Abstract:We describe a probabilistic model for deriving, from the database of assigned chemical shifts, a set of random coil chemical shift values that are "unbiased" insofar as contributions from detectable secondary structure have been minimized (RCCSu). We have used this approach to derive a set of RCCSu values for 13Calpha and 13Cbeta for 17 of the 20 standard amino acid residue types by taking advantage of the known opposite conformational dependence of these parameters. We present a second probabilistic approach that utilizes the maximum entropy principle to analyze the database of 13Calpha and 13Cbeta chemical shifts considered separately; this approach yielded a second set of random coil chemical shifts (RCCSmax-ent). Both new approaches analyze the chemical shift database without reference to known structure. Prior approaches have used either the chemical shifts of small peptides assumed to model the random coil state (RCCSpeptide) or statistical analysis of chemical shifts associated with structure not in helical or strand conformation (RCCSstruct-stat). We show that the RCCSmax-ent values are strikingly similar to published RCCSpeptide and RCCSstruct-stat values. By contrast, the RCCSu values differ significantly from both published types of random coil chemical shift values. The differences (RCCSpeptide - RCCSu) for individual residue types show a correlation with known intrinsic conformational propensities. These results suggest that random coil chemical shift values from both prior approaches are biased by conformational preferences. RCCSu values appear to be consistent with the current concept of the "random coil" as the state in which the geometry of the polypeptide ensemble samples the allowed region of (phi, psi)-space in the absence of any dominant stabilizing interactions and thus represent an improved basis for the detection of secondary structure. Coupled with the growing database of chemical shifts, this probabilistic approach makes it possible to refine relationships among chemical shifts, their conformational propensities, and their dependence on pH, temperature, or neighboring residue type.

Protein secondary structure prediction using NMR chemical shift data

Accurate Prediction of Chemical Shifts for Aqueous Protein Structure for "Real World" Cases using Machine Learning

Prediction of Protein 13cα NMR Chemical Shifts Using a Combination Scheme of Statistical Modeling and Quantum-Mechanical Analysis

HSQC2STRUC: A Machine Learning Model for Protein Secondary Structure Prediction using Unassigned NMR Spectra

Multiple Linear Regression for Protein Secondary Structure Prediction.

EFG-CS: Predicting chemical shifts from amino acid sequences with protein structure prediction using machine learning and deep learning models

A new method on reconstructing protein structure from NOESY distances

AFNMR: Automated Fragmentation Quantum Mechanical Calculation of NMR Chemical Shifts for Biomolecules

Protein structure prediction: Combining de novo modeling with sparse experimental data

Building native protein conformation from NMR backbone chemical shifts using Monte Carlo fragment assembly.

Scaled Alternating Steepest Descent Algorithm Applied for Protein Structure Determination from Nuclear Magnetic Resonance Data.

Protein Chemical Shift Prediction

Consistent blind protein structure generation from NMR chemical shift data

Predicting protein secondary structure and solvent accessibility with an improved multiple linear regression method.

Rapid Quantification of Protein Secondary Structure Composition from a Single Unassigned 1D 13C Nuclear Magnetic Resonance Spectrum

Rapid Quantification of Protein Secondary Structure Composition from a Single Unassigned 1D 13C NMR Spectrum

Protein Structure Determination Using a Riemannian Approach

CS23D: a web server for rapid protein structure generation using NMR chemical shifts and sequence data

Improving prediction of protein secondary structure, backbone angles, solvent accessibility and contact numbers by using predicted contact maps and an ensemble of recurrent and residual convolutional neural networks

Probabilistic approach to determining unbiased random-coil carbon-13 chemical shift values from the protein chemical shift database

Quantum Calculation of Protein Nmr Chemical Shifts Based on the Automated Fragmentation Method