GEOMETRIC MORPHOMETRICS IN R
January 20th-24th, 2020, Barcelona (Spain)
Course Overview
Concepts in geometric morphometrics will be taught using a series of original data sets and working in R for solving a series of tasks. The course will start with an introduction to R and will rapidly go into shape analysis with measurements, landmark data and outlines. The participants are welcome to bring their own data and problems so that we may find R solutions.
This is not an introductory course to Geometric Morphometrics, therefore basic knowledge of Multivariate Statistics, R and Geometric Morphometric is recommended in order to take this course.
Requirements
Graduate or postgraduate degree in any Life Sciences discipline.
Knowledge of Multivariate Statistics, R and Geometric Morphometrics. Participants with that are not familiar with R environment are strongly recommended to read the book ‘R for beginners‘ and practice before the course. Participants who are not familiar with Geometric Morphometrics are recommended to take first the course Introduction to Geometric Morphometrics.
All participants must bring their own personal laptop (Windows, Macintosh, Linux).
Register here (mention Bioblogia to get a discount)
Program
Monday, January 20th, 2020.
1. An Introduction to R / Image Processing / Organizing Morphometric Data.
1.1. Some Basics in R.
1.1.1. The R Environment.
1.1.2. R objects, Assigning, Indexing.
1.1.3. Generating Data in R.
1.1.4. 2D and 3D Plots in R; Interacting with the Graphs.
1.2. Organizing Data for Morphometrics.
1.2.1. Data-frame, Array and List.
1.2.2. Converting and Coercing Objects.
1.2.3. Read and Write Morphometric Data in R.
1.3. Image Processing in R.
1.3.1. Reading Various Image Files.
1.3.2. Obtaining Image Properties.
1.3.3. Modifying Image Properties: Contrast, Channels, Saturation Directly from R or by Interfacing R with Imagemagick.
1.4. Simple Tests, Simple Linear Modelling, Alternatives to Linear Modelling, an example using traditional morphometrics.
1.4.1. Defining size and shape using PCA and log-shape ratio approaches.
1.4.2. Getting stats and test outputs.
1.4.3. Testing assumptions of linear modelling.
1.4.4. Testing for allometry and isometry.
1.4.5. Solutions when assumptions of linear modelling are not met.
Tuesday, January 21st, 2020.
2. Landmark data.
2.1. Acquiring Landmark Data in R.
2.2. Plotting Landmark Configurations in 2 and in 3D.
2.2.1. Using Different Symbols and Setting the Graphical Parameters.
2.2.2. Labeling Landmarks.
2.3. Geometric Transformation with Landmark Configurations.
2.3.1. Translation.
2.3.2. Scaling using Baseline or Centroid Size.
2.3.3. Rotation.
2.4. Superimposing and Comparing Two Shapes.
2.4.1. Baseline Superimposition.
2.4.2. Ordinary Least Squares Superimposition.
2.4.3. Resistant Fit.
2.5. Representing Shape Differences.
2.5.1. Plotting Superimposed Shape with Wireframe.
2.5.2. Lollipop Diagrams and Vector Fields.
2.5.3. Thin Plate Splines and Warped Shapes.
2.6. Superimposing More Than Two Shapes.
2.6.1. Baseline Registration.
2.6.2. Full Generalized Procrustes Analysis.
2.6.3. Partial Generalized Procrustes Analysis.
2.6.4. Dimensionality of Superimposed Coordinates.
Wednesday, January 22nd, 2020.
2.7. Exploring Shape Variation and Testing Hypotheses.
2.7.1. PCA.
2.7.2. Multivariate Linear Modelling (Multivariate Regression and MANOVA).
2.7.3. Allometry free approaches (Burnaby correction).
2.7.4. Linear discriminant and Canonical Analysis.
3. Outlines.
3.1. Acquiring outline Data in R.
3.2. Fourier Analysis.
3.2.1. Principles.
3.2.2. Fourier Analysis of the Tangent Angle.
3.2.3. Radius Fourier Analysis.
3.2.4. Elliptic Fourier Analysis.
3.2.5. Reduction of Shape Variables.
3.2.6. Statistical Analysis of Shape Variation with Fourier Analysis.
3.2.6.1. Exploring Shape Variation and Testing Hypotheses.
3.2.6.2. PCA.
3.2.6.3. Multivariate Linear Modelling (Multivariate Regression and MANOVA).
3.2.6.4. Canonical Analysis.
Thursday, January 23rd, 2020.
3.3. Combining Landmarks and Curves.
3.3.1. Hybrid Methods between Fourier and Procrustes Analysis.
3.3.2. Sliding Semi Landmarks.
3.4. Solutions for Open Curves.
4. Specific Applications.
4.1. Testing Measurement Error.
4.2. Partitional Clustering.
4.2.1. K-means, Partition Around Medoids.
4.2.2. Mclust.
4.2.3. Combining Genetic, Geographic and Morphometric Data.
Friday, January 24th, 2020.
4.3. Modularity / Integration Studies.
4.3.1. Two-block Partial Least Squares.
4.3.2. Testing Among Various Sets of Modules.
4.4.Fluctuating Asymmetry and Directional Asymmetry.
4.4.1. Inter-Individual and Intra-Individual Variation.
4.4.2. Object and Matching Symmetry.
4.5.Bending Energy, Uniform and Non-uniform Shape Variation.
Literature
“Morphometrics with R“, Julien Claude. R. Gentleman, K. Hornik and G. Parmigiani, eds.
