Proteomics is a research area that analyze protein molecules in order to determine their sequence, structure and functions. It is very popular because proteins play an important part in the processes in the cells of organisms. There are plethora of methods for protein annotation, which try to found out the functions of the proteins in the interactions between the proteins. However, they cannot follow the speed of determination of protein structures, thus the gap between the known protein structures and those that are functionally annotated continually rises. Therefore, there is a noticeable necessity for developing fast and accurate computational methods for determination of protein functions. In this paper, we present an approach for prediction of the amino acid residues that are part of binding region where an interaction occurs with other protein structure. These predictions could be used for protein annotation. We present some results of the evaluation of our approach, and we compare it with some approaches found in the literature.
Protein molecule, protein function, protein binding site