The documentation for that software is rather minimal! That was because it was developed for a specific project and I had good intentions of coming back to it soon to make a more complete program with the usual help file.
Answers to the three questions: 1) As with the use of any Fourier method, you need to standardize the starting point on an outline, trace the curve in the same direction (clockwise or counter-clockwise), orient the outlines in some standard way, and scale for size. The 2D program efa had options to attempt to do this standardization automatically. There were problems with that so this software assumes that one does that by during the creation of the data file. One problem with the other software was that it placed the starting point at the end of the PCA axis - but the choice of which end is poorly defined. More work needs to be done on this. Until then one should make sure one collects data in a standardized fashion. 2) Yes, the curves should all be placed in the same file. Each curve begins with its own header line (see the next question) that indicated the number of points. 3) That header line is in the NTS file format. The initial '1' means a rectangular data matrix. The next '1' means the matrix has one row. The '180' means 180 columns. The '0' means no missing data. The 'dim=3' does mean 3 dimensions. As a result, the 180 columns actually corresponds to 60 points along the curve - each consisting of 3 coordinate values. I hope this helps. ---------------------------------------------- F. James Rohlf, Distinguished Professor Dept. Ecology & Evolution, Stony Brook University Stony Brook, NY 11794-5245 http://life.bio.sunysb.edu/ee/rohlf > -----Original Message----- > From: morphmet [mailto:[EMAIL PROTECTED] > Sent: Friday, October 27, 2006 3:55 PM > To: [email protected] > Subject: Curve analysis > > Dear All > > I am interested in using EFA3D to calculate Fourier > coefficients for some curve data taken on human crania. > However, this program only has limited supplementary > information as yet so I have a few questions: > > 1) Do all homologous curves first need to be placed in the > same frame of reference given three common landmarks before > co-efficients can be computed? > > 2) Can I then place all homologous curves in the same input > file even if they each have varying number of points making them up? > > 3) If anyone has used this program before perhaps you can > spell out what the first line of the input file refers to: > i.e. 1 1 180 0 DIM=3(that's presumably dimensions?) > > Perhaps some people have used alternative programs for such > analysis? I am aware of several 2D versions but no other 3D ones. > > Any comments, suggestions and advise are gratefully accepted! > Many thanks in advance > Noreen > > -- > Noreen von Cramon-Taubadel > PhD Candidate > Leverhulme Centre for Human Evolutionary Studies Department > of Biological Anthropology University of Cambridge The Henry > Wellcome building, Fitzwilliam street Cambridge CB2 1QH > Phone: +44 1223 764700 > http://www.human-evol.cam.ac.uk/voncramon.htm > > > > -- > 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
