JClavel
commented
4 months ago Hello,
The matrix must have the species in rows and coordinates in columns. Just use the superimposed coordinates (you can convert arrays to matrices using the "vecx" function in package Morpho). I think you can follow examples in the "mvgls", "gic" or "eic" help page. There's also a tutorial with examples on the phyllostomid bats dataset (mandibles in 2D) in the supplementary of one of our papers (http://dx.doi.org/10.5061/dryad.jsxksn052).
HTH
Best regards,
Julien
De : emaseculi @.> Envoyé : mardi 28 mai 2024 18:30 À : JClavel/mvMORPH @.> Cc : JClavel @.>; Mention @.> Objet : [JClavel/mvMORPH] Analyzing 2d morphometric data (Issue #16)
Hi @JClavelhttps://github.com/JClavel how are you? I been working with the abeliaruids maxilla evolution using 2d geometrics morphometrics technique. I employed the 90% of phylogenetics PCs to explore the evolution of this structure using the function of mvMORPH (mvOU, mvEB and mvBM) in different hyphotesis of evolution. Recently I discovered the mvgls function and I want to compore how the result change when i considere all the dimension of my data with the Penalized Likelihood Framework approach, additionally the PhyloPcs can produce bias results. But I do not know how can i should convert the procustes coordinate (matrix, data frame, etc) and if the coordinates should be in the columns or rows. Moreover i do not understand how i have to use the arguement formula. If you can give some advice i really appreciate
Best regards!
— Reply to this email directly, view it on GitHubhttps://github.com/JClavel/mvMORPH/issues/16, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ACSCJSN445BUT56AKNQLFJLZESWJFAVCNFSM6AAAAABINI7TU6VHI2DSMVQWIX3LMV43ASLTON2WKOZSGMZDCNBYGY3DOMA. You are receiving this because you were mentioned.
emaseculi
Hi @JClavel how are you? I been working with the abeliaruids maxilla evolution using 2d geometrics morphometrics technique. I employed the 90% of phylogenetics PCs to explore the evolution of this structure using the function of mvMORPH (mvOU, mvEB and mvBM) in different hyphotesis of evolution. Recently I discovered the mvgls function and I want to compore how the result change when i considere all the dimension of my data with the Penalized Likelihood Framework approach, additionally the PhyloPcs can produce bias results. But I do not know how can i should convert the procustes coordinate (matrix, data frame, etc) and if the coordinates should be in the columns or rows. Moreover i do not understand how i have to use the arguement formula. If you can give some advice i really appreciate
Best regards!