<<   Research

Last update: Thursday, 13-Mar-2014 12:56:32 EDT

 
 

The following software is distributed "as is", and it is free for anyone to use. Even though I have made every effort to ensure that computations are performed correctly, it is always a good idea to check for mathematical errors if you intend to publish results produced with this software. Also, make sure you are using the latest release (compiled version) of each program, and particularly young programs, as previous versions may contain problems, which may or may not produce error messages. You can help improve these programs by sending comments and reporting bugs by clicking here. You can also contact me for support about using the software, or if you need specific information about algorithms, formulae, etc.

  
 
New for March 11th, 2014: ALL NON-BETA PROGRAMS UPDATED!
   In response to many users' reports of software on this page crashing even before opening, I decided to update all of the stable programs using a new set of Matlab and Microsoft libraries (that is, SAGE, MACE, MACE3D, SEMITHINNER, CORIANDIS, MINT, LORY, and CPR). The problem, it seems, was introduced during a Windows update at some point last year, and affected, as far as I know, only installations running on Windows 7+ and Windows emulators for Mac. I have not heard any reports of the programs crashing in Windows Vista or XP. In any case, the programs, with one exception, are the same. Keep in mind that to run these versions, you will also need to install a new set of Matlab MCR libraries (version 8.1). I've updated the links below to point to the correct files (still password-gated, read further for instructions). Alternatively, Matlab now offers these libraries on their site. Just make sure you get the correct (32-/64-bit version 8.1) MCRinstaller.
   If you are still experiencing problems, please drop me a line letting me know. For the time being, I am keeping the old versions of these programs here (MCR libraries ver. 7.11: 32-bit/64-bit).


PLEASE READ BEFORE DOWNLOADING

These programs have been developed and compiled using Matlab® software and will only work on Windows machines. In order to run, the necessary Matlab libraries must be instaled prior to their use. These libraries are specific for Matlab version 7.9.0 (R2009b), and the programs will not work on different versions. The get the libraries, you must run and install the file MCRInstaller.exe, linked to below, if you have not done so before. To use any of the programs listed here, simply run the .exe file included with each program zip-file. You only need to download and install MCRInstaller once.

Due to licensing requirements, download of Matlab Component Runtime libraries is now password restricted. If you are a user of the software listed in this page, please contact me using this form to request username and password. Alternatively, you can download these libraries directly from Matlab website.

MCRInstaller.exe (ver.8.1/32-bit Windows) || MCRInstaller.exe (ver.8.1/64-bit Windows)



Program/version Purpose and features Links
Sage* 32-bit|64-bit
Symmetry and Asymmetry in Geometric Data, version 1.21
(last compiled 03/11/14)
  • Sage accepts 2D landmark data (in X1,Y1,...,Xk,Yk; X1,Y1,...,Xk,Yk,CS; or basic TPS formats) and decomposes variance into symmetric and asymmetric components. The latter is further decomposed in directional and fluctuating asymmetry.
  • Procrustes ANOVA and MANOVA tests are performed to assess the significance of symmetry (=individual), directional asymmetry, and fluctuating asymmetry of shape and size (when applicable), given samples with at least two replicates per specimen.
  • Both Object (bilaterally symmetric structures) and Matching (bilaterally symmetric parts) types of symmetry are handled.
  • Covariance matrix correlations are also computed between symmetric and asymmetric components of variation.
  • Sage allows saving symmetrized datasets (average of relabeled, reflected, and superimposed configurations), residuals from symmetric component, plus shape configurations representing each component of variation (symmetric, asymmetric, and error), plus expected covariance matrices from each of these effects.
  • A preliminary 3-D version of Sage is available upon request only. Although it does most computations (ANOVA, MANOVA, permutation tests), it is not yet capable to perform Procrustes superimposition (you need to superimpose your data yourself, using, for example, David Sheet's IMP software). Current version of Sage3D cannot yet produce any visualization either.
  • Manual (PDF)
  • Screenshot
  • References
  • Version history
    •
    Mace* 32-bit|64-bit
    Matrix correlations for landmark data, version 1.03
    (last compiled 03/11/14)
    • Mace accepts three forms of data: covariance matrices obtained from landmark data, actual landmark data (same formats as Sage), paired non-landmark variables, such as partial warp scores, and unpaired (regular) data, such as linear distances.
    • Significance of matrix correlations is computed from permuting rows and columns of covariance matrices, maintaining the pairing of X,Y coordinates. Inclusion of covariance diagonals in permutation tests is optional.
    • Significance tests include computation of repeatability, and bootstrap estimates of correlation statistics.
    • Covariance matrices can be transformed into an "isotropic" version, in which all elements except for the diagonal are zero and all elements in diagonals are equal, and a version in which variances are left as observed and all remaining elements are set to zero. The resulting matrices can be saved and reintroduced in the analysis.
  • Manual (PDF)
  • Screenshot
  • References
  • Version history
    •
    Mace3D* 32-bit|64-bit
    Matrix correlations for 3-D landmark data, version 1.02
    (last compiled 03/11/14)
    • Mace3D has the same functionality as Mace, except that does not process unpaired (traditional) data.
  • Manual (PDF)
  • References
    •
    SemiThinner* 32-bit|64-bit
    Utility to convert curves in semi-landmarks, version 1.02
    (last compiled 03/11/14)
    • SemiThinner does only one thing: accepts (2-D) TPS files with CURVES statements and reduce the number of points to a specified number in all specimens in the TPS file.
    • SemiThinner allows saving only semi-landmarks, landmarks only, or both landmarks and semi-landmarks in a XY... format.
    • You can contact me if you need this program to accept formats other than TPS, or 3-D data. Send me an annotated sample of your format to add it to the program.
  • Manual (PDF)
  • Screenshot
    •
    Coriandis* 32-bit|64-bit
    Correlation analysis based on distances , version 1.12
    (last compiled 03/11/14)
    • Coriandis provides a set of graphical and analytical tools to study associations among multivariate datasets (e.g. shapes), using distances among measured individual or species.
    • Coriandis accepts 2D landmark, as well as non-landmark data, including distance matrices (irrespective of how they are computed).
    • Referenced in Márquez & Knowles (2007) [PDF][abstract].
  • Manual (PDF)
  • Screenshot
  • References
  • Version history
    •
    Mint* 32-bit|64-bit
    Modularity and Integration analysis tool for morphometric data, version 1.61
    (last compiled 03/11/14)
    • Mint is a tool for testing a priori models of morphological integration and modularity on multivariate data.
    • Mint accepts general multivariate data, as a matrix with n individuals and m variables, as well as 2-D landmark data (XY and TPS formats).
    • Models tested by Mint assign each variable or landmark to a module, and tests are carried out on the covariance matrix derived from a full set of models.
    • Models can be loaded individually, in batch, or created and edited within Mint. All loaded/edited/created models are then tested simultaneously for relative of goodness of fit.
    • Fit statistics are derived using a parametric approach and a resampling (jackknife) method.
    • Mint also includes a tool to carry out a variant of Partial Least Squares analysis on which a partition or putative module is regressed against the full set of traits.
    • Referenced in Márquez (2008) [PDF][abstract].
  • Manual (PDF)
  • Screenshot
  • References
  • Version history
    •
    Lory**,*** 32-bit|64-bit
    Model-based Estimation of Local Shape Deformations, version 1.0
    (last compiled 03/11/14)
    • Lory is a tool for estimation and visualization of local deformations derived from landmark data.
    • Lory accepts 2-D landmark data (XY and TPS formats).
    • Lory uses shape information contained in all landmarks to estimate, via interpolation, local deformations.
    • Interpolations are based (for now) on two popular models: thin-plate splines (TPS) and elastic body splines (EBS).
    • In this approach, landmarks are not treated as data, but as parameters of functions that describe shape differences continuously distributed over entire configurations.
    • Evaluation of functions at pseudo-homologous points provide estimates of local deformations that are directly interpretable as local shape differences, amenable for statistical analyses.
    • Estimates provided as either Jacobian matrices or their (scalar) determinants.
    • Lory also provides a variety of useful visualizations (deformation grids, parrot plots, deformation movies), which can be exported at publication-quality resolution level.
    • Referenced in Márquez et al. (2012) [PDF][abstract].
  • Manual (PDF)
  • Screenshot
  • References
  • Version history
    •


    Training grounds: Software currently in active development

    The following software is in beta-testing phase and as such it may still contain bugs relevant for proper program function. Final versions of each program as well as full support and Windows compatibility will be available soon, along with the relevant publications. They are posted here for evaluation purposes, mainly intended for direct collaborators. General users can still use it at their own risk; any comments or bug reports are appreciated (click here to submit).

    CPR** (External link) Tool for reading, conversion, and visualization of CP files from WingMachine, version 1.11 (last compiled 03/11/14)

    Sift** Tool for filtering genomewide scan data based on frequencies of resolved and unresolved sites, version 0.11 (last compiled 03/31/11)

    Winnow** Tool for mapping of multivariate phenotypic data on genomewide scans, version 0.2 (last compiled 03/31/11)

    flyWISH** Utility for comparing and sorting shape variation with respect to a target shape, version 0.1 (last compiled 03/31/11)

    *Supported by the National Science Foundation Doctoral Dissertation Improvement Grant 0407570.
    **Supported by the National Science Foundation Grant DEB 0950002.
    ***Supported by the National Institutes of Health through the NIH Roadmap for Medical Research, Grant U54 RR021813.

    Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation or the National Institutes of Health.

    References:

    • Abdi, H.; Valentin, D.; O’Toole, A.J.; Edelman, B. 2005. DISTATIS: The analysis of multiple distance matrices. In: Proceedings of the IEEE Computer Society: International Conference on Computer Vision and Pattern Recognition, San Diego, pp. 42–47.
    • Abdi, H.; Valentin, D.; Chollet, S.; Chrea, C. 2007. Analyzing assessors and products in sorting tasks: DISTATIS, theory and applications. Food Quality and Preference, 18:627-640.
    • Escoufier, Y. 1973. Le traitement des variables vectorielles. Biometrics 29: 751–760.
    • Klingenberg, C.P.; Barluenga, M.; Meyer, A. 2002. Shape analysis of symmetric structures: quantifying variation among individuals and asymmetry. Evolution 56:1909-1920.
    • Klingenberg, C.P.; McIntyre, G.S. 1998. Geometric morphometrics of developmental instability: analyzing patterns of fluctuating asymmetry with Procrustes methods. Evolution 52:1363-1375.
    • Krzanowski, W.J. 2000. Principles of Multivariate Analysis: A User’s Perspective. Oxford University Press, Oxford.
    • Mardia, K.V.; Bookstein, F.L.; Moreton, I.J. 2000. Statistical assessment of bilateral symmetry of shapes. Biometrika 87:285-300.
    • Marroig, G.; Cheverud, J.M. 2001. A comparison of phenotypic variation and covariation patterns and the role of phylogeny. Ecology, and ontogeny during cranial evolution of new world monkeys. Evolution 55:2576-2600.
    • Márquez, E. J.; Knowles L.L. 2007. Correlated evolution of multivariate traits: detecting co-divergence across multiple dimensions. Journal of Evolutionary Biology 20:2334-2348.
    • Márquez, E. J. 2008. A statistical framework for testing modularity in multidimensional data. Evolution 62:2688-2708.
    • Márquez, E. J.; Cabeen, R.; Woods, R. P.; Houle, D. 2012. The measurement of local variation in shape. Evolutionary Biology 39:419-439.
    • Palmer, A.R.; Strobeck, C. 1986. Fluctuating asymmetry: measurement, analysis, patterns. Annual Review of Ecology and Systematics 17:391-421.
    • Parsons, K. J.; Márquez, E. J.; Albertson, R. C. 2012. Constraint and opportunity: the genetic basis and evolution of modularity in the cichlid mandible. The American Naturalist 179:64-78.
    © 2003-2014 Eladio J. Márquez
    Visitors since 05/30/04: