Abstract
Short hydrogen bonds with the heteroatom distances below 2.7 Å are widely observed in proteins. Due to their compact structures, short hydrogen bonds have strong covalent nature arising from the quantum delocalization of both the electrons and protons. From a statistical analysis of the top 1% highest quality biomolecular structures in the Protein Data Bank, we have characterized the structural and chemical features of short hydrogen bonds that form from amino acids and ligands, and developed a machine learning model that effectively predicts the presence of short hydrogen bonds between amino acids in proteins. We have then combined quantum chemistry calculations, energy decomposition analysis and first principles simulations to unravel how the quantum nature of short hydrogen bonds results in unique proton behavior and distinctive downfield 1H NMR chemical shifts.
Original language | English (US) |
---|---|
Title of host publication | Comprehensive Computational Chemistry, First Edition |
Subtitle of host publication | Volume 1-4 |
Publisher | Elsevier |
Pages | V4-735-V4-754 |
Volume | 4 |
ISBN (Electronic) | 9780128219782 |
DOIs | |
State | Published - Jan 1 2023 |
All Science Journal Classification (ASJC) codes
- General Chemistry
Keywords
- ab initio molecular dynamics
- ab initio path integral molecular dynamics
- Electronic quantum effects
- Enzyme catalysis
- Hydrogen bonds
- Low-barrier hydrogen bonds
- Machine learning model
- Nuclear quantum effects
- Protein structure
- Proton delocalization
- Short hydrogen bonds