Which amino acid property is commonly analyzed in a hydropathy plot to predict transmembrane domains?
Computational prediction of membrane spans relies on physicochemical property hydrophobicity quantifying tendency to partition into non-polar phase. Kyte-Doolittle hydropathy scale derived from experimental water vapor transfer energies assigns isoleucine 4.5, valine 4.2, leucine 3.8, phenylalanine 2.8 as hydrophobic; arginine -4.5, lysine -3.9, asparagine -3.5, aspartate -3.5 as hydrophilic. Sliding window average length nineteen matches bilayer thickness smooths sequence revealing peaks sustained above cutoff indicating candidate transmembrane helices. Complementary scales include Goldman Engelman Steitz detecting helical hydrophobic moment and White Wimley interfacial octanol partitioning reflecting translocon insertion energetics. Algorithms combine hydrophobicity with charge distribution applying positive-inside rule where cytosolic loops enriched Arg Lys due to membrane potential, and with length constraints. Molecular weight does not correlate with membrane insertion, net charge alone insufficient due to loop variability, polarity ambiguous. Therefore hydrophobicity remains primary property analyzed in hydropathy plots for mapping membrane protein topology before high-resolution structures confirm assignments. Inclusion of positive-inside rule and signal peptide detection enhances prediction distinguishing signal anchors from true transmembrane helices for cloning strategies.
Ref: Kyte and Doolittle, Hydropathy Plot for Transmembrane Prediction, J Mol Biol 1982.