Deciphering Genetic Overlaps - A Comprehensive Study On Viral Host Determination Using Machine Learning And Deep Learning Models

Aug 4, 2026

Upload by Ivy

The study uses machine learning and deep learning models to analyze the relationship between viral genetic DNA sequences and host organisms. A comprehensive dataset is assembled from databases such as ExPASy and NCBI, then viruses are classified into eight host categories. Genetic overlaps reduce separability, limiting gains across tested models. By consolidating classes into three groups (plants, animals, microorganisms), evaluation improves, reaching up to 70% accuracy with Random Forest and over 85% with LSTM. Shared genomic segments challenge conventional molecular distinctions and refine computational viral host determination.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

The Significance of Virus Host Categorization

Data Set & Domain

Journal of Advanced Zoology ISSN: 0253-7214 Volume 45 Issue 3 Year 2024 Page 776-788 Deciphering Genetic Overlaps: A Comprehensive Study On Viral Host Determination Using Machine Learning And Deep Learning Models Pankaj Agarwal1*, Sapna Yadav2 1*K.R Mangalam University, Gurgaon, [pankaj.agarwal7877@gmail.com](pankaj.agarwal7877@gmail.com)[ ](pankaj.agarwal7877@gmail.com)2Jamia Millia Islamia, Delhi, [sapnayadav0821@gmail.com](sapnayadav0821@gmail.com) *Corresponding Author: Pankaj Agarwal *K.R Mangalam University, Gurgaon, [pankaj.agarwal7877@gmail.com](pankaj.agarwal7877@gmail.com) | CC License
CC-BY-NC-SA 4.0 | Abstract
The study uses machine learning and deep learning models to study the intricate relationship between viral genetic DNA sequences and host organisms. It uses a comprehensive dataset from databases like ExPASy andNCBI, which encodes crucial genetic information for viral replication.
The study aimed to create a viral DNA dataset and develop robust machine learning and deep learning models to classify viruses into eight host categories. Despite extensive experimentation using various models, performance improvement was elusive due to genetic overlaps. Viral genomes from different classes had significant shared genetic sequences, making it difficult for these models to identify unique class-specific features, blurring the lines of differentiation.
The study reduced the number of classes from eight to three, focusing on plants, animals, and microorganisms. This resulted in improved evaluation metrics, with the Random Forest Machine learning model reaching a maximum accuracy of 70% and the LSTM deep learning model surpassing 85%, overcoming earlier challenges.
The discovery that viral genomes from different classes share significant genetic overlaps challenges conventional molecular distinctions, emphasizing the complexity of molecular differentiation in viral genomes. This pragmatic approach aligns molecular understanding with genetic data in viral host determination.
Key Terms: Machine Learning; Deep Learning; Viral Host Specificity; DNA Sequences; Disease Surveillance. | | --- | --- | 1. Introduction Understanding viral dynamics and host organism interactions is crucial in virology and molecular biology. Advancements in technology, including machine learning and deep learning, offer new opportunities to unravel these complexities. This work aims to develop a predictive model using advanced machine learning and deep learning models to determine virus host organisms based on genetic DNA sequences, using a comprehensive dataset from databases like ExPASy and NCBI.This research aims to generate a vast viral DNA dataset and develop machine learning and deep learning models to classify viruses into eight host categories, reflecting the diversity of viruses and their potential hosts, reflecting the molecular tapestry of life. Machine learning models like Decision Trees, Random Forest, Naïve Bayes, KNN, and deep learning models like CNN and LSTM face a challenge due to genetic overlaps among different viral classes. These shared genetic sequences make it difficult for conventional models to identify unique class-specific features. The research reduced the number of classes from eight to three, focusing on plants, animals, and microorganisms, to better understand viral genetic information. This approach aligns molecular understanding with genetic data, resulting in improved metrics and clarity in viral genome classification. The discovery that viral genomes from different classes share significant genetic overlaps challenges traditional molecular distinctions and emphasizes the need for advanced computational approaches to navigate genetic data. This not only advances computational virology but also offers a nuanced perspective on viral host classification. Advanced computational methodologies and molecular insights will enhance our understanding of viruses and host organisms,

FAQs

How does the study determine viral hosts using genetic information?

It trains machine learning and deep learning models on viral DNA sequence data collected from ExPASy and NCBI to classify viruses by host categories.

Why were classification model improvements difficult in the original eight-class setup?

Genetic overlaps between viral classes caused shared genetic sequences, making it hard for models to learn distinct class-specific features.

What change improved the classification performance, and what accuracies were achieved?

The study reduced eight host classes to three groups (plants, animals, microorganisms). Random Forest reached up to 70% accuracy, while LSTM surpassed 85%.

Aug 4, 2026

Upload by Ivy