Biomedical Engineering Reference
In-Depth Information
sequences. The first step to achieve this goal is the successful prediction of
membrane spanning regions in TMH proteins. Accordingly, several methods have
been proposed to identify TMH proteins and predict their membrane spanning
regions. Most of the methods use amino acid properties and machine-learning
methods have also been developed for prediction.
3.2.1 Statistical Methods
The statistical methods are mainly based on hydrophobicity profiles, physicochem-
ical properties, conformational parameters, and the combination of them.
Kyte and Doolittle [
12
] developed a hydrophobicity profile method for locating
the membrane spanning helical segments in amino acid sequences. To identify
membrane regions, they implemented a moving-window approach in which the
hydrophobicity indices of the amino acid residues have been summed over a stretch
of residues and took the average. If the average hydrophobicity for a segment
exceeds a threshold, the segment has been suggested as a transmembrane helix.
Eisenberg et al. [
13
] developed the helical hydrophobic moment as a measure of the
amphiphilicity of
helix. This hydrophobic moment differed between transmem-
brane, and globular helices, and could thus be explored to predict transmembrane
regions [
14
]. Jayasinghe et al. [
15
] attempted to improve hydropathy analysis by
directly improving the hydropathy scales. The commonly used hydrophobicity
scales neglect the thermodynamic constraints
a
-helices impose on transmembrane
stability, and hence they have derived a whole-residue hydropathy scale from the
Wimley-White experiments that took into account the backbone constraints.
Ponnuswamy and Gromiha [
16
] developed a surrounding hydrophobicity scale
applicable to both globular and membrane proteins and proposed a “surrounding
hydrophobicity profile” method for predicting transmembrane helices from the
amino acid sequence of a protein. This profile is simply the plot of the surrounding
hydrophobicity indices of the residues against their sequence numbers. In this plot,
the hydrophobic and hydrophilic parts are distinguished by a horizontal line
representing the average hydrophobicity value, which is obtained from the
surrounding hydrophobicity values for all the amino acid residues in the sample
set of proteins. The surrounding hydrophobicity profile thus constructed projects
the transmembrane helices as a sequence of peaks and valleys above the average
middle line (or with a few valleys crossing down the average line), and the other
parts as peaks and valleys frequently crossing the middle line, of falling below the
middle line. The criterion of a continuous sequence of 20-24 points above the
average line with a maximum of two nonadjacent exceptions was used to determine
the length of a predicted transmembrane helix.
Hirokawa et al. [
17
] developed a method, SOSUI, for discriminating transmem-
brane helical proteins and predicting their membrane spanning segments using the
combination of a variety of physicochemical parameters. Specifically, the
parameters, Kyte-Doolittle hydropathy index, amphiphilicity, relative and net
charges, and protein length have been used to detect transmembrane
a
a
-helices.
