Hi, In the meantime I received an answer, reading Adams 2014.
Thanks 2016-05-20 13:08 GMT+03:00 filipe oliveira <filipe.fi...@gmail.com>: > Hi Emma, > > Thank you. My co-worker calculated the nodes, MorphoJ does not seem to > provide that info but some papers do mention they used it for that end (if > I am not wrong). > > By the way, we obtained a K=0.54 and P=0.001. If I understand it > correctly, that indicates that there is substantial convergence in the > data, and that pattern observed isn't random. Does that mean then that the > phylogenetic signal is week? Or am I interpreting things the wrong way. > > In MorphoJ I obtain a significant phylogenetic signal. > > Thanks > Filipe > > 2016-05-20 8:43 GMT+03:00 Emma Sherratt <emma.sherr...@gmail.com>: > >> Hi Filipe, >> >> I believe you cannot directly. When you use the "Map onto phylogeny" >> option in MorphoJ, you will have appear in the scene tree the tree icon, >> and two dataset icons below that. "BranchDiffs" and "IndContrasts". From >> the MorphoJ helpGuide: >> *Each analysis produces two output data sets: one with the values of the >> changes along the branches of the phylogeny for the values of shape >> variables, centroid sizes and covariates in the original dataset, and the >> other with scores for phylogenetically independent contrasts (Felsenstein >> 1985) for the same variables.* >> >> You could retroactively calculate node position in PC space based on the >> IndCont and branch lengths. However it may be simpler to run >> plotGMPhyloMorphoSpace in *geomorph* in R and extract the ancestral >> states (which although are calculated by maximum likelihood using >> *phytools* fastanc *) - note that the anc.states matrix returned are >> states prior to PCA, so all you'd need to do is: >> >> all.data <- rbind(two.d.array(mydata), anc.states) # combine tips and >> nodes >> pcdata <- prcomp(all.data)$x # PCA >> >> Em >> >> * ML and weighted squared change parsimony are equivalent assuming BM >> model of evolution >> >> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >> >> Emma Sherratt, PhD. >> >> >> Postdoctoral Researcher in the Keogh Lab >> <http://biology-assets.anu.edu.au/hosted_sites/Scott/> >> Division of Evolution, Ecology & Genetics >> Research School of Biology >> 116 Daley Road >> The Australian National University >> Acton, ACT 2601 >> AUSTRALIA >> >> email: emma.sherr...@gmail.com >> tel: +61 2612 53029 >> mob: +61 4234 19966 >> Twitter: @DrEmSherratt <https://twitter.com/DrEmSherratt> >> >> Caecilians are legless amphibians... >> >> * __ >> (\ .-. .-. /_") >> \\_//^\\_//^\\_// >> `"` `"` `"`* >> >> learn more about them here: www.emmasherratt.com/caecilians >> >> >> >> >> On 17 May 2016 at 01:32, filipe oliveira <filipe.fi...@gmail.com> wrote: >> >>> Hi, >>> >>> do you know how to save the PC values of the ancestral nodes estimated >>> with Square change Parsimony in MorphoJ? >>> >>> Thank you. >>> Best >>> Filipe >>> >>> -- >>> MORPHMET may be accessed via its webpage at http://www.morphometrics.org >>> --- >>> You received this message because you are subscribed to the Google >>> Groups "MORPHMET" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to morphmet+unsubscr...@morphometrics.org. >>> >> >> > > > -- > Filipe Oliveira da Silva > > M.Sc. in Evolutionary Biology (MEME) > Uppsala University (Sweden) and University of Munich - LMU (Germany) > > B.Sc. (Honors). Universidade Federal de Vicosa (Brazil). > > "In evolution, selection may decide the winner of a given game but > development non-randomly define the players" Alberch (1980) > -- Filipe Oliveira da Silva M.Sc. in Evolutionary Biology (MEME) Uppsala University (Sweden) and University of Munich - LMU (Germany) B.Sc. (Honors). Universidade Federal de Vicosa (Brazil). "In evolution, selection may decide the winner of a given game but development non-randomly define the players" Alberch (1980) -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org --- You received this message because you are subscribed to the Google Groups "MORPHMET" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet+unsubscr...@morphometrics.org.
