Re: [MORPHMET] Re: semilandmarks in biology
Agreed. In addition, I think it’s important to note that, in the original implementations of the sliding algorithm, semilandmarks were slid not along the curve itself, but along tangents to the curve (= off the boundary outline). How much distortion this induces is, of course, a function of how much the semilandmarks are displaced from their original positions. However, it’s always seemed problematic to me that, after sliding, you end up with shapes that have been distorted to a greater or lesser extent. Of course, if the displacement is small the amount off distortion will (likely) be small and the results might not be all that different. Moreover, as Phillip notes, in terms of many types of analyses, linear data transformations make no difference to the outcome of an analysis. But given these facts, the point of sliding the semilandmarks at all seems questionable in many contexts. Moreover, in the case of complex boundary outline curves - in other words, the curves semilandmarks are usually called upon to quantify - since the magnitude of the slide is, to a large extent, determined by the density of the semilandmark placements, large semilandmark displacements will never occur. So, if you have a curve that is so smooth it only needs a few semilandmarks to tie down, you run the risk of generating some (presently unspecified) degree of distortion in your data by sliding the semilandmarks so long as the sliding takes place along tangents. But if your curve is complex it’s unlikely that sliding the semilandmarks will make much difference because the distance along which sliding can take place is constrained. Sliding semilandmarks is an interesting strategy in principle. But in many cases the (current) practice is fraught with problems that are rarely acknowledged. Norm MacLeod > On 6 Nov 2018, at 19:53, mitte...@univie.ac.at wrote: > > Yes, it was always well known that sliding adds covariance but this is > irrelevant for most studies, especially for group mean comparisons and shape > regressions: the kind of studies for which GMM is most efficient, as Jim > noted. > If you consider the change of variance-covariance structure due to (a small > amount of) sliding as an approximately linear transformation, then the > sliding is also largely irrelevant for CVA, relative PCA, Mahalanobis > distance and the resulting group classifications, as they are all based on > the relative eigenvalues of two covariance matrices and thus unaffected by > linear transformations. In other words, in the lack of a reasonable > biological null model, the interpretation of a single covariance structure is > very difficult, but the way in which one covariance structure deviates from > another can be interpreted much easier. > > Concerning your example: The point is that there is no useful model of > "totally random data" (but see Bookstein 2015 Evol Biol). Complete > statistical independence of shape coordinates is geometrically impossible and > biologically absurd. Under which biological (null) model can two parts of a > body, especially two traits on a single skeletal element such as the cranium, > be complete uncorrelated? > > Clearly, semilandmarks are not always necessary, but making "cool pictures" > can be quite important in its own right for making good biology, especially > in exploratory settings. Isn't the visualization one of the primary strengths > of geometric morphometrics? > > It is perhaps also worth noting that one can avoid a good deal of the > additional covariance resulting from sliding. Sliding via minimizing bending > energy introduces covariance in the position of the semilandmarks _along_ the > curve/surface. In some of his analyses, Fred Bookstein just included the > coordinate perpendicular to the curve/surface for the semilandmarks, thus > discarding a large part of the covariance. Note also that sliding via > minimizing Procrustes distance introduces only little covariance among > semilandmarks because Procrustes distance is minimized independently for each > semilandmark (but the homology function implied here is biologically not so > appealing). > > Best, > > Philipp > > > > Am Dienstag, 6. November 2018 18:34:51 UTC+1 schrieb alcardini: > Yes, but doesn't that also add more covariance that wasn't there in > the first place? > Neither least squares nor minimum bending energy, that we minimize for > sliding, are biological models: they will reduce variance but will do > it in ways that are totally biologically arbitrary. > > In the examples I showed sliding led to the appearance of patterns > from totally random data and that effect was much stronger than > without sliding. > I neither advocate sliding or not sliding. Semilandmarks are different > from landmarks and more is not necessarily better. There are > definitely some applications where I find them very useful but many > more where they seem to be there just to mak
Re: [MORPHMET] Interpreting PCA results
I agree with Jim. However, this discussion does beg the question of what the status of landmark, semilandmark, or indeed pixel brightness configurations within multivariate spaces is? Very similar spaces have been used in the area of theoretical morphology to conduct various sorts of experiments dealing with the nature of morphological evolution, especially the development of patterns based on null models to which empirical patterns can be compared. Moreover, machine learning specialists are now using morphologies generated artificially, in ways that aren’t very different from the ways in which such visualisations can be created by morphometricians, to train their AI systems. McGhee distinguishes "theoretical morphospaces” derived from graphics equations (e.g., Raup’s coiling models) from (what he terms) the “empirical morphospaces” we deal with as morphometricians and that lie at the heart of this conversation. But are the two really that different? If so, why and in what cases? If not what does that mean for the ways in which we might use such spaces? I’ve long found this an interesting question to ponder. Any thoughts from the community? Norm MacLeod > On 15 May 2017, at 18:28, F. James Rohlf wrote: > > What is important is not the fact that one is going +/- one standard > deviation along each axis. When shape changes are subtle one may need to go > beyond the observed range to make it more obvious to the eye what the changes > are. Exactly how far one goes away from the mean is arbitrary. It is a > visualization – not statistics. > > -- > F. James Rohlf New email: f.james.ro...@stonybrook.edu > Distinguished Professor, Emeritus. Dept. of Ecol. & Evol. > & Research Professor. Dept. of Anthropology > Stony Brook University 11794-4364 > WWW: http://life.bio.sunysb.edu/morph/rohlf > P Please consider the environment before printing this email > _ Professor Norman MacLeod The Natural History Museum, Cromwell Road, London, SW7 5BD (0)207 942-5204 (Office Landline) (0)785 017-1787 (Mobile) http://paleonet.org/MacLeod/ Department of Earth Sciences, University College London, Gower Street, London WC1E 6BT, UK Nanjing Institute of Geology & Palaeontology, Chinese Academy of Sciences, 39 Beijing, Donglu, Nanjing, China _ -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org --- You received this message because you are subscribed to the Google Groups "MORPHMET" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet+unsubscr...@morphometrics.org.
Re: [MORPHMET] model II regression statistics PAST
I can’t help you with PAST, but this reference will allow you to compute the correct statistics for a standardized major axis (often erroneously called a reduced major axis) regression. Warton, D. I., et al., 2006, Bivariate line-fitting methods for allometry: Biological Reviews, v. 81, no. 2, p. 259–291. Norm MacLeod > On 12 Mar 2016, at 14:49, Patrick Arnold wrote: > > Dear morphometrics, > > I am frequently using PAST (paleontological statistics; what a great free > software) for the analysis of morphometric data (geometric and traditional). > Now I want to do a reduced major axis regression on bivariate data (only one > Y for each X). However, I do not found the possibility to perform a > significance test for the regression (F statistics) or for the slope and > intercept (t statistics). > It is provided for least square regression (e.g., under polynomial option) > but not for model II regression. Does anyone know a possibility to obtain > regression statistics in PAST for model II regressions? > Thanks in advance. > > Cheers > Patrick > > > -- > Patrick Arnold, M.Sc. > > wissenschaftlicher Mitarbeiter und Doktorand > Institut für Spezielle Zoologie und Evolutionsbiologie > mit Phyletischem Museum > Friedrich-Schiller-Universität Jena > Erbertstraße 1 > 07743 Jena > Germany > > Phone: +49 (0)3641 9-49165 > Fax:+49 (0)3641 9-49142 > E-mail: patrick.arn...@uni-jena.de > > -- > MORPHMET may be accessed via its webpage at http://www.morphometrics.org > ---You received this message because you are subscribed to the Google Groups > "MORPHMET" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to morphmet+unsubscr...@morphometrics.org. > _ Professor Norman MacLeod Dean of Postgraduate Education and Training The Natural History Museum, Cromwell Road, London, SW7 5BD (0)207 942-5204 (Office Landline) (0)785 017-1787 (Mobile) http://www.nhm.ac.uk/hosted_sites/paleonet/MacLeod/ Department of Earth Sciences, University College London, Gower Street, London WC1E 6BT, UK Nanjing Institute of Geology & Palaeontology, Chinese Academy of Sciences, 39 Beijing, Donglu, Nanjing, China _ -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org --- You received this message because you are subscribed to the Google Groups "MORPHMET" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet+unsubscr...@morphometrics.org.
Re: Trusses
If you are interested and I have an old application for performing Burnaby-PCA I'd be happy to send you. Norm MacLeod ___ Prof. Norman MacLeod Keeper of Palaeontology The Natural History Museum Cromwell Road, London, SW7 5BD (0)207 942-5204 (Office) (0)207 942-5546 (Fax) http://www.nhm.ac.uk/palaeontology/a&ss/nm/nm.html (Web Page) ___ On 10/6/05 15:57, "morphmet" <[EMAIL PROTECTED]> wrote: > I want to perform comparative anlyses of shape variation using trusses > and newer geometric analyses (including outlines). I have already > carried out a TPS analysis and now wish to proceed with an anlysis based > on truss measurements. I would like to measure truss lengths between > pairs of landmarks, can I do this using the formula (i.e. for the truss > measurment between L1 and L6): > > ã (y1-y6)2+ (x1-x6)2 * scale factor (scale set for images in TPSDig2) > > I need to correct size and think I would like to use the Burnaby method > (1966) but am unsure how to proceed. > > Any advice would be gratefully received. > > Thanks, > > Catherine > > [EMAIL PROTECTED] > > > > > __ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com
Algorithmic Approaches to the Identification Problem in Systematics Symposium
Title: Algorithmic Approaches to the Identification Problem in Systematics Symposium The second circular for the upcoming Algorithmic Approaches to the Identification Problem in Systematics symposium has just been released (see below). This symposium will be held on 19 August 2005 in the Flett Theater of the Natural History Museum, London. It’s purpose is to provide leaders of research groups, researchers, post-doctoral research assistants, and students working or studying in any area of systematics with an opportunity to (1) learn about current trends in quantitative approaches to the group-recognition problem, (2) become familiar with the capabilities of various software systems currently available for identifying systematic objects/groups and (3) evaluate various applications of this technology to present and future systematic problems. Special attention will be paid to showing how different approaches to automated identification can be applied to various organismal groups and in various applied research contexts (e.g., biodiversity studies, biostratigraphy, conservation, agriculture). Ample programme time will also be provided for discussions of issues relating to how these approaches and technologies can play a larger role in meeting the needs of current and future systematists. This free, one-day symposium is sponsored by The Systematics Association and the Natural History Museum London, and is part of The Systematics Association’s Biennial Meeting. Please visit the web site for additional information (http://www.nhm.ac.uk/hosted_sites/paleonet/aaips_symposium/). If you are not able to attend the meeting, a symposium volume is being assembled and will be published as part of the Systematics Associations Special Volume series in 2006. Below is a list of symposium presentations. If you have any questions about the symposium please see the symposium web site or contact me at [EMAIL PROTECTED] or at the address below. Norman MacLeod ___ Dr. Norman MacLeod Keeper of Palaeontology The Natural History Museum Cromwell Road, London, SW7 5BD (0)207 942-5204 (Office) (0)207 942-5546 (Fax) http://www.nhm.ac.uk/palaeontology/a&ss/nm/nm.html (Web Page) ___ Second Circular Algorithmic Approaches to the Identification Problem in Systematics Date: 19 August 2005 Venue: Flett Theatre, The Natural History Museum, Cromwell Road, London Sponsors: The Systematics Association and The Natural History Museum, London Authors and Presentation Titles (alphabetical listing by author) Homology and Morphometrics: An Old Theme Revisited F. L. Bookstein Institute of Geontology, University of Michigan, Ann Arbor, Michigan, USA and Institute of Anthropology, University of Vienna, Austria. Is Automated Species Identification Feasible? David Chesmore Intelligent Systems Research Group, Department of Electronics, University of York, Heslington, York, YO10 5DD, England. Identification of Botanical Taxa Using Artificial Neural Networks Jonathan Y. Clark Neural Computing Group, Department of Computing, School of Electronics and Physical Sciences, University of Surrey, Guildford, Surrey, GU2 7XH, UK. Natural Object Recognition – Machines Versus Humans Phil Culverhouse Centre for Interactive Intelligent Systems, School of Computing, Communications & Electronics, University of Plymouth PL4 8AA, UK. Drawing the Line: the Differentiation Between Morphological Plasticity and Interspecific Variation David Jones and Mark Purnell Department of Geology, University of Leicester, University Road, Leicester, England LE1 7RH, UK. Plastic Self Organising Maps Robert Lang Flat 3, 15 Christchurch Gardens, Reading, Berks RG2 7AH, UK. Forging a Synthesis Between 3D Object Ordination and 3D Object Recognition Norman MacLeod1, P. David Polly2, Stig Walsh1, Mark O’Neill3 1Department of Palaeontology, Natural History Museum, Cromwell Road, London SW7 5BD, UK; 2Department of Biological Sciences, Queen Mary, University of London, Mile End Road, London E1 4NS,UK; 3Centre for Neuroecology, Henry Wellcome Building, University of Newcastle upon Tyne, Newcastle, UK. Decision Trees: A Machine Learning Methodology to Determine Ungulate Feeding Behavior from Craniodental Morphology Manuel Mendoza Department of Ecology and Evolutionary Biology, Brown University, Providence, Rhode Island 02912, USA. Pattern Recognition for Ecological Science and Environmental Monitoring Eric N. Mortensen School of Electrical Engineering and Computer Science, Oregon State University, Corvallis, Oregon 97331-4501, USA. DAISY: A Practical Computer Based Tool for Semi-Automated Species Identification Mark A. O'Neill Centre for Neuroecology, Henry Wellcome Building, University of Newcastle upon Tyne, Newcastle, UK. Introducing SPIDA-web: An Automated Identification System for Biological Species Kimberly Norris Russell, Martin T. D
