Hydrophobicity Revisited: a Molecular Story

2020-06-08T13:01:45Z (GMT) by David Cavanaugh Krishnan Chittur
In a previous paper we have introduced a new hydrophobicity proclivity scale and justified its superior performance characteristics, particularly in the context of a scale for protein alignments, but also for its strong correlation with many other amino-acid physico-chemical properties. Within that paper, we calculated a corrected free energy of residue burial of each amino-acid in folded proteins from a linear regression of amino-acid free energy of transfer from water to n-Octanol (F&P octanol scale dGow, Y axis) and our Hydrophobicity Proclivity Scale
(HPS, X axis). In this present paper we pursue the latter general findings in more detail by considering the relationship of hydrophobicity and other physico-amino-
acid scales with the molecular geometry of amino-acids and secondary group structure/surface chemistry, with a concommitant discussion of the dimensions/geometry
of the caveties that amino-acids make in water. We identify a series of molecular physico-chemical properties that uniquely define the natural selection and geometry of the 20 natural amino-acids. We use the corrected free energy of amino-acid burials in proteins (Y axis) and a multiple linear regression to identify the AA molecular physico-chemical properties (X1, X2, ...) that explain the energetics of amino-
acid water contacts in an unfolded protein state to that of the folded protein state by modeling these two states as a solvent-solvent transfer, thus, providing a thermodynamical model for the initial stages of protein folding. Between our previous paper and the current paper we can greatly simplify and reduce the very large number of amino-acid scales in the literature to a small number of amino-acid property scales. Finally, we explore the numerical relationship between the structure of the genetic code and molecular physico-chemical properties of AA’s that in turn can be related directly to hydrophobicity. We validate and explain our novel models we describe herein with extensive data from the literature.