisted modeling of protein structures.

Mateusz KogutZhou GongChun TangJozef Adam Liwo
Published in: Journal of computational chemistry (2021)
Pseudopotentials for the chemical cross-links comprising the glutamic- and aspartic-acid side chains bridged with adipic- (ADH) or pimelic-acid hydrazide (PDH), and the lysine side chains bridged with glutaric (BS2 G) or suberic acid (BS3 ) for coarse-grained cross-link-assisted simulations were determined by canonical molecular dynamics with the Amber14sb force field. The potentials depend on the distance between side-chain ends and on side-chain orientation, this preventing from making cross-link contacts across the globule in simulations. The potentials were implemented in the UNRES coarse-grained force field and their effect on the quality of models was assessed with 11 monomeric and 1 dimeric proteins, using synthetic or experimental cross-link data. Simulations with the new potentials resulted in improvement of the generated models compared to unrestrained simulations in more instances compared to those with the statistical potentials.
Keyphrases
  • molecular dynamics
  • density functional theory
  • single molecule
  • electronic health record
  • high resolution
  • amino acid
  • big data
  • artificial intelligence
  • molecular dynamics simulations
  • binding protein