Related Experiment Videos
Understanding the recognition of protein structural classes by amino acid composition
I Bahar1, A R Atilgan, R L Jernigan
1Molecular Structure Section, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892-5677, USA.
Proteins
|November 5, 1997
Summary
Amino acid composition alone can predict protein structural class with 81% accuracy. This is driven by how amino acids interact, influencing protein folding through enthalpic and entropic factors.
Area of Science:
- Biophysics
- Computational Biology
- Protein Science
Background:
- Protein structure prediction is crucial for understanding protein function.
- Amino acid composition is a fundamental property of proteins.
Purpose of the Study:
- To determine if amino acid composition alone is sufficient for predicting protein structural class.
- To investigate the physical factors governing protein folding preferences.
Main Methods:
- Exhaustive enumeration of lattice model conformations.
- Eigenvalue analysis of Kirchhoff matrices to assess non-bonded contacts.
- Analysis of non-bonded contact distributions in known protein structures.
Main Results:
- Amino acid composition accurately predicts protein structural class (alpha, beta, alpha + beta, alpha/beta) with 81% accuracy.
- Protein folding is influenced by the distribution of non-bonded contacts, specifically coordination numbers and cluster geometry.
- Enthalpic and entropic effects compete to determine favored contact distributions for specific amino acid compositions.
Conclusions:
- Amino acid composition is a powerful predictor of protein structural class.
- The distribution of non-bonded contacts is a key determinant of protein fold.
- Understanding these factors provides insights into the physical basis of protein folding.