1. Abstract

DNA virus detection plays an important role in modern bioinformatics and medical diagnostics. Identifying whether a person carries a virus based on DNA sequences helps in early disease detection and prevention. Traditional laboratory methods can be time-consuming and require complex analysis.

This project focuses on building a deep learning model trained on a DNA promoter dataset to determine whether a virus is present or not. DNA sequences consist of combinations of four molecules: Adenine (A), Cytosine (C), Guanine (G), and Thymine (T). These molecules form patterns in DNA that can be analyzed using machine learning techniques.

In this project, the dataset is preprocessed and used to train a Multi-Layer Perceptron (MLP) classifier, which is a type of neural network used for classification tasks. After building the model, a web application is developed using the Django framework to allow users to input DNA sequences and receive predictions.

Finally, the web application is hosted on Amazon Web Services (AWS) using an EC2 instance, making the system accessible from anywhere through the internet. This project demonstrates the integration of deep learning, web development, and cloud computing to create a practical virus detection system.

2.Objectives

The main objectives of this project are:

1. To analyze DNA promoter dataset using deep learning techniques.
2. To build a neural network model for detecting viruses from DNA sequences.
3. To preprocess DNA data for machine learning applications.
4. To implement a Multi-Layer Perceptron (MLP) classifier for classification.
5. To develop a web application using the Django framework.
6. To integrate the trained machine learning model with the web application.
7. To deploy the application on AWS cloud for public access.

3. Existing System

In traditional virus detection systems, biological laboratories perform manual analysis of DNA samples using specialized equipment and chemical testing procedures.

Limitations of the Existing System

1. The process is time-consuming.
2. Requires expensive laboratory equipment.
3. Requires skilled professionals to analyze DNA samples.
4. Difficult to analyze large amounts of genetic data quickly.
5. Limited accessibility for remote analysis.

Due to these limitations, automated systems using machine learning and deep learning techniques are becoming increasingly important.

4. Proposed System

The proposed system introduces an automated approach to detect viruses using deep learning models trained on DNA sequence data.

In this system:

1. DNA sequence data containing combinations of A, C, G, and T molecules is used.
2. The dataset is preprocessed into a format suitable for machine learning algorithms.
3. A Multi-Layer Perceptron (MLP) classifier is trained to recognize patterns in DNA sequences.
4. The trained model predicts whether a virus is present or not.
5. A Django-based web application is created to provide a user interface.
6. Users can input DNA sequence information through the website.
7. The model processes the input and displays the prediction result.
8. The website is deployed on AWS EC2 cloud server to make it accessible online.

 5. Implementation Procedure

The project is implemented through several stages.

 Step 1: Data Collection

The dataset used in this project is a DNA promoter dataset which contains DNA sequences encoded using four molecules:

1. Adenine (A)
2. Cytosine (C)
3. Guanine (G)
4. Thymine (T)

These molecules combine to form DNA sequences.

 Step 2: Data Preprocessing

The dataset is preprocessed to convert the DNA sequence information into a machine learning friendly format.

Preprocessing steps include:

1. Cleaning the dataset
2. Encoding DNA characters (A, C, G, T) into numerical values
3. Preparing the dataset for training

Step 3: Model Building

A Multi-Layer Perceptron (MLP) classifier is used to train the deep learning model.

MLP is a type of neural network that contains multiple layers including:

1. Input Layer
2. Hidden Layers
3. Output Layer

This neural network learns patterns from DNA sequences to classify whether a virus is present or not.

Step 4: Model Training

The dataset is divided into:

1. Training data
2. Testing data

The MLP classifier is trained using the training dataset to learn DNA sequence patterns.

 Step 5: Model Evaluation

After training, the model performance is evaluated using testing data.

Evaluation metrics may include:

1. Accuracy
2. Precision
3. Recall
4. Confusion matrix

Step 6: Creating Django Web Application

A Django web framework is used to create a web application for the model.

Django provides built-in functionalities such as:

1. User authentication
2. Form handling
3. Database management
4. File uploads

Users can input DNA sequence data through the website interface.

Step 7: Integrating Model with Website

The trained machine learning model is integrated with the Django application.

The workflow is:

1. User enters DNA sequence on the website.
2. Django sends the input data to the trained model.
3. The model processes the input sequence.
4. The prediction result is displayed on the website.

Step 8: Deployment on AWS

The final web application is deployed on Amazon Web Services (AWS) using an EC2 T2 Micro Instance.

AWS provides:

1. Cloud servers
2. Virtual machines
3. Remote hosting environment

After deployment, the system can be accessed online from anywhere.

6. Software Requirements

The software required for this project includes:

1. Python Programming Language
2. Django Web Framework
3. Scikit-Learn Library
4. NumPy and Pandas
5. Jupyter Notebook / VS Code
6. AWS Management Console
7. Web Browser

7. Hardware Requirements

Minimum hardware requirements for this project include:

1. Processor: Intel i3 or higher
2. RAM: 4 GB minimum (8 GB recommended)
3. Storage: 256 GB hard disk or higher
4. System: Laptop or Desktop Computer

 8. Advantages of the Project

1. Provides automated virus detection using DNA data.
2. Reduces the need for manual analysis.
3. Faster processing of genetic data.
4. Integrates deep learning with web applications.
5. Accessible from anywhere via cloud deployment.
6. Useful in bioinformatics and medical research.
7. Demonstrates integration of AI, web development, and cloud computing.