Statistics for biomedical scientists

25, 26, 27 October 2021

 

While looking at the experimental data we try to recognize patterns and to make sense of the observations. However, our intuition is often wrong, and so there is a need to impose some objectivity and the methods by which observations are converted into knowledge. Biostatistics is the branch of statistics that provides such methods. Understanding, interpreting data, drawing the conclusions are linked to comprehension of a prescribed set of mathematical expressions. This course will introduce participants to the basics of statistics on a more intuitive way without being encumbered by equations. This introductory course is to foster an appreciation for the role of statistics and associated data analysis approaches in research and in our everyday lives.

 

Aim of the course:

To provide an introduction to biostatistics by explaining statistical principles with a focus on the scientific interpretation of statistical tests rather than on the mathematical logic of the tests themselves.

 

By the end of the course, the participants should be able to:

            - Describe the main terms of descriptive statistics

            - Explain the p-value and its meaning

            - Explain the problem of multiple testing

            - Discuss the tests to account for multiple testing

            - Perform the main statistical test on a data set

            - Describe the Bayesian statistics

 

Target group: PhD candidates in the beginning of their PhD trajectory. The course is limited to 14 participants.

 

Prerequisites: The basics knowledge of statistics, probability theory and combinatorics, basics knowledge of the R programming language.

 

Duration of the course:  3 days

 

Location: GIGA B34 +5, Ghuysen room

 

Workload: 3 days x 8 hours per day = 24 hours

 

Educator: Michel Georges (GIGA – Medical Genomics, ULiège)

 

Course Syllabus/schedule

 

Day 1: Statistical significance 9:00-18:00 (13:00-14:00 - Lunch)

  • Definition of p-value and confidence intervals
  • Permutation tests
  • Bootstrapping
  • Accounting for confounders

 

Day 2: Multiple testing 9:00-18:00 (13:00-14:00 - Lunch)

  • Multiple testing: the issue.
  • Adjusting the thresholds
  • Exploiting the p-value distribution: false discovery rate

 

Day 3: Advanced topics 9:00-18:00 (13:00-14:00 - Lunch)

  • Likelihood based tests

Share this page