Kidney transplant is often the treatment of choice to people with end-stage kidney disease. Understanding the characteristic of stable patients and patients who experienced rejection can be one of the ways to understand more about kidney transplant rejection. In this assignment, select a microarray or RNA-seq data that measure gene expression of patients' cells and will use the gene expression to predict for the patient outcome (i.e., stable versus rejection) using a selected machine learning approach.
This multi-media discipline report will consist of three components:
- An executive summary
- A shiny app
- Video presentation (worth 10%)
Some suggestions of gene expression data are
|RNA-seq||GSE120396 (provided in lab)||GSE131179 (provided in lab)||GSE120649 (provided in lab)||GSE86884|
Part A) Prepare an executive summary with no more than 750 words providing an overview that will highlight a particular analytical approach addressing the question of interest and a brief guide to the shiny app. Points to includes are:
- Provide a clear statement of the question you intend to address or the topic that you intend to focus on your multi-media discipline report.
- What is your approach to addressing the question stated in (1) and what is the key technique in your approach (e.g. random forest, lasso, Bayesian network etc.)? Select ONE method and provide a concise technical description.
- Identify potential shortcomings or issues associated with the data analytics that you have performed and discuss a possible approach to address the issue. Here, a strategy doesn't necessarily refer to a model, but it must address the issue.
- Create and describe the interactive graphics (or shiny app) that illustrate one aspect of your report, and please provide the link to the shiny app (this can either be a web page or a GitHub link).
Part B) Create a Shiny app and demonstrate this app in a video presentation.
Part C) Oral presentation submitted via video recording (length – 3 mins), this component of the assessment is part of your discipline assessment, and the video recording alone is worth 10% of your final mark in DATA3888. Here, you will be assessed on
- the technical content, and
- your presentation skills, measured by criteria covering slides, the flow of information, engagement with the audience and clarity of message.
The technical content will include a "student-led learning" component where you will describe key ML techniques for your peers as well as a clear demonstration of your shiny app.