Clustering Protein Super Secondary Structure with Artificial Neural Network Based Self Organizing Map Machine Learning Approach Using In-Memory Computing Environment

Abstract

Protein’s super secondary structure knowledge is important to understand protein functions which is vital for all living organisms. In this research, an in-memory computing approach is created to identify the super secondary structure of protein on primary protein sequences. The Artificial Neural Network (ANN) based Self Organizing Map (SOM) algorithm was used to develop a three-way predictive model. The developed clustering model helps to assign protein super secondary structures mainly into â-Turns. Also this model is implemented in an in-memory computing environment—SPARK. The analysis results using ANN-SOM approach in clustering protein super secondary structures provide better high accuracy in the in-memory computing environment.