Multivariate Statistics in Ecology and Quantitative Genetics SS 2016

Lecture with Exercises (Vorlesung mit Übung)

Instructors

Time and rooms

June 27 2016 until July 15 2016
Lecture: Each day from 9:00 to 10:30
Exercises: Usually Tuesday to Friday from 10:45 to 12:00 and from 1 pm to 5 pm (sometimes 6 pm).
Rooms: on Mondays in E02.023, on Tuesdays, Wendesdays and Fridays in E03.052, on Thursdays in the first and the third week in D00.013, and on the Thursday in the 2nd week in G 00.031.
Exercises: Usually Tuesday to Friday from 10:45 to 12:00 and from 1 pm to 5 pm (sometimes 6 pm). We will have the lab room D00.027 for the exercise sessions.

Exam

The make-up exam will be on 15. September at 2:00 pm in lecture hall B01.019. To take the make-up exam you need to register by e-mail to Dirk Metzler by 12. September. You will have 90 minutes to answer the questions. Participants can bring non-programmable pocket calculators an their personal assortment of formulas on an A4 sheet. This A4 can have any contents on both sides BUT ONLY IN YOUR OWN HANDWRITING! (no copy, not print-out, not another person's handwriting) Please note: If you fully attend the course (which is required to get the ECTS points) you will need the time between the 11 a.m. and 5 p.m. to solve exercises, which will be given in the morning lectures and discussed in the afternoon courses.

Requirements

Basic knowledge of statistics. If you are not already familiar with the free R software, it may be a good idea to install it on your computer and get a bit familiar with it before the course starts. Some examples (data file FinchesSulloway.txt), R tutorial, Martin's introductory R course.

In the exercises you will practice to analyse data with the methods listed below using the R software. We will discuss the interpretations of the results and problems that may arise during the analysis. In the lectures we will explain the theoretical concepts behind the methods and show how these methods can be applied.

If you would like to discuss analyses of your own datasets during the course, please contact D. Metzler as soon as possible.

Fundamental methods: Multivariate linear regression, Model Selection, Analysis of Variance (ANOVA), Experimental designs (nested, balanced, unbalanced,...), Graphical Inspection of Fitted Models, Generalized Linear Models (GLM), Mixed Effects Models
R-package: lme4

Special Methods for complex ecological datasets:
ANCOVA, Redundancy Analysis, Principal Component Analysis (PCA), Canonical Correspondence Analysis,....
R-package: vegan (see also the ecology task view)

Analysis of Gene Expression Data
Multiple-Testing Problems, Variance-Stabilizing Normalization, Regularization, reconciliation with gene ontologies (GO)...
R-packages: DESeq, vsn, limma, multtest, GOstats

Special Methods in Genetics: Mapping Quantitative Traits Loci (QTLs).
Keywords: Haley-Knott regression, Composite Interval Mapping, Multiple-QTL models,...
R-package: qtl (see also the genetics task view)

Language: English

Course material

Basics on anova: slides
Linear Models: slides, R commands for Darwin finches, Darwin finches data, exercises sheet 1, abcdx.txt, bp.txt
Balanced design ANOVAs, parameter transformations: slides, R-script, exercises sheet 2, bacteria_trainig.txt, bacteria_predict.txt
Generalized Linear Models: slides, exercise sheet 3, TbDeerAndBoar.txt
Hotellings T2-test slides, R-script, raspberry.csv exercise sheet 4, grapes.txt
Principal Component Analysis (PCA): slides, R-script, exercise sheet 5, HeightShoeWeight.txt EWU.txt
PCA Regression: slides
Redundancy Analysis (RDA): slides, R-script, exercise sheet 6, EWU.txt, HSWoutlier.txt, artificialFishes.txt, RIKZGroups.txt
Correspondence Analysis (CA): slides, R-script, MexicanPlants.txt
Canonical Correspondence Analysis (CCA): slides, R-script, exercise sheet 7
Mixed-effects models slides, lme4a.R, mcmcglmm.R, exercise sheet 8, fruits.txt
QTL Mapping slides, R-script, exercise sheet 9
RNA-Seq gene expression slides , R-script

Material will be added during the course. Here is the course material from 2015.

Literature

Introductory statistics with R / Peter Dalgaard (Springer 2002)
Biometry: The Principles and Practice of Statistics in Biological Research (3rd Ed.) / Sokal, Rohlf, (Palgrave Macmillan 1995)
Mixed effects models and extensions in ecology with R / Alain F. Zuur, Elena N. Ieno, N. J. Walker, Anatoly A. Saveliev, Graham M. Smith. (Springer 2009)
Modern applied statistics with S / W. N. Venables ; B. D. Ripley. - 4. ed. (Springer 2002)
A Guide to QTL Mapping with R/qtl / Karl W. Broman & Saunak Sen (Springer 2009)
Analysing ecological data / Alain F. Zuur ; Elena N. Ieno ; Graham M. Smith. (Springer 2007)
Genetics and analysis of quantitative traits / Michael Lynch ; Bruce Walsh. (Sinauer 1998)
Introduction to quantitative genetics / D. S. Falconer and Trudy F. C. Mackay. - 4. ed. (Pearson 1996)
Numerical ecology / Legendre & Legendre (Elsevier 1998)
Statistical Genetics of Quantitative Traits / Wu, Ma & Casella. (Springer 2007)
Bioinformatics and Computational Biology Solutions Using R and Bioconductor / R. Gentleman, V.J. Carey, W. Huber, R.A. Irizarry, S. Dudoit (Eds.). (Springer 2005)
Likelihood, Bayesian, and MCMC Methods in Quantitative Genetics / D. Sorensen, D. Gianola (Springer 2002)
Mixed effects models in S and S-PLUS / Jose C. Pinheiro ; Douglas M. Bates. (Springer, 2004)

Last update: 5. August 2016