BIOS4000 – Design and analysis of biological studies

Course content

This course is a thorough introduction to design of biological studies and statistical analysis in biology. The focus is on the use of statistical models for analyzing biological patterns and processes. Students are taught fundamental skills in modern biological research through project work, exercises and computer exercises. The statistical environment R is used throughout the course.

Learning outcome

After completing this course, you are expected to:

  • Understand the difference between observational studies and experiments, and be able to assess the results from different types of studies in a biological context
  • Understand the importance of the terms pseudoreplication, confounding effects, experimental control, randomization, sampling skewness, stratified sampling and blocking in analysis of biological studies
  • Be able to carry out Monte Carlo simulations to assess different study design and statistical models
  • Be able to fit biologically relevant models based on the normal, binomial, and Poisson distributions (GLM), and calculate linear contrasts and predictions with confidence intervals, as well as evaluate how well these models fit the data (goodness of fit).
  • Know how to fit hierarchical models with normally distributed response variables and interpret these
  • Know how to assess the sources of bias in models fitted to biological data, including the effects of sampling skewness, measurement error in the predictive variables (attenuation) and loss of study units during the course of the study.
  • Be aware of common fallacies in statistical inference.
  • Know about principles for choice of models depending on specific research questions.
  • Be able to comprehend literature based on the contents of the course and be able to present this as a lecture for fellow students.

Formal prerequisite knowledge

STK1000 – Introduction to Applied Statistics or equivalent.

A background in elementary programming equivalent to the content of BIOS1100 – Introduction to computational models for Biosciences is strongly recommended.

Other recommended background courses are BIOS1110 – Celle- og molekyl?rbiologi (Cell and Molecular Biology), BIOS1120 – Fysiologi (Physiology), BIOS1140 – Evolusjon og genetikk (Evolution and Genetics) and?BIOS2100 – General Ecology.

Overlapping courses

Teaching

The course consists of:

  • Lectures

  • Problem-solving with guidance

  • Mandatory hand-ins

  • A compulsory literature review project resulting in a lecture a short lecture for fellow students (individually or in groups of two).

The curriculum is in English and all reports must be written in English.

Attendance to the first lecture is mandatory. If you are unable to attend the first lecture you will lose your seat on the course if you do not inform the student administration?studieinfo@ibv.uio.no?prior to the first lecture.

Approved mandatory course work is valid for 3 years.

Access to teaching?

A student who has completed compulsory instruction and coursework and has had these approved, is not entitled to repeat that instruction and coursework. A student who has been admitted to a course, but who has not completed compulsory instruction and coursework or had these approved, is entitled to repeat that instruction and coursework, d