ABG-30806 Modern Statistics for the Life Sciences

Course

Credits 6.00

Teaching methodContact hours
Lecture15
Tutorial14
Practical56
Independent study0
Course coordinator(s)dr. ir. BJ Ducro
Lecturer(s)dr. ir. BJ Ducro
dr. B Engel
dr. ir. CA Maliepaard
dr. G Gort
dr. ir. P Bijma
Examiner(s)dr. ir. BJ Ducro
dr. G Gort
dr. ir. CA Maliepaard
dr. B Engel
dr. ir. P Bijma

Language of instruction:

English

Assumed knowledge on:

It is assumed that all students who take this course have a basic understanding of statistics and genetics. It is strongly recommended to take the courses MAT-15303 + MAT-15403 and MAT-20306 Advanced Statistics before taking part in the present course.

Contents:

Note: This course can not be combined in an individual programme with PBR-34803 Experimental Design and Data Analysis of Breeding Trials and/or PBR-32803 Markers in Genetics and Plant Breeding and/or PBR-33303 Quantitative and Population Genetics.
In this course students will learn about a number of statistical models and associated methods for statistical inference. Applications of models and methods in quantitative genetics and epidemiology will be discussed. For this purpose we will make use of plenary lectures, intensive computer practicals and working on two cases, i.e. analyse two data sets and write a small report on the analysis. This course consists of 6 modules where each module takes one week. The first three modules are general in nature and introduce the student to the Analysis of Variance, Regression analysis, the Likelihood concept, Mixed models, Generalized Linear Models and Bayesian statistics. The second three modules are dedicated to the application of the statistical methods to: the estimation of genetic parameters, QTL mapping, association studies and epidemiology.

Learning outcomes:

After successful completion of this course students are expected to be able to:
- explain the general differences between a linear model (LM), linear mixed model (LMM) and generalized linear model (GLM) in terms of model assumptions;
- present simple examples of data structures that should either be analysed by a LM, LMM or GLM;
- execute an analysis with a standard LM (ANOVA or regression), LMM (split-plot) or GLM (logistic regression or log linear model) for a given data set and interpret the results of such an analysis;
- explain the principle of maximum likelihood estimation;
- describe the differences between Bayesian and frequentistic statistics;
- explain how the heritability of a trait can be estimated in pedigreed populations and in a cross between inbred lines;
- explain how Quantitative Trait Loci (QTL) can be detected in an outcross population and in a cross between inbred lines and design QTL mapping experiments;
- design an experiment for estimating heritabilities or for QTL mapping;
- explain the difference between a linkage study and an association study.

Activities:

Theory will be introduced during plenary lectures. The theory will be illustrated during the computer practicals where data sets will be analysed. For the analysis use will be made of the statistical software R and other more specific programs.

Examination:

The examination consists of two written reports and an oral examination.
The contribution of each of these elements to the final mark is as follows: report 1 (25%); report 2 (25%); oral examination 50%.
For the oral examination a minimum score of 5 is required.

Literature:

A study guide and lecture notes will be provided.
Further, lecturers will put their presentations on MyPortal.

ProgrammePhaseSpecializationPeriod
Restricted Optional for: MASAnimal SciencesMScE: Molecule, Cell and Organ Functioning5AF
MASAnimal SciencesMScA: Genetics and Biodiversity5AF
MPSPlant SciencesMScD: Plant Breeding and Genetic Resources5AF
MBFBioinformaticsMSc5AF
MinorPeriod
Restricted Optional for: WUPBRBSc Minor Plant Breeding5AF