-------- Original Message -------- Subject: RE: Of Landmarks and Semilandmarks. Date: Mon, 13 Apr 2009 11:51:34 -0700 (PDT) From: F. James Rohlf <ro...@life.bio.sunysb.edu> Reply-To: <ro...@life.bio.sunysb.edu> Organization: Stony Brook University To: <morphmet@morphometrics.org> References: <49e385e0.4000...@morphometrics.org> The points used for computation in tpsRelw must all be provided as landmark points in the tps file. The original purpose of the 'draw curves' feature was just a way to draw informal curves on a specimen. Once the semilandmark method was introduced it became a convenient way to draw the curves - however, the tpsUtil program must be used to move their position in the tps file from 'curves' to additional landmark points. After that conversion it is essential that the landmark points have matching numbers. Point number 10, for example, must correspond across all specimens or else subsequent computations will make little sense. This means in practice that one must have the same number of landmarks and the same number of semilandmarks on all specimens and the order of digitization must be identical across all specimens. One can slide points along an arbitrary number of curves as long as the points all correspond. One does not have to use the tpsUtil program. What it does is pretty simple so one can create a slider file by hand. The tpsUtil program just provides a convenient graphical interface that may reduce the chance for making an error. ========================= F. James Rohlf Distinguished Professor, Stony Brook University http://life.bio.sunysb.edu/ee/rohlf
-----Original Message----- From: morphmet [mailto:morphmet_modera...@morphometrics.org] Sent: Monday, April 13, 2009 2:35 PM To: morphmet Subject: Of Landmarks and Semilandmarks. -------- Original Message -------- Subject: Of Landmarks and Semilandmarks. Date: Mon, 13 Apr 2009 10:35:06 -0700 (PDT) From: Parzer, Harald Friedrich <hpar...@indiana.edu> To: morphmet@morphometrics.org Dear All, I am rather new to Geometric Morphometrics...at the moment I am trying to use a semilandmark + landmark approach to estimate evolutionary rates in different insect traits. However, I am stuck at a very basic problem: I created outlines with the "draw curves mode" and landmarks in tpsDig (several curves, landmarks in between). I then wanted to create a "slider" file in tpsUtil, which is necessary for further analysis in tpsRelw (?). However, the semilandmarks did not show up. Thus, I thought of something else: I created changed all the SLM into landmarks and then tried to assign the SLM. However, the landmarks assigned do not correspond between the individuals (thus, the points are clearly not homologous - the slided...already), which will change the alignment. So, my question might be: what should I do? Or, can I have several curves - which I do - on the same trait, without have the problem of sliding and slipping LM? How can I keep the LM stable across individuals? Is there a way to load the LM/SLM data I created in tpsDig directly into tpsRelw, avoiding tpsUtil (if I do it, it says: there is no slider file)? Thank you very much and sorry about the question - I checked the archive of morphmet several times, but could not find any information: harald -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
-- Replies will be sent to the list. For more information visit http://www.morphometrics.org
