Amino Acid Hydrophobicity for Structural Prediction of Peptides and Proteins

Hydrophobicity refers to the physical property of molecules that repel water. Hydrophobic molecules tend to be nonpolar and therefore prefer to interact other neutral and nonpolar solvents. In water, hydrophobic molecules often cluster together to form micelles. The hydrophobicity of molecules such as amino acids can be experimentally measured as reported by Richard Wolfenden.

Physiochemical properties routinely allow the characterization of peptide or protein sequences of known or unknown function. “Peptide Property Calculators,” such as the one found at BSIs website for the structural prediction or detection of membrane-associated or embedded β-sheets or α-helices, mostly present in larger peptides or proteins, are based on the Kyte-Doolittle Hydrophobicity scale. Several experimentally defined hydrophobicity parameters for amino acids now allow the calculation of peptide properties. In addition, many different scales for calculating peptide properties using hydrophobicity parameters have been developed in recent decades. However, the Kyte-Doolittle Hydrophobicity scale appears to be used the most in peptide calculators.

Often, to achieve a more accurate prediction for the peptide or protein structure of unknowns, five different hydrophobicity parameters are employed:    

1) the overall hydrophobicity, 

2) the hydrophobic moment for detection of α-helical and β-sheet membrane segments, 

3) the alternating hydrophobicity, 

4) the alternating hydrophobicity, and 

5) the exact β-strand score. 

Simm et al. in 2016 reported that most scales allow discriminating between transmembrane α-helices and transmembrane β-sheets. However, using the five different hydrophobicity parameters enables the assignment of peptides to pools of soluble peptides of different secondary structures. But using alternating hydrophobicity is not significantly beneficial. All scales appear to have limitations in separation capacity. Simm et al. observed that the scales derived from the evolutionary approach performed best in separating different peptide pools. The 98 hydrophobicity scales known today have differently defined hydrophobicity values for the 20 amino acids. Unfortunately, because different experimental approaches were used for their definition, a high variance between different scales is observed. Importantly, Simm et al. found that the scale, as defined by Naderi-Manesh, developed in 2001, performed somewhat better than the other hydrophobicity scales. Therefore Simm et al. proposed a rule of thumb for the use of a hydrophobicity scale for the identification of peptides with transmembrane segments from a pool of peptides. Place the hydrophobicity value of arginine (Arg, R) and tyrosine (Tyr, Y) most distant from the value for glutamate (Glu, E). Select the hydrophobicity values for asparagine (Asn, N), aspartate (Asp, D), histidine (His, H), and lysine (Lys, K) such that they are in the center of the scale.

{Simm S, Einloft J, Mirus O, Schleiff E. 50 years of amino acid hydrophobicity scales: revisiting the capacity for peptide classification. Biol Res. 2016 Jul 4;49(1):31. doi: 10.1186/s40659-016-0092-5. PMID: 27378087; PMCID: PMC4932767. https://www.ncbi.nlm.nih.gov/pubmed/27378087}.