Deoxyribose Nucleic Acid Nucleotide Virus Classification with Machine Learning

Abstract

Deoxyribose Nucleic Acid (DNA) viruses are a major focus in virology because they cause many different types of human illness. DNA viruses' biology, transmission, and pathogenicity can be better comprehended if they are first properly categorized. Recently, machine learning has proven to be an effective method for studying massive amounts of biological data, such as DNA viral sequences. Here, we give a high-level overview of utilizing machine learning to categorize DNA viral sequences. We address supervised, unsupervised, and deep learning strategies that have been used for DNA viral sequence classification. The data's high dimensionality, highly variable sections, and the requirement to differentiate between closely related viral strains are only a few of the difficulties we emphasize in DNA virus sequence classification. The FASTA Tool was used to retrieve the dataset from the NCBI database. The data collected included three human gene family sequences and six virus sequences. We employed six machine learning methods to classify the DNA viruses (Sars-Cov-1, Mers-Cov-2, Ebola, Dengue, Influenza, Synthase, Ion Channel, and Transcription Factor) with 98% accuracy.