Prediction of MHC class II binders using the ant colony search strategy

Abstract

Objective: Predictions of the binding ability of antigen peptides to major histocompatibility complex (MHC) class II molecules are important in vaccine development. The variable length of each binding peptide complicates this prediction.

Methodology: Motivated by the search properties of the ant colony system (ACS), a method for the identification of an alignment for a given set of short protein peptides has been developed. This alignment is further used for the derivation of a position specific scoring matrix. The distinguishing feature of this method is the use of the collective optimized search strategy of ants for the selection of the alignment.

Results: The performance of the new model has been evaluated with several benchmark datasets. It achieves better or comparable results as compared to the performance of existing methods.

Conclusion: The experiments demonstrate that the predictive performance of the scoring matrix embodies several promising characteristics.