Table (II).
Chart summarizing the different hydrophobicity scales and their applicability
| scale | ref | year | derivation* | α/β | 2-state/3-state | characteristics/applicability |
|---|---|---|---|---|---|---|
| Hopp & Woods | 1 | 1981 | exp | n/a | 2 | • hydrophilicity scale for antigenic sites on the protein surface; • derived from the values of Levitt44; • some values were adjusted to fit immunochemical data of 12 proteins; • for the proteins only the primary sequence was available!; • window used is 6 residues ≈ length of antigenic determinant |
| Goldman, Engelman, Steitz | 2 | 1986 | exp | α | 2 | • hydrophobicity scale for single trans-membrane helices; • semi-theoretical approach based on energetic considerations of residues undergoing hydrogen bonds in helices derived from experimental data in the literature; • hydrophobicity scale as a sum of hydrophilic and hydrophobic components |
| Wimley & White | 3,4 | 1996 | exp | α | 2 + 3 | • derived by measuring the partitioning energies of host-guest penta-peptides; • whole residue scale that considers the polar peptide bond; • interface: POPC vesicle interface; bilayer: n-octanol; • for unfolded peptides in all 3 phases (solution, interface, bilayer) |
| Hessa et al. | 23,24 | 2005/2007 | exp | α | 2 / pot | • designed TM helix within the Lep protein that is inserted via the Sec61 translocon; • TM helix is 19-residue helix with amino acid in question incorporated in the center; • measured fraction of singly vs. doubly glycosylated Lep molecules to derive the scale; • therefore applicable to folded MPs; • scale has been extended to position-dependent free energy scale (2007) |
| Eisenberg & Weiss | 45 | 1982 | cons | n/a | 2 | • normalized consensus scale of five different scales |
| Kyte & Doolittle | 10 | 1982 | cons | n/a | 2 | • normalized consensus scale based on experimental observations of different scales; • refinement by studying hydropathy plots of proteins of known X-ray structure; |
| Guy | 18 | 1985 | cons | n/a | 2 | • based on experimental and statistical results from several studies; • considers solvent accessibility according to accessible layers of amino acids in globular proteins |
| Janin | 7 | 1979 | KB | n/a | 2 | • derived from X-ray structures of 22 soluble proteins; • looked at molar fraction of buried and accessible residues |
| Punta & Maritan | 8 | 2003 | KB | α | 2 | • derived two membrane propensity scales from two TM helix databases using a simple perceptron algorithm; • databases contained 118/228 TM helices; • sequence identity of the proteins was 30% |
| Beuming & Weinstein | 25 | 2004 | KB | α | n/a | • calculated surface propensities of amino acids (probability of finding a residue on the surface of a TM protein); • based on surface fractions of residues; • considered 28 α-helical MPs |
| Senes et al. | 9 | 2007 | KB | α | 2 / pot | • calculated membrane depth-dependent potential for amino acid side-chains; • considered 24 α-helical MPs |
| UHS | 2008 | KB | α/β | 2 + 3 | • derived from 60 known structures of folded MPs; • considers folded structures both in solution and membrane bilayer; • both α, β, and α/β structures were taken into account with approximately equal distribution of helices and strands; • considers only depth in membrane bilayer and no accessibility or secondary structure | |
| MHS | 2008 | KB | α | 2 + 3 | • derived from 16 known structures of folded MPs from mammalian organisms; • only α-helical structures could be taken into account; • considers folded structures both in solution and membrane bilayer; • considers only depth in membrane bilayer and no accessibility or secondary structure |
exp: experimental; cons: consensus; KB: knowledge-based; pot: potential