Multi Agent Network-propelled Data Extraction for Protein Research

摘要

To comprehend the structure function model, a novel procedure for proteins classification and prediction is proposed. It uses multi agent system technique that represents a new standard for building software systems to predict and classify protein constructs. To categorize the proteins, support vector machine (SVM) has been developed to extract features from the protein sequences. This paper describes a method for predicting and classifying the subordinate structure of proteins. Support vector machine (SVM) modules were developed using multi-agent system principle for predicting the proteins and its functions there-by achieving the goals of accuracy, specificity, sensitivity, of 92%, 94.09%, and 91.59% respectively. The proposed algorithm provides an understanding of the protein structure, which can greatly improve biological science by analyzing the relationships amongst proteins.