----- Forwarded message from Stefan Schlager <stefan.schla...@uniklinik-freiburg.de> -----

Date: Mon, 17 Jun 2013 04:39:18 -0400
From: Stefan Schlager <stefan.schla...@uniklinik-freiburg.de>
Reply-To: Stefan Schlager <stefan.schla...@uniklinik-freiburg.de>
Subject: Morpho 0.24 available - includes important bug-fix
To: morphmet@morphometrics.org

Dear all,

I just pushed an updated version to CRAN. Please update ASAP, as it contains bug-fixes!
You can either download the binary from https://sourceforge.net/projects/morpho-rpackage/
or once CRAN has published the binaries, simply issue the command "update.packages()"

Changes:

* bugfix: fixed inaccurate scaling in procSym, wit scale=TRUE (affects specimen far from the average) - caused by changes in a function called from another package
-> Thanks go to Paolo Psiras for finding the bug by using synthetic data with large shape differences.

* new: pls2B - two block PLS calculation.
* new: covDist and covPCA to calculate distances between covariance matrices and calculate PC-scores in the space of covariance matrices (as suggested in Mitteroecker P, Bookstein F. 2009. The ontogenetic trajectory of the phenotypic covariance matrix, with examples from craniofacial shape in rats and humans. Evolution 63:727-737.)
* new: facenormals - calculate facenormals and face barycenters of a triangular mesh.

* change: added option "distfun" in procSym to select between approx. Riemann distance (angle) and more accurate Riemann distance to calculate distances to meanshape (differences are negligible when shape differences are small).
* change: added option "scale" in regdist.raw


Best regards
Stefan

--
Stefan Schlager M.A.
Anthropologie
Medizinische Fakultät der der Albert Ludwigs- Universität Freiburg
Hebelstr. 29
79104 Freiburg

Anthropology
Faculty of Medicine, Albert-Ludwigs-University Freiburg
Hebelstr. 29
D- 79104 Freiburg

phone +49 (0)761 203-5522
fax +49 (0)761 203-6898



----- End forwarded message -----



Attachment: signature.asc
Description: PGP signature

Reply via email to