Subcellular localization of proteins using automated fuzzy inference system

Protein sub-cellular localization prediction involves the computational prediction of where a protein resides in a cell. It is an active area of research in bioinformatics-based prediction of protein function and genome annotation, and research finding from this area can aid the identification of drug targets. Different machine learning and data mining techniques are used to do this prediction; however, there is still scope of improvement with higher accuracy. In this paper, a fuzzy rule based system is used to predict the sub-cellular localization of protein. This method takes some time in constructing the rules from the given data initially, but once the model established, it can predict the localization of unknown proteins very fast. The adaptive nature of fuzzy rules makes this technique to automatically incorporate new protein localization information once available. Initial finding from this research is also encouraging. An average 86% accuracy has been achieved that suggests for further exploration and future scope of fuzzy based theory in the field of biological sciences.