Prediction of protein cellular localization site by using data mining techniques

In recent years data mining has been intensively used in the medical field, bioinformatics and biomedical research. This paper focus on the prediction of the protein localization site on the basis of their amino acid sequences in Escherichia Coli (E coli) bacteria. This is of great importance because information on cellular location is helpful for annotation of proteins and genes. So there is need to develop a simple method with high prediction accuracy. To accomplish this various classification techniques are considered on the dataset by performing various experiments. Based on these experiments various measures such as classification accuracy, error rate, F-Measure, etc are calculated. The dataset used is the E coli dataset. The maximum accuracy is achieved by the proposed hybrid model of Support Vector Machine and the LogitBoost technique. The classification accuracy achieved by this model is 95.23%.