Preference of Amino Acids in Different Protein Structural Classes: A Database Analysis

Understanding sequence-structure relationship is the key step in protein modeling and de novo protein design. Although almost 55,000 protein structures are solved and stored in protein data bank, elucidating sequence-structure relationship is still a challenging task. To understand sequence-structure relationship better, a statistical analysis of amino acid residues in four major structural classes of protein viz. α proteins, β proteins, α/β proteins and α+β proteins is performed. We use non-homologous proteins from (<; 30% identity) October 2008 release Brookhaven Protein Data Bank (PDB) with resolution better than 2.5 angstrom. Interestingly, in comparison to the helical protein, the helical propensities of hydrophobic residues in mix proteins (containing both α helix and β sheet) are increased significantly. On the other hand, the helical propensities of hydrophilic residues are reduced in mixed proteins. A reverse trend is observed in strand propensity. The difference in helical propensity of hydrophobic and hydrophilic residues in different fold may be due to differential folding mechanism. The size of protein may also play a crucial role. A position specific analysis of helices is also done in all α and α/β proteins. The detailed analysis of helix dissection revealed that, the presence of β sheet influences the individual preference of amino acids in different positions within helix. This result indicates that the preference of amino acid in different positions (N terminus, C terminus and middle) within α helix are influenced by long range interactions with other structural elements.