Deep Learning Algorithms for Secure and Efficient Compression of Genomic Sequence Data

Abstract

The idea that DNA encodes the human body is uncontested. Everything about a person, including behaviour and physical characteristics, is influenced by genetic information. Genomic data is a rich source of information when it is analysed, genomic data has a significant potential for use in disease diagnosis. Despite the fact that One size does not fit all, the medical advice is usually generic. The intelligent machine may provide us with additional human body knowledge on the contrary to build intelligent robots. The ongoing development of genomic data is either beneficial for engineering as well as medical research. Our search for personalised medicine may have a solution in the overlapping fields of engineering and medicine. The introduction of new emergent multimedia applications with an increase in demand and consumption of genetic data, as well as the invasion of digital media into every aspect of diagnosis and prognosis, are compelling new challenges for next-generation sequencing. The processing of genomic data is intended to meet hitheito unrecognised requirements, including compression, robustness, security, unambiguity, and the prediction of future anomalies, in addition to delivering individualised treatment. Deep neural network based deep learning has been intensively researched in recent years as one of the possible methods for overcoming these difficulties. Because digital data must be maintained effectively, with very high quality, and shouldn’t be easily modified by computer standards. The goal is to develop a safe genome compression method that significantly outperforms existing sequencing methods in terms of all performance parameters.