Hands-on-Biomedical-Data

Practical exercises for the course "Hands-on Biomedical Data - Resources and Analysis Tools"

Comments up front

Make sure you read instructions in detail. Especially getting the Setup right.

Exercises

• Day 1: Basic R programming and visualizations
• Day 2: Introduction in differential expression analysis
• Day 3: A more complex example
• Day 4: Interaction effects
• Day 5: Assignment - analyze a dataset of your choosing

Setup

You can run the practicals on: * Your personal computer(ideal case) * Within Galaxy * On the PLUS Server RICARDA

Evaluation

For the evaluation, you will get points based on the exercises indicated like this: Exercise X:

• The evaluation is based on a protocol that you will prepare.
• In this protocol your should address all exercises. Each exercise counts for 1 point unless otherwise stated.
• Usually exercises are just one or two plots. If you are asked to respond to questions, max. 2-3 sentences per exercise should be sufficient.
• You can should ideally use Markdown (see instructions below) to create the protocol. This contains code and plots together and makes it very easy to track and evaluate your progress.
• Alternatively, you can copy/paste plots and write answers in Powerpoint, Word or similar (convert and submit a PDF file). In this case, you also have to submit R Scripts which document your code.

Instructions

• Save one R script for each day to not mix exercises and R sessions from different days.
• Save the R scripts as day1.R, day2.R, day3.R, day4.R, and day5.R.
• While you can execute commands from your script in any order, make sure your finally submitted script runs through from top to bottom if started from an empty environment!
• Submit all files through Blackboard (see deadline there).

Markdown

• If you work on your personal computer, you can combine code and answers to questions using Markdown. See the following: Markdown instructions
• If you do use Markdown (on your personal computer), create HTML files using File > Knit Document in R. Please DO NOT use File > Knit Document on Ricarda.

The example dataset

In this part of the practical, we will study transcriptomics data of structural cells in mice upon cytokine stimulation from Krausgruber & Fortelny et al., Nature, 2020.

Basic analyses steps covered

• Quality control using sample correlations and dimensionality reduction
• Data normalization
• Differential expression
• Model diagnostics and quality control
• Plotting of results
• Interpretation of top genes
• Gene set enrichment analysis

How to get help?

• Most commands should be explained in this practical.
• If you do not understand certain functions, type the question mark plus the function name in R, e.g.: "?median".
• If you need additional commands, Google is your friend.
• Also consult this list of function names, which contains key functions relevant for this course.
• Don't forget to raise you hand if lost!