Turin 2009 Geometrics Morphometrics Workshop

Turin University
31 August-2 September 2009

Geometric morphometrics is a set of statistical tools for the analysis and visualization of morphological variation. For its power it has been considered a 'revolution' in the field of morphometrics. The number of scientific papers using geometric morphometrics has increased exponentially over the last couple of decades. Geometric morphometric analyses have been successfully performed in a variety of disciplines ranging from taxonomy, palaeontology and evodevo to forensics, ergonomics and archaeology.
A short course in geometric morphometrics will be held at Turin University from August 31st to September 2nd 2009, immediately after the end of the ESEB 2009 meeting. The workshop will cover both introductory and more advanced topics. Lectures and practicals will be given by Chris Klingenberg (University of Manchester, UK) and Andrea Cardini ( Università di Modena e Reggio Emilia, Italy, and the Hull York Medical School, UK).

WORKSHOP PROGRAM
a) INTRODUCTORY TOPICS

• - Introduction on morphometrics: the analysis of size and shape variation in organisms.
• - Landmark-based geometric morphometrics: data collection; computation of size and shape
• variables using the generalized Procrustes analysis.
• - Summary and visualizartion of shape variation: principal component analysis and thin plate spline deformation grids.
• - Statistical testing: an example using permutation tests for mean size and mean shape differences of two samples.

b) ADVANCED TOPICS

• - Test for measurement error.
• - Analysis of symmetric structures.
• - Phylogenetic generalized least square.

c) STUDENTS PRESENTATIONS

• - Participants who are willing to try their own analysis on their study data or the example files provided by the instructors will be given the possibility to briefly show their work in a short (5min.) presentation.

A registration fee of 100 euros is requested to cover the costs of the workshop organization.
Participants will have to provide accomodations, meals etc. by themselves. Software and example data will be avaialable for participants to practice and try their own analyses.

The total numer of participants is limited to 25. A request of pre-booking should be sent to Prof. Claudia Palestrini, Dipartimento di Biologia Animale, Università degli Studi di Torino, via Accademia Albertina 13, Turin, Italy before 30 April 2009.

Participants will be accepted following a strict booking order. After booking confirmation you will be requested to pay the registration fee (more information on this will be provided directly by Prof. Claudia Palestrini).
Please, if you are interested to participate, send your pre-booking request as soon as possible. We look forward to meeting you in Turin.

Que son los Analisis de Componentes Principales "Evolutivos"?

Confieso mi ingenuidad e ignorancia: me ha sorprendido leer que existen "Análisis de Componentes Principales Evolutivos"!
La fuente de la iluminación es el articulo:
Jesse M Meik & Andre Pires-daSilva. 2009. Evolutionary morphology of the rattlesnake style. BMC Evolutionary Biology 9:35 doi:10.1186/1471-2148-9-35
Article URL http://www.biomedcentral.com/1471-2148/9/35

Así que me dispuse a leerlo y averiguar que es este método, pues solo con el nombre me inspira una predisposición psicológica muy favorable.

Los datos: 8 medidas de los cascabeles de serpientes (el estilo), cuatro medidas de otras partes del cuerpo y el numero de escamas subcaudales. Nada extraordinario: distancias entre marcas y conteos.
Los ajustes: todas las variables transformadas en logs base 10. Se calcularon los residuales de la regresión de las variables contra una de ellas (snout–vent length) como una manera de estandarizar el efecto del tamaño.
La referencia: una filogenia ad hoc para colocar 34 especies en cinco grupos.
Los análisis ACPE: un circo extraordinario de tres pistas y necesito que alguien me dirija para disfrutar el espectáculo.
No quiero arriesgarme a hacer una mala traducción o una mala interpretación. La explicación de los autores dice:

"In cross-species comparisons, EPCA has an advantage over standard PCA in that EPCA extracts axes that maximize evolutionary change rather than ordinating static variation. In this variant of PCA, ancestral states are first reconstructed using squared-change parsimony and then eigenanalysis is performed on vectors of evolutionary change along each branch [18]. Thus, by incorporating phylogenetic information, EPCA is potentially able to reduce non-independence of cross-species data sets due to similarity through common descent."

El software: Dyreson E, Maddison WP. 2006. Rhetenor package for morphometrics. 1.11 edition: Mesquite module.
Los resultados: Se graficaron los dos primeros componentes y se identificaron las variables correlacionadas con estos, como es usual.

Las interpretaciones:

"Thus we interpret the main spectrum of interspecific differences in style shape to reflect a continuum from relatively longer styles with deeper prongs, wider gaps between prongs (i.e., greater angle), more attenuated prongs, and greater constriction (or notching) near the juncture with the prongs and the anterior body of the style, to relatively shorter styles with shallow, but broad, prongs (or, in the extreme condition, prongs absent), and less constriction at the juncture of the prongs and the anterior
style body (see Fig. 1)."

Seguro que alguien ya había oído de este método! Que opinan? Realmente es la manera de revelar la información de las diferencias evolutivas entre linajes? Que otras alternativas existen?

Mi preferencia es tradicional: reconstruir la serie transformacional optimizando los cambios de estados en el cladograma. El diagnostico de los estados por supuesto puede ser referido a un espacio morfometrico. Mi preferencia es el morfoespacio mediante los Análisis de Variables Canonicas extraido del espacio configurado por las variables que describen la característica estudiada.

Espero sus comentarios y aclaraciones, pues seguro se me escapa algo sobre los ACPE.

Meik, J., & Pires-daSilva, A. (2009). Evolutionary morphology of the rattlesnake style BMC Evolutionary Biology, 9 (1) DOI: 10.1186/1471-2148-9-35