I work on morphology work and use Morphometric data to analysis, such as PCA. Now could I use a matrix produced by MorphoJ to build a phylogenetic tree?
Could you please tell me which literature could tell me or which software I can use?
Only some of morphometric data can be used for meanful phylogenetic analysis.
Read for an example:
Article The Holy Grail of the Perfect Character: the Cladistic Treat...
The simplest way of doing this is by estimating mshape for all individuals under a morphological species. mshape is not a statistical mean but the coordinate chosen among existing coordinates of same landmark using the least square technique. Once you get the mshape of all species, you can estimate the euclidean distance matrix per-say using dist function available in base R. Once you get the distance matrix, you can visualize the species relationship in morphological space using clustering algorithms like hclust. I have found this way more meaningful among several available methods. you can see my dissertation at :https://scholarworks.uark.edu/cgi/viewcontent.cgi?article=4720&context=etd for details.
Binod Regmi Thanks for your detail methods, but maybe somewhat difficult for me to do so for R. And do you have any recommended software?
Alex Ignatov Sorry for my English, what I want to express to use the matrix produced by Morphometric Software such as MorphoJ to build. I don't want to build a phylogenetic tree subjectively.
You can not build a phylogenetic tree via morphometric data! Because characters can easily change in various environmental conditions (e.g. Stem height, petal color, hair types and even their variation, etc). Homoplasy is another meaningful reason than can create a great bias in the results!
You can only use such kind of data to show the similarity or dissimilarity between or among species and accessions using phenograms.
Useful software for morphometric analyses are as follow:
MVSP, R, PAST, NTSYS, GeneAlex, etc.
Atena Eslami Thanks for your advice and I don't want to use artificial characters and I want to use the matrix obtained from landmarks and semi landmarks. And many specimens would be involved in the study, so even the variation could be omitted.
Your welcome, i cannot understand what do you mean by landmarks, but if you need any help in software (s) manuals, please don't hesitate to ask.
If you want obtain a phylogenetic hypothesis, TNT has a version where you can employ landmarks matrix. You can download it from http://www.lillo.org.ar/phylogeny/tnt/
NTSYSpc can be used to discover pattern and structure in multivariate data. The program takes advantage of the Windows environment and allows long file names and the processing of large datasets. NTS data files are ASCII files that can be shared with other programs.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix