Related Experiment Videos
To be folded or to be unfolded?
Sergiy O Garbuzynskiy1, Michail Yu Lobanov, Oxana V Galzitskaya
1Institute of Protein Research, Russian Academy of Sciences, 142290 Pushchino, Moscow Region, Russia.
Protein Science : a Publication of the Protein Society
|October 23, 2004
Summary
Intrinsic amino acid properties predict protein structure. The number of residue contacts in a globular state can identify natively unfolded proteins with 89% accuracy, distinguishing them from folded proteins.
Area of Science:
- Protein structure and dynamics
- Biophysics
- Computational biology
Background:
- Natively unfolded proteins lack fixed three-dimensional structure under physiological conditions.
- The intrinsic properties of amino acid residues are investigated as potential determinants of this lack of structure.
Purpose of the Study:
- To determine if intrinsic amino acid residue properties can predict whether a protein will be natively unfolded or folded.
- To identify simple indicators for natively unfolded proteins based on amino acid sequence.
Main Methods:
- Calculating the expected average number of contacts per residue from amino acid sequences.
- Utilizing hydrophobicity as a predictive parameter.
- Employing a Monte Carlo algorithm to optimize artificial parameters for distinguishing protein states.
Main Results:
- The expected average number of contacts per residue accurately predicts natively unfolded proteins with 89% accuracy.
- Hydrophobicity achieved 83% prediction accuracy.
- An optimal set of artificial parameters yielded a maximum prediction accuracy of 95%.
Conclusions:
- The propensity of amino acid residues to form contacts in a globular state is a key determinant of protein folding.
- Simple sequence-based calculations can serve as effective indicators for identifying natively unfolded proteins.
- Predictive models based on amino acid properties offer high accuracy in distinguishing folded from unfolded protein states.