1.1 --- a/grant.html Tue Apr 21 03:36:06 2009 -0700 1.2 +++ b/grant.html Tue Apr 21 04:05:54 2009 -0700 1.3 @@ -592,16 +592,14 @@ 1.4 _____________________________ 1.5 195-fold cross-validation. 1.6 Research Design and Methods 1.7 -Further work on flatmapping 1.8 -Often the surface of a structure serves as a natural 2-D basis for anatomical organization. Even when the shape of the 1.9 -surface is known, there are multiple ways to map it into a plane. We will compare mappings which attempt to preserve 1.10 -size (such as the one used by Caret[7]) with mappings which preserve angle (conformal maps). Although there is much 2-D 1.11 -organization in anatomy, there are also structures whose anatomy is fundamentally 3-dimensional. We plan to include a 1.12 -statistical test that warns the user if the assumption of 2-D structure seems to be wrong. 1.13 -Automatic segmentation of cortical layers 1.14 -Extension to probabalistic maps Presently, we do not have a probabalistic atlas which is registered to the ABA 1.15 -space. However, in anticipation of the availability of such maps, we would like to explore extensions to our Aim 1 techniques 1.16 -which can handle probabalistic maps. 1.17 +Flatmapping and segmentation of cortical layers** 1.18 +There are multiple ways to flatten 3-D data into 2-D. We will compare mappings from manifolds to planes which attempt 1.19 +to preserve size (such as the one used by Caret[7]) with mappings which preserve angle (conformal maps). Our method will 1.20 +include a statistical test that warns the user if the assumption of 2-D structure seems to be wrong. 1.21 +We have not yet made use of radial profiles. While the radial profiles may be used “raw”, for laminar structures like the 1.22 +cortex another strategy is to group together voxels in the same cortical layer; each surface pixel would then be associated 1.23 +with one expression level per gene per layer. We will develop a segmentation algorithm to automatically identify the layer 1.24 +boundaries. 1.25 Develop algorithms that find genetic markers for anatomical regions 1.26 1.Develop scoring measures for evaluating how good individual genes are at marking areas: we will compare pointwise, 1.27 geometric, and information-theoretic measures. 1.28 @@ -620,33 +618,30 @@ 1.29 # Linear discriminant analysis 1.30 Decision trees todo 1.31 20. 1.32 -Apply these algorithms to the cortex 1.33 -1.Create open source format conversion tools: we will create tools to bulk download the ABA dataset and to convert 1.34 -between SEV, NIFTI and MATLAB formats. 1.35 -2.Flatmap the ABA cortex data: map the ABA data onto a plane and draw the cortical area boundaries onto it. 1.36 -3.Find layer boundaries: cluster similar voxels together in order to automatically find the cortical layer boundaries. 1.37 -4.Run the procedures that we developed on the cortex: we will present, for each area, a short list of markers to identify 1.38 -that area; and we will also present lists of “panels” of genes that can be used to delineate many areas at once. 1.39 +# confirm with EMAGE, GeneAtlas, GENSAT, etc, to fight overfitting, two hemis 1.40 +# mixture models, etc 1.41 Develop algorithms to suggest a division of a structure into anatomical parts 1.42 -# mixture models, etc 1.43 1.Explore dimensionality reduction algorithms applied to pixels: including TODO 1.44 2.Explore dimensionality reduction algorithms applied to genes: including TODO 1.45 3.Explore clustering algorithms applied to pixels: including TODO 1.46 -_________________________________________ 1.47 - 20Already, for each cortical area, we have used the C4.5 algorithm to find a decision tree for that area. We achieved good classification accuracy 1.48 -on our training set, but the number of genes that appeared in each tree was too large. We plan to implement a pruning procedure to generate 1.49 -trees that use fewer genes 1.50 4.Explore clustering algorithms applied to genes: including gene shaving, TODO 1.51 5.Develop an algorithm to use dimensionality reduction and/or hierarchial clustering to create anatomical maps 1.52 6.Run this algorithm on the cortex: present a hierarchial, genoarchitectonic map of the cortex 1.53 # Linear discriminant analysis 1.54 # jbt, coclustering 1.55 # self-organizing map 1.56 -# confirm with EMAGE, GeneAtlas, GENSAT, etc, to fight overfitting, two hemis 1.57 # compare using clustering scores 1.58 # multivariate gradient similarity 1.59 # deep belief nets 1.60 -# note: slice artifact 1.61 +Apply these algorithms to the cortex Using the methods developed in Aim 1, we will present, for each cortical area, 1.62 +a short list of markers to identify that area; and we will also present lists of “panels” of genes that can be used to delineate 1.63 +_________________________________________ 1.64 + 20Already, for each cortical area, we have used the C4.5 algorithm to find a decision tree for that area. We achieved good classification accuracy 1.65 +on our training set, but the number of genes that appeared in each tree was too large. We plan to implement a pruning procedure to generate 1.66 +trees that use fewer genes 1.67 +many areas at once. Using the methods developed in Aim 2, we will present one or more hierarchial cortical maps. We will 1.68 +identifyand explain how the statistical structure in the gene expression data led to any unexpected or interesting features 1.69 +of thesemaps. 1.70 Bibliography & References Cited 1.71 [1]Chris Adamson, Leigh Johnston, Terrie Inder, Sandra Rees, Iven Mareels, and Gary Egan. A Tracking Approach to 1.72 Parcellation of the Cerebral Cortex, volume Volume 3749/2005 of Lecture Notes in Computer Science, pages 294–301.