1.1 --- a/grant.html Fri Jul 03 05:17:28 2009 -0700 1.2 +++ b/grant.html Wed Jul 08 05:18:30 2009 -0700 1.3 @@ -44,25 +44,25 @@ 1.4 Even the questions of how many areas should be recognized in cortex, and what their arrange- 1.5 ment is, are still not completely settled. A proposed division of the cortex into areas is called a 1.6 cortical map. In the rodent, the lack of a single agreed-upon map can be seen by contrasting the 1.7 -recent maps given by Swanson[21] on the one hand, and Paxinos and Franklin[16] on the other. 1.8 +recent maps given by Swanson[22] on the one hand, and Paxinos and Franklin[17] on the other. 1.9 While the maps are certainly very similar in their general arrangement, significant differences re- 1.10 main. 1.11 The Allen Mouse Brain Atlas dataset 1.12 - The Allen Mouse Brain Atlas (ABA) data[13] were produced by doing in-situ hybridization on 1.13 + The Allen Mouse Brain Atlas (ABA) data[14] were produced by doing in-situ hybridization on 1.14 slices of male, 56-day-old C57BL/6J mouse brains. Pictures were taken of the processed slice, 1.15 and these pictures were semi-automatically analyzed to create a digital measurement of gene 1.16 expression levels at each location in each slice. Per slice, cellular spatial resolution is achieved. 1.17 Using this method, a single physical slice can only be used to measure one single gene; many 1.18 different mouse brains were needed in order to measure the expression of many genes. 1.19 - Mus musculus is thought to contain about 22,000 protein-coding genes[26]. The ABA contains 1.20 + Mus musculus is thought to contain about 22,000 protein-coding genes[27]. The ABA contains 1.21 data on about 20,000 genes in sagittal sections, out of which over 4,000 genes are also measured 1.22 in coronal sections. Our dataset is derived from only the coronal subset of the ABA2. An auto- 1.23 mated nonlinear alignment procedure located the 2D data from the various slices in a single 3D 1.24 coordinate system. In the final 3D coordinate system, voxels are cubes with 200 microns on a 1.25 -side. There are 67x41x58 = 159,326 voxels, of which 51,533 are in the brain[15]. For each voxel 1.26 -and each gene, the expression energy[13] within that voxel is made available. 1.27 - The ABA is not the only large public spatial gene expression dataset[8][25][5][14][24][4][23][20][3]. 1.28 -However, with the exception of the ABA, GenePaint[25], and EMAGE[24], most of the other re- 1.29 +side. There are 67x41x58 = 159,326 voxels, of which 51,533 are in the brain[16]. For each voxel 1.30 +and each gene, the expression energy[14] within that voxel is made available. 1.31 + The ABA is not the only large public spatial gene expression dataset[9][26][6][15][25][4][24][21][3]. 1.32 +However, with the exception of the ABA, GenePaint[26], and EMAGE[25], most of the other re- 1.33 sources have not (yet) extracted the expression intensity from the ISH images and registered the 1.34 results into a single 3-D space. 1.35 The remainder of the background section will be divided into three parts, one for each major 1.36 @@ -79,9 +79,9 @@ 1.37 regions may be expressed as a function. The input to this function is a voxel, along with the gene 1.38 expression levels within that voxel; the output is the regional identity of the target voxel, that is, the 1.39 ____________________________________ 1.40 - 2The sagittal data do not cover the entire cortex, and also have greater registration error[15]. Genes were selected 1.41 + 2The sagittal data do not cover the entire cortex, and also have greater registration error[16]. Genes were selected 1.42 by the Allen Institute for coronal sectioning based on, “classes of known neuroscientific interest... or through post hoc 1.43 -identification of a marked non-ubiquitous expression pattern”[15]. 1.44 +identification of a marked non-ubiquitous expression pattern”[16]. 