Data Analysis and Visualization

Much data is quantitative, and there is a wide range of methods available for the analysis of such data. After a brief introduction to data types and normalisation, a number of visualisation methods will be discussed. Next, methods will be introduced to find groups (clustering), dependencies (regression), significant differences between conditions (hypothesis testing) and to predict classes (classification). In addition, ways of assessing the relevance of findings and of interpreting results will be discussed. Students will learn to apply all these methods in practice in R.

• Describe qualitatively a number of analysis methods from statistics, clustering, and classification

• Apply the correct normalization and visualization methods given specific data types and research questions

• Implement quantitative analysis methods for a specific dataset in R scripts

• Interpret analysis results and their statistical significance and relevance through observation

• Assess the outcome of data analysis with different parameters and settings using appropriate measures and visualization methods

• Select the appropriate analysis methods to answer a given domain-specific research question

• Written test with open questions (50%)
• Assignment report (25%) Final project is done in groups of 2. A re-sit of the Final project is possible and deadlines to hand in the reworked assignment are the same as for the written test re-sit.
• Assignment report (25%) Projects are performed in groups of 2. Projects consist of both code blocks and text that are handed in. Projects take place in block 1 to block 6; the five (out of six) highest grades are used to calculate the average grade for the assignment. Re-doing projects from individual blocks is not possible. If this partial grade is insufficient, a replacement assignment will be organised, focusing on the topic of the block(s) for which the grades were lowest. Deadlines to hand in this replacement assignment are the same as the re-sit of the written exam.

BIF21806 Practical Computing for Biologists or INF22306 Programming in Python

• Slides, readers, handouts, and the online freely-available book 'Modern Statistics for Modern Biology', written and kept up-to-date by Susan Holmes and Wolfgang Huber, https://www.huber.embl.de/msmb/.