Introduction
Multidrug And Toxin Extrusion (MATE) proteins, first characterized as bacterial drug transporters, are now present within almost all species, and are therefore, one of the most conserved transporter families. It plays a very important function in the secretion of cationic drugs across the cell membrane.MATE proteins have been characterized as important transporters that mediate the final excretion of cationic drugs into bile and urine
We have developed an automatic tool for prediction of Multidrug and Toxin Extrusion (MATE) proteins using supprot vector machine method. The SVM program is much more flexible and user friendly. SVM based PSSM model is the best model (accuracy = 86.77%,,sensitivity = 73.02% , specificity = 93.65% ,MCC= 0.695).
Sequence Information :
The server accepts protein sequence in FASTA format as an input. The output displays sequence number, score and decision of the model. The results will be send to the user via email-id.
Evaluation of Performance:
Applying the following equations accuracy, sensitivity, specificity and Matthew correlation coefficient (MCC) were calculated for evaluating the accuracy of SVM classifiers:
Sensitivity: It is determined as the percentage of MATE that is correctly predicted as MATE.
Sensitivity = TP/TP+FN*100
Specificity: It is the percentage of non-MATE that is correctly predicted as non-MATE.
Specificity = TN/TN+FP*100
Accuracy: It is the percentage of correct predictions out of the total number of predictions.
Accuracy =TP+TN/TP+FP+TN+FN*100
Matthews correlation coefficient (MCC): It is a measure of both sensitivity and specificity.
MCC= (TP*TN)-(FN*FP)/(sqrt(TP+FN)*(TN+FP)*(TP+FP)*(TN+FN)
Multidrug And Toxin Extrusion (MATE) proteins, first characterized as bacterial drug transporters, are now present within almost all species, and are therefore, one of the most conserved transporter families. It plays a very important function in the secretion of cationic drugs across the cell membrane.MATE proteins have been characterized as important transporters that mediate the final excretion of cationic drugs into bile and urine
We have developed an automatic tool for prediction of Multidrug and Toxin Extrusion (MATE) proteins using supprot vector machine method. The SVM program is much more flexible and user friendly. SVM based PSSM model is the best model (accuracy = 86.77%,,sensitivity = 73.02% , specificity = 93.65% ,MCC= 0.695).
The server accepts protein sequence in FASTA format as an input. The output displays sequence number, score and decision of the model. The results will be send to the user via email-id.
Applying the following equations accuracy, sensitivity, specificity and Matthew correlation coefficient (MCC) were calculated for evaluating the accuracy of SVM classifiers:
Sensitivity: It is determined as the percentage of MATE that is correctly predicted as MATE.
Specificity: It is the percentage of non-MATE that is correctly predicted as non-MATE.
Accuracy: It is the percentage of correct predictions out of the total number of predictions.
Matthews correlation coefficient (MCC): It is a measure of both sensitivity and specificity.