1.45 2 1.46 1.47 region to which the target voxel belongs. We call this function a classifier. In general, the input to 1.48 @@ -255,7 +255,7 @@ 1.49 is selectively underex- 1.50 pressed in area SS. As noted above, the GIS community has developed tools for supervised 1.51 classification and unsupervised clustering in the context of the analysis 1.52 - of hyperspectral imaging data. One tool is Spectral Python6. Spectral 1.53 + of hyperspectral imaging data. One tool is Spectral Python[5]. Spectral 1.54 Python implements various supervised and unsupervised classification 1.55 methods, as well as utility functions for loading, viewing, and saving 1.56 spatial data. Although Spectral Python has feature extraction methods 1.57 @@ -263,36 +263,35 @@ 1.58 new features computed based on the original features, it does not have 1.59 feature selection methods, that is, methods to select a small subset 1.60 out of the original features (although feature selection in hyperspectral 1.61 - imaging has been investigated by others[19]. 1.62 + imaging has been investigated by others[20]. 1.63 There is a substantial body of work on the analysis of gene expression data. Most of this con- 1.64 -cerns gene expression data which are not fundamentally spatial7. Here we review only that work 1.65 +cerns gene expression data which are not fundamentally spatial6. Here we review only that work 1.66 which concerns the automated analysis of spatial gene expression data with respect to anatomy. 1.67 - Relating to Goal 1, GeneAtlas[5] and EMAGE [24] allow the user to construct a search query by 1.68 + Relating to Goal 1, GeneAtlas[6] and EMAGE [25] allow the user to construct a search query by 1.69 demarcating regions and then specifying either the strength of expression or the name of another 1.70 gene or dataset whose expression pattern is to be matched. Neither GeneAtlas nor EMAGE allow 1.71 one to search for combinations of genes that define a region in concert. 1.72 - Relating to Goal 2, EMAGE[24] allows the user to select a dataset from among a large number 1.73 + Relating to Goal 2, EMAGE[25] allows the user to select a dataset from among a large number 1.74 of alternatives, or by running a search query, and then to cluster the genes within that dataset. 1.75 EMAGE clusters via hierarchical complete linkage clustering. 1.76 - [15] describes AGEA, ”Anatomic Gene Expression Atlas”. AGEA has three components. Gene 1.77 + [16] describes AGEA, ”Anatomic Gene Expression Atlas”. AGEA has three components. Gene 1.78 Finder: The user selects a seed voxel and the system (1) chooses a cluster which includes the 1.79 seed voxel, (2) yields a list of genes which are overexpressed in that cluster. Correlation: The user 1.80 selects a seed voxel and the system then shows the user how much correlation there is between 1.81 the gene expression profile of the seed voxel and every other voxel. Clusters: AGEA includes a 1.82 +preset hierarchical clustering of voxels based on a recursive bifurcation algorithm with correlation 1.83 ____________________________________ 1.84 - 6http://spectralpython.sourceforge.net/ 1.85 - 7By “fundamentally spatial” we mean that there is information from a large number of spatial locations indexed by 1.86 + 6By “fundamentally spatial” we mean that there is information from a large number of spatial locations indexed by 1.87 spatial coordinates; not just data which have only a few different locations or which is indexed by anatomical label. 1.88 6 1.89 1.90 -preset hierarchical clustering of voxels based on a recursive bifurcation algorithm with correlation 1.91 as the similarity metric. AGEA has been applied to the cortex. The paper describes interesting 1.92 results on the structure of correlations between voxel gene expression profiles within a handful of 1.93 cortical areas. However, that analysis neither looks for genes marking cortical areas, nor does it 1.94 suggest a cortical map based on gene expression data. Neither of the other components of AGEA 1.95 can be applied to cortical areas; AGEA’s Gene Finder cannot be used to find marker genes for the 1.96 cortical areas; and AGEA’s hierarchical clustering does not produce clusters corresponding to the 1.97 -cortical areas8. 1.98 +cortical areas7. 1.99 1.100 1.101 Figure 3: The top row shows the two 1.102 @@ -303,11 +302,11 @@ 1.103 area AUD, according to gradient sim- 1.104 ilarity. From left to right and top to 1.105 bottom, the genes are Ssr1, Efcbp1, 1.106 -Ptk7, and Aph1a. [6] looks at the mean expression level of genes within 1.107 +Ptk7, and Aph1a. [7] looks at the mean expression level of genes within 1.108 anatomical regions, and applies a Student’s t-test to de- 1.109 termine whether the mean expression level of a gene is 1.110 significantly higher in the target region. This relates to 1.111 - our Goal 1. [6] also clusters genes, relating to our Goal 1.112 + our Goal 1. [7] also clusters genes, relating to our Goal 1.113 2. For each cluster, prototypical spatial expression pat- 1.114 terns were created by averaging the genes in the cluster. 