An Evolutionary Algorithm to Predict Super Secondary Structures of Proteins from Secondary Ones, A Case Study: β-LACTAMAZE Enzyme

The protein’s motifs (called super secondary structures) are dense three-dimensional structures of proteins consisting of several secondary structures in a specific geometric arrangement. The prediction of motifs is a matter of concern and has been studied. The previous studies dealt with motif prediction based on the polypeptide chain; however, the prediction of motifs based on the secondary structures leads to more accurate prediction. This study aims to address such a prediction. First, several secondary structures are constructed and then, based on the energy level and using a metaheuristic (evolutionary) algorithm called Imperialist Competitive Algorithm. (ICA) The protein’s motifs are predicted. The advantage of our approach over existing approaches is that secondary structural data as input to our algorithm leads to a more accurate prediction that is closer to the real protein third than previous algorithms. We applied our method to predict super secondaries of the enzyme β−LACTAMASE, whose specification was obtained from the PDB file in Yasara. This enzyme is produced by bacteria and provides multi-resistance to antibiotics β−LACTAMA. Then we evaluated our prediction using Root-Mean-Square Deviation (RMSD). It shows the average distance between the two proteins structurally having the same alignment. Having determined the structural alignment of the two proteins, we determined the similarity of their ۳D structures using RMSD. If the RMSD between two structures is less than ۲, it denotes they are very similar. Accordingly, we used RMSD to show how much similarity exists between the motif obtained by our proposed algorithm for β−LACTAMASE and its native structure.

کلیدواژه ها:

Motif prediction ، Secondary structure ، Imperialist Competitive Algorithm (ICA) ، Root-Mean-Square Deviation (RMSD)