I tried to use Morphologika to format my data, but even the sample data set won't work (a message pops up "error in data") and the new software, EVAN toolbox, costs money to use. I want to be able to look at vectors to examine sex differences in shape to determine the severity and directionality of the sex variation at each landmark. But whenever I attempt to do the compare vector analyses, it says that there isn't enough appropriate data to analyze. Does anyone have this problem? Maybe there's a better software that I can use to study vectors/angles? Any suggestions?
The best way to deal with this is to ask Paul O'Higgins, the brain behind Morphologika, for help. He can really help you! Find his info at https://sites.google.com/site/hymsfme/contactus.
Or better yet here : https://sites.google.com/site/hymsfme/profpaulo'higgins
You can do this analysis most easily in geomorph, a geometric morphometrics package in R. It isn't quite clear from your question whether you are planning to compare two samples (males vs females) and to depict the difference between them or two compare two vectors for their length and direction. If you have two samples, one male and one female, and you want to know first whether they differ in shape, and if so, by how much, after superimposing the samples, you'd use the procD.lm function (which does a Procrustes Anova) to determine if the means differ. If they do, you could calculate the mean male and female shapes, using the mshape function, and plot the difference between the two means using plotRefToTarget. That draws the vector between the two means shapes that you just calculated. If you want to compare vectors, such as two regressions of shape on size (given ontogenetic series of males and females) you could use the pairwise.slope.test function. If you have multiple groups, with males and females in all, and you want to determine if the groups differ in the extent and/or direction of sexual dimorphism, you'd use the trajectory.analysis function.
Our Checkpoint software (below) can help you collect 3D landmark points as well as perform some 3D shape analysis on the collected points, including PCA, Procrustes, and more.
https://www.stratovan.com/products/checkpoint?utm_source=rg&utm_medium=ans&utm_campaign=meas3
Hello! I am sorry. What is vector analysis? I am just entering this field and I am learning geometric morphometrics. I saw this method in the manual of MorphoJ. But I don' t know what is vector analysis. Could you help me? Thank you very much!
The method, if I understand the question correctly, it not usually called "vector analysis", but it involves measuring the difference between two vectors by the angle between them, and testing the hypothesis that the vectors either are no more similar than expected by chance, or are no more different than expected by chance. MorphoJ and geomorph differ in their statistical analysis because MorphoJ tests the hypothesis that they are no more similar than expected by chance and geomorph tests the hypothesis that they are no more different than expected by chance. The length of the vectors can also be compared. This method is not an analysis, landmark by landmark, rather, it is an analysis of the vectors in the full space of shape data.
Dear professor@ Miriam Leah Zelditch :
I am very grateful for your reply! I also want to know what does the vector mean? Is the angle between the vectors the direction in which the deformation occurs?Which function can do this analysis in geomorph? Please forgive me for knowing too little. Thank you very much!
The angle between the vectors is a measure of the difference in the directions of shape change. If the two vectors point in the same direction, the angle between them is zero. So (in geomorph) the statistical test determines whether that angle exceeds what would be expected by chance. There are two functions that perform the analysis in geomorph, depending on whether the vectors are trajectories between means or regression coefficients (slopes). The functions that did the analyses up until this most recent version of geomorph (3.10) are two of the major changes The function for comparing trajectories between means is trajectory.analysis, which is now in the RRPP package (but geomorph depends on RRPP so this is a minor change except in the manual and in the arguments for the function). The function that compares slopes is now procDlm used in conjunction with anova.lm.rrpp, and pairwise (in the previous version, you'd use trajectory.analysis and advanced.procD.lm) You can see examples of these analyses in the vignette for the newest version of geomorph.
Dear professor@ Miriam Leah Zelditch :
Thank you for your help very much! I understand now. Best wishes!
- Badges
- Science method
- Similar topics
- Biological Science
- Ecology