1.115 The prototypes were analyzed manually, without cluster- 1.116 @@ -323,27 +322,27 @@ 1.117 sults). Figures 4, 2, and 3 in the Preliminary Results 1.118 section contain evidence that each of our three choices 1.119 is the right one. 1.120 - [10] describes a technique to find combinations of 1.121 + [11] describes a technique to find combinations of 1.122 marker genes to pick out an anatomical region. They 1.123 use an evolutionary algorithm to evolve logical operators which combine boolean (thresholded) 1.124 images in order to match a target image. They apply their technique for finding combinations of 1.125 marker genes for the purpose of clustering genes around a “seed gene”. 1.126 Relating to our Goal 2, some researchers have attempted to parcellate cortex on the basis of 1.127 -non-gene expression data. For example, [17], [2], [18], and [1] associate spots on the cortex with 1.128 -the radial profile9 of response to some stain ([12] uses MRI), extract features from this profile, and 1.129 +non-gene expression data. For example, [18], [2], [19], and [1] associate spots on the cortex with 1.130 +the radial profile8 of response to some stain ([13] uses MRI), extract features from this profile, and 1.131 then use similarity between surface pixels to cluster. 1.132 - [22] describes an analysis of the anatomy of the hippocampus using the ABA dataset. In 1.133 + [23] describes an analysis of the anatomy of the hippocampus using the ABA dataset. In 1.134 addition to manual analysis, two clustering methods were employed, a modified Non-negative 1.135 Matrix Factorization (NNMF), and a hierarchical bifurcation clustering scheme using correlation as 1.136 +similarity. The paper yielded impressive results, proving the usefulness of computational genomic 1.137 ____________________________________ 1.138 - 8In both cases, the cause is that pairwise correlations between the gene expression of voxels in different areas but 1.139 + 7In both cases, the cause is that pairwise correlations between the gene expression of voxels in different areas but 1.140 the same layer are often stronger than pairwise correlations between the gene expression of voxels in different layers 1.141 but the same area. Therefore, a pairwise voxel correlation clustering algorithm will tend to create clusters representing 1.142 cortical layers, not areas. 1.143 - 9A radial profile is a profile along a line perpendicular to the cortical surface. 1.144 + 8A radial profile is a profile along a line perpendicular to the cortical surface. 1.145 7 1.146 1.147 -similarity. The paper yielded impressive results, proving the usefulness of computational genomic 1.148 anatomy. We have run NNMF on the cortical dataset, and while the results are promising, other 1.149 methods may perform as well or better (see Preliminary Results, Figure 6). 1.150 Comparing previous work with our Goal 1, there has been fruitful work on finding marker genes, 1.151 @@ -379,7 +378,7 @@ 1.152 Allen Brain Atlas raw data, and produce as output all 1.153 numbers and charts found in publications resulting from 1.154 the project. Source code to be released will include ex- 1.155 - tensions to Caret[7], an existing open-source scientific 1.156 + tensions to Caret[8], an existing open-source scientific 1.157 imaging program, and to Spectral Python. Data to be 1.158 released will include the 2-D “flat map” dataset. This 1.159 dataset will be submitted to a machine learning dataset 1.160 @@ -393,9 +392,9 @@ 1.161 area. Finding marker genes will be useful for drug discovery as well as for experimentation be- 1.162 cause marker genes can be used to design interventions which selectively target individual cortical 1.163 areas. 