WORKSHOP: ECOLOGICAL APPROACHES TO ANALYZING COMPLEX COMMUNITY DATASETS
JULY 25 -26, 2009 MSA/BSA meeting, Snowbird, Utah
Primary organizer: Peter Kennedy
For information, also contact Kabir Peay firstname.lastname@example.org
Organizers: Peter Kennedy, Greg Gilbert, Tom Bruns, Kabir Peay
The study of fungi has developed largely outside the realm of mainstream ecology, so many of the more commonly used statistical analyses are not well-known to the mycological community. The artificial division between these two fields, however, is quickly disappearing and mycologists are increasingly eager to apply rigorous statistical approaches to the data they are collecting. The goal of this workshop is to introduce mycologists to a diverse array of statistical tools available for analyzing the composition and structure of fungal communities. The emphasis of the workshop is to provide researchers with the opportunity to learn about as well as actually begin using various freely available web-based statistical packages. The workshop will consist of a small group of experts giving relatively brief presentations on their package of expertise followed by a "lab" period where researchers can begin to analyze data (their own or data provided) with the help of speakers. In the labs, attendees will work individually or in small groups and come back together at the end to discuss successes/challenges.
We have confirmed speakers for four analysis programs, EstimateS, Vegan in R, PhyloCom, and UniFrac. The workshop will be held over a two day period on July 25-26, the Saturday and Sunday before the main MSA/BSA meeting. The introductions and labs for each type of analysis will be held back-to-back, with each lasting around 3 hours total. Each day will begin at 8:30 am and last until 5 pm.
Session 1: Biodiversity Estimation & Community Structure - Susan Letcher and Jari Oksanen (Saturday July 25, 2009)
The first day of the FESIN workshop will focus on statistical tools for analyzing ecological communities. The two primary topics covered will be Estimation of Biodiversity and Analysis of Community Structure. Because the observed number of species in a sample or plot can be biased by factors such as sampling effort and species abundance distributions, ecologists have turned to statistical tools such as rarefaction and non-parametric richness estimators to allow more accurate comparison of diversity between study units. In the first part of the session we will introduce participants to the program EstimatesS, one of the most popular tools for implementing these types of biodiversity analysis. Multivariate statistical techniques have also revolutionized the analysis and visualization of ecological community structure. However, techniques such as non-metric multidimensional scaling (NMDS), principle components analysis (PCA) and analysis of similarity (ANOSIM) are complex to calculate or require expensive software. One free option, the R computing language, is challenging to learn but has become increasingly popular among ecologists. This is particularly due to the easy implementation of a wide variety of multivariate analyses through a community ecology package known as Vegan. The second half of the Session 1 will be dedicated to teaching participants the basics of the R language and learning how to implement the community ecology statistics implemented in Vegan.
Susan Letcher: Speaker's Notes
EstimateS Lab Notes
Jari Oksanen :Speaker's Notes
R and Vegan
Resources to help you learn and use R:
R starter kit: http://www.ats.ucla.edu/stat/r/sk/
Session 2: Community Phylogenetics (Sunday July 26) Steve Kembel and Elizabeth Costello
The second day will focus on the emerging field of community phylogenetics. Because closely related species are often more similar in their ecology than distantly related species, the phylogenetic structure of local communities can yield novel insights into the processes underlying community assembly. For example, communities in which species are more closely related than expected by chance may be produced when habitat filtering is working on functional traits that are evolutionarily conserved. The two programs that will be introduced on Sunday – UniFrac and PhyloCom – have become the most widely used in studies of bacterial and plant phylogenetic community ecology. Given that most fungal community studies involve the collection of phylogenetically useful DNA sequence data, these programs can be easily adapted to the needs of fungal ecologists. The session will be aimed at helping users learn to format their data, navigate the interface and interpret phylogenetic community statistics such as Net Relatedness Index (NRI) and the P-test generated by these packages.
Steve Kembel Speaker's Notes
Steve Kembel's website (http://phylodiversity.net/skembel/fesin/) with
more resources on Picante / Phylocom.
Please install the latest version of picante (0.7-1) and its dependencies in R. To do this, type the following in a R console:
Code for tutorial: picante-walkthrough.R
Phylocom tutorial by Sharon Strauss and Jean Burns: http://bodegaphylo.wikispot.org/Community_Phylogenetics
Community phylogenetics review papers
Cavender-Bares J, Kozak KH, Fine PVA, and Kembel SW (2009) The merging of community ecology and phylogenetic biology. Ecology Letters 12:693-715.
Emerson BC, Gillespie RG (2008) Phylogenetic analysis of communityassembly and structure over space and time. Trends in Ecology & Evolution 23:619–630.
Vamosi SM, Heard SB, Vamosi JC, and Webb CO. (2009) Emerging patterns in the comparative analysis of phylogenetic community structure. Molecular Ecology 18:572-592.
Webb CO, Ackerly DD, McPeek MA, and Donoghue MJ (2002) Phylogenies and community ecology. Annual Review of Ecology and Systematics 33:475–505.
UniFrac: http://bmf2.colorado.edu/unifrac/index.psp UniFrac is a web interface so there is no software to download / install.
Elizabeth Costello: Speaker's notes: