-- computational reconstruction and transformation of flattened retinae
Retistruct is an R package to morph a flat surface with cuts (a dissected flat-mount retina) onto a curvilinear surface (the a standard retinal shape). It can estimate the position of a point on the intact adult retina to within 8° of arc (3.6% of nasotemporal axis). The coordinates in reconstructed retinae can be transformed to visuotopic coordinates.
Reconstruction is achieved by: stitching the marked-up cuts of the flat-mount outline; dividing the stitched outline into a mesh whose vertices then are mapped onto a curtailed sphere; and finally moving the vertices so as to minimise a physically-inspired deformation energy function.
Retistruct has been tested on GNU/Linux (Ubuntu 12.04), Mac OS X 10.8 and Microsoft Windows Vista. Installing the graphical user interface on Mac OS X 10.9 (Mavericks) and 10.10 (Yosemite) is possible, but requires the GTK library to be installed first; see Chris von Csefalvay's instructions.
To install the stable version of Retistruct hosted on CRAN, follow the instructions in the User Guide. The installation contains a number of demonstration retinae, and instructions for how to handle retinal flat-mount images in Retistruct.
The development version of Retistruct contains the most recent bug fixes and improvements, but is not stable. Builds of the package are usually available on R-forge and can be installed using the R install command given there. If the R-forge packages are not available, you will have to build from the source code using R CMD build.
The source code can be checked out from the R-forge subversion repository or Github.
For reference purposes, this zip file contains the version of Retistruct that generated the reconstructions in Sterratt & al. (2013; PLoS Computational Biology 9). The file also contains some Matlab code to read data directories contained by Retistruct.
As well as the built-in demo data, there are some sample images to practise on:
Retistruct was written by David Sterratt at the University of Edinburgh, and tested by Daniel Lyngholm and Ian Thompson at the MRC Centre for Developmental Neurobiology, KCL.
This work was supported by a Programme Grant from the Wellcome Trust (G083305).