1.164 + The application of the marker gene finding algorithm to the cortex will also support the develop- 1.165 8 1.166 1.167 - The application of the marker gene finding algorithm to the cortex will also support the develop- 1.168 ment of new neuroanatomical methods. In addition to finding markers for each individual cortical 1.169 areas, we will find a small panel of genes that can find many of the areal boundaries at once. 1.170 The method developed in Goal 2 will provide a genoarchitectonic viewpoint that will contribute 1.171 @@ -410,12 +409,12 @@ 1.172 _ 1.173 Preliminary Results 1.174 Format conversion between SEV, MATLAB, NIFTI 1.175 -We have created software to (politely) download all of the SEV files10 from the Allen Institute 1.176 -website. We have also created software to convert between the SEV, MATLAB, and NIFTI file 1.177 -formats, as well as some of Caret’s file formats. 1.178 +We have created software to (politely) download all of the SEV files9 from the Allen Institute web- 1.179 +site. We have also created software to convert between the SEV, MATLAB, and NIFTI file formats, 1.180 +as well as some of Caret’s file formats. 1.181 Flatmap of cortex 1.182 We downloaded the ABA data and selected only those voxels which belong to cerebral cortex. 1.183 -We divided the cortex into hemispheres. Using Caret[7], we created a mesh representation of the 1.184 +We divided the cortex into hemispheres. Using Caret[8], we created a mesh representation of the 1.185 surface of the selected voxels. For each gene, and for each node of the mesh, we calculated an 1.186 average of the gene expression of the voxels “underneath” that mesh node. We then flattened 1.187 the cortex, creating a two-dimensional mesh. We converted this grid into a MATLAB matrix. We 1.188 @@ -434,11 +433,11 @@ 1.189 Correlation Recall that the instances are surface pixels, and consider the problem of attempt- 1.190 ing to classify each instance as either a member of a particular anatomical area, or not. The target 1.191 area can be represented as a boolean mask over the surface pixels. 1.192 - 10SEV is a sparse format for spatial data. It is the format in which the ABA data is made available. 1.193 + We calculated the correlation between each gene and each cortical area. The top row of Figure 1.194 +1 shows the three genes most correlated with area SS. 1.195 + 9SEV is a sparse format for spatial data. It is the format in which the ABA data is made available. 1.196 9 1.197 1.198 - We calculated the correlation between each gene and each cortical area. The top row of Figure 1.199 -1 shows the three genes most correlated with area SS. 1.200 Conditional entropy 1.201 For each region, we created and ran a forward stepwise procedure which attempted to find 1.202 pairs of genes such that the conditional entropy of the target area’s boolean mask, conditioned 1.203 @@ -547,9 +546,9 @@ 1.204 all genes at once, we ran a support vector machine to 1.205 classify cortical surface pixels based on their gene ex- 1.206 pression profiles. We achieved classification accuracy of 1.207 - about 81%11. However, as noted above, a classifier that 1.208 + about 81%10. However, as noted above, a classifier that 1.209 ____________________________________ 1.210 - 115-fold cross-validation. 1.211 + 105-fold cross-validation. 1.212 11 1.213 1.214 looks at all the genes at once isn’t as practically useful 1.215 @@ -577,7 +576,7 @@ 1.216 Our plan: what remains to be done 1.217 Flatmap cortex and segment cortical layers 1.218 There are multiple ways to flatten 3-D data into 2-D. We will compare mappings from manifolds to 1.219 -planes which attempt to preserve size (such as the one used by Caret[7]) with mappings which 1.220 +planes which attempt to preserve size (such as the one used by Caret[8]) with mappings which 1.221 preserve angle (conformal maps). We will also develop a segmentation algorithm to automatically 1.222 identify the layer boundaries. 