Biomedical Engineering Reference
In-Depth Information
The systematic analysis of amino acid residues in these transporters showed a
preference of specific amino acid residues. The residue Asn is dominant in
channels/pores among all the transporters. Interestingly, Asn plays an important
role to the stability and function of
-barrel membrane proteins [
43
,
80
]. Glu is
another amino acid that shows the difference of more than one with electrochemical
transporters. It has been showed that the residues Glu166 and Glu148 are important
for the channel function in CIC chloride channel proteins [
81
]. The residues Phe
and Leu are dominant in electrochemical transporters. In addition, the composition
of Ala, Ile, Val, and Trp is higher in this class of proteins compared with other two
transporters. Interestingly, in glycerol-3-phosphate transporter the space between
helices 1 and 7 is filled by nine aromatic side chains and the occurrence of bulky
aromatic residues helps to close the pore completely [
82
]. The higher occurrence of
hydrophobic residues is due to the presence of long stretches of these residues in
membrane spanning segments of
b
a
-helical membrane proteins.
4.4 Discrimination of Transporters Based on Different
Classes and Families
We have utilized different machine-learning methods for discriminating channels/
pores, electrochemical and active transporters [
78
]. We observed that the average
accuracy of discriminating channels/pores, electrochemical and active transporters
lies in the range of 56-64% for different machine-learning techniques. The highest
accuracy of 64% is obtained for neural network-based method using fivefold cross-
validation and jack-knife tests. Interestingly, this method has similar values for
sensitivity and specificity, indicating the ability of picking up the specific class of
transporters and eliminating others with similar accuracy. On the contrary, homol-
ogy search-based methods such as BLAST could discriminate the three classes of
transporters with an average accuracy of 51.5%. We have developed a web server
for predicting the transporters into three classes. The server takes the amino acid
sequence as input (Fig.
7c
) and displays the predicted type of the transporter in the
output (Fig.
7d
). Recently, we have utilized PSSM profiles and amino acid
properties for discriminating the three classes of transporters, channels/pores,
electrochemical and active transporters and obtained an average accuracy of 78%
in a test set of 118 proteins [
83
].
In addition, we have analyzed the influence of PSSM profiles and amino acid
properties in six different families of transporters developed a method to discrimi-
nate them. We observed that the PSSM with amino acid properties could discrimi-
nate six different classes of transporters with an average accuracy of 69%, with an
improvement of 8% over amino acid composition [
83
]. We have developed a
strategy for the annotation of transporters in genomic sequences and it is depicted in
Fig.
8
. First our method classifies the query protein into transporter or nontransporter.
For a transporter, it predicts the class as channels/pores, electrochemical or active
