Hi all, I have examined several phenotypic traits of a set of species belonging to a genus. For each species there were multiple observations. Later I performed statistical tests (ANOVA) to know if the species show significant variation for a given phenotypic trait.
The critics of the work say that statistical tests used in this study are not proper to compare phylogenetically dependent samples, so Independent contrast methods (Felsenstein, 1985) should be used to correct different divergence time among species.
While use of independent contrasts for comparison of two traits has been frequently applied, but I am confused how to carry out statistical tests on the phylogenetically corrected data, i.e., independent contrasts. I can think of using mean values for each species for computing contrasts, but how to go ahead with these contrasts?
Your comments will be highly appreciated.
Thanks
Vivek
Dear Vivek,
hope this work is helpful:
Yates ML, Andrew NR, Binns M, Gibb H. Morphological traits: predictable responses to macrohabitats across a 300 km scale. Higley L, ed. PeerJ. 2014;2:e271. doi:10.7717/peerj.271.
Best wishes,
Nils
Would a phylogenetic least squares approach work? It's commonly used for geometric morphometrics (i.e. multivariate data) across species of a clade.
Hi Vivek,
Going with with what Dr. Sergei said, it is not clear whether you have a phylogenetic framework (a phylogenetic tree!) at hand to use the method of ICs.
There are several software packages that help test evolutionary hypotheses and perform comparative studies (using comparative phylogenetic methods). But, all depends on what kind of comparison you have to do.
[Useful Ref:
Applications of Phylogenetically Independent Contrasts: A Mixed Progress Report
Author(s): Robert E. Ricklefs and J. Matthias Starck
Source: Oikos, Vol. 77, No. 1 (Oct., 1996), pp. 167-172 ]
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA