An Framework for Simple Sequence Repeat Reduction with Information Extraction using Machine Learning

Abstract

Demand for the agricultural improvements using the advanced computer algorithms have increased in the recent years. The primary focus is on the higher crop production rates with least damage to the crops due to various diseases. In the recent times, a good number of research attempts are observed to formulate multiple computerized algorithms to identify the Amino Acid sequence and further protein sequences, which are responsible for diseases to the plants and crops. However, due to the higher complexity of DNA structure and further the complex process for DNA to Amino Acid extraction, these recent researches have produced unsatisfactory outcomes.  Henceforth, in order to solve the primary challenge of higher time complexity of the DNA processing methods, this work proposes two algorithms to reduce the DNA sequence length without losing vital information using machine learning. Firstly, the use of clustering method to reduce the size ensures least information loss and best processing time. Secondly, the look up based indexed Amino Acid extraction process ensures higher correctness of the extraction and again in best possible time. The proposed framework produced nearly 98% accuracy in 0.107 sec time frame, which is relatively 5% improvement in accuracy and 10% improvement in time complexity