e17-6sp-NGS-Data-AnalysingToolkit
Team
- E/17/015, ARSHAD MRM , e17015@eng.pdn.ac.lk
- E/17/327, KAVINAYA Y, e17159@eng.pdn.ac.lk
- E/17/379, NISHANKAR S, e17230@eng.pdn.ac.lk
Table of Contents
- Introduction
- Problem Overview
- Solution Overview
- Solution Architecture
- Data Flow
- Requirements Analysis
- Use Case Diagrams
- Machine Learning Proposal
- Machine Learning Workflow
Introduction
This is a Online toolkit to detect the biomarkers for Alzheimer’s Disease using the Next generation Sequencing Data Set. Simple and self explanatory GUI elements of our toolkit will help to analyse the data and visualize them in different ways (different graps and tables).
Problem Overview
Problem related to Alzheimer’s Disease
- Found in people aged 65 or above
- Still has no efficient cure
- Sixth most leading death cause in world.
- Spreads in human body before 20 years when the symptoms are shown
- Can be found at early stage
Problem related to detecting biomarkers
- High cost and time
- Lack of online tools
- People without programming knowledge tends to have difficulties when using existing tools
3.Solution Overview
Better biomarker could provide an early sign before symptoms start to show, enabling more effective prevention treatment. So, we are proposing a online tool which can be detect the biomarkers for Alzheimer’s disease using the Ngs data set of Diseased samples and Control samples.
4.Solution Architecture
5.UI & UX
6.Requirements Analysis
6.1.Functional Requirements
Users
- Login and sign up
- View the results of provided data sample
Registered users
- Browse a table of provided data or upload their own files
- Manage their files and data
- Search bar
- Able to visualize the data using different methods (Box plot, Scatter plot, heatmap, ROC)
- Able to normalize the data
- Able to use different feature selection methods
- Make future predictions
- The output and results should be downloadable
6.2.Non-Functional Requirements
Scalability
- System should be utilized.
- Scalability of the system should be high in order to handle a huge amount of data.
Usability
- Take minimal time to detect biomarkers
- User’s data need to be recoverable.
- Good data processing pipeline
- No need for programming skills
Availability
- System should be available for 24/7.
- Able to access across the globe.
7.Use case Diagrams
8.Machine Learning Proposal
We are proposing a online tool where user can experimenting with different feature selection and classification methods to create a accurate model to detect the biomarkers and classify the samples.
9.MACHINE LEARNING WORKFLOW