1.223 Develop algorithms that find genetic markers for anatomical regions 1.224 @@ -646,9 +645,9 @@ 1.225 An area may be difficult to identify because the boundaries are misdrawn in the atlas, or be- 1.226 cause the shape of the natural domain of gene expression corresponding to the area is different 1.227 from the shape of the area as recognized by anatomists. We will develop extensions to our pro- 1.228 -cedure which (a) detect when a difficult area could be fit if its boundary were redrawn slightly12, 1.229 +cedure which (a) detect when a difficult area could be fit if its boundary were redrawn slightly11, 1.230 ____________________________________ 1.231 - 12Not just any redrawing is acceptable, only those which appear to be justified as a natural spatial domain of gene ex- 1.232 + 11Not just any redrawing is acceptable, only those which appear to be justified as a natural spatial domain of gene ex- 1.233 pression by multiple sources of evidence. Interestingly, the need to detect “natural spatial domains of gene expression” 1.234 in a data-driven fashion means that the methods of Goal 2 might be useful in achieving Goal 1, as well – particularly 1.235 13 1.236 @@ -685,7 +684,7 @@ 1.237 spond to interesting spatial regions. 1.238 Clustering and segmentation on pixels We will explore clustering and image segmentation 1.239 algorithms in order to segment the pixels into regions. We will explore k-means, spectral cluster- 1.240 -ing, gene shaving[9], recursive division clustering, multivariate generalizations of edge detectors, 1.241 +ing, gene shaving[10], recursive division clustering, multivariate generalizations of edge detectors, 1.242 multivariate generalizations of watershed transformations, region growing, active contours, graph 1.243 partitioning methods, and recursive agglomerative clustering with various linkage functions. These 1.244 methods can be combined with dimensionality reduction. 1.245 @@ -705,7 +704,7 @@ 1.246 identify spatial regions. It remains to be seen whether removal of redundancy would help or hurt 1.247 the ultimate goal of identifying interesting spatial regions. 1.248 Co-clustering We will explore some algorithms which simultaneously incorporate clustering 1.249 -on instances and on features (in our case, pixels and genes), for example, IRM[11]. These are 1.250 +on instances and on features (in our case, pixels and genes), for example, IRM[12]. These are 1.251 called co-clustering or biclustering algorithms. 1.252 Compare different methods In order to tell which method is best for genomic anatomy, for 1.253 each experimental method we will compare the cortical map found by unsupervised learning to a 1.254 @@ -729,7 +728,7 @@ 1.255 dence. There are three ways we will validate our marker genes to guard against this. First, we 1.256 will confirm that putative combinations of marker genes express the same pattern in both hemi- 1.257 spheres. Second, we will manually validate our final results on other gene expression datasets 1.258 -such as EMAGE, GeneAtlas, and GENSAT[8]. Third, we may conduct ISH experiments jointly with 1.259 +such as EMAGE, GeneAtlas, and GENSAT[9]. Third, we may conduct ISH experiments jointly with 1.260 collaborators to get further data on genes of particular interest. 1.261 Using the methods developed in Goal 2, we will present one or more hierarchical cortical 1.262 maps. We will identify and explain how the statistical structure in the gene expression data led to 1.263 @@ -746,8 +745,9 @@ 1.264 1.265 References Cited 1.266 [1] Chris Adamson, Leigh Johnston, Terrie Inder, Sandra Rees, Iven Mareels, and Gary Egan. 1.267 - A Tracking Approach to Parcellation of the Cerebral Cortex, volume 3749/2005 of Lecture 1.268 - Notes in Computer Science, pages 294–301. Springer Berlin / Heidelberg, 2005. 1.269 + A tracking approach to parcellation of the cerebral cortex. In Medical Image Computing 1.270 + and Computer-Assisted Intervention MICCAI 2005, volume 3749/2005 of Lecture Notes in 1.271 + Computer Science, pages 294–301. Springer Berlin / Heidelberg, 2005. 1.272 [2] J. Annese, A. Pitiot, I. D. Dinov, and A. W. Toga. A myelo-architectonic method for the struc- 1.273 tural classification of cortical areas. NeuroImage, 21(1):15–26, 2004. 1.274 [3] Tanya Barrett, Dennis B. Troup, Stephen E. Wilhite, Pierre Ledoux, Dmitry Rudnev, Carlos 1.275 @@ -757,34 +757,36 @@ 1.276 [4] George W. Bell, Tatiana A. Yatskievych, and Parker B. Antin. GEISHA, a whole-mount in 1.277 situ hybridization gene expression screen in chicken embryos. Developmental Dynamics, 1.278 229(3):677–687, 2004. 1.279 - [5] James P Carson, Tao Ju, Hui-Chen Lu, Christina Thaller, Mei Xu, Sarah L Pallas, Michael C 1.280 + [5] Thomas Boggs. Spectral python. http://spectralpython.sourceforge.net/, July 2008. 1.281 + [6] James P Carson, Tao Ju, Hui-Chen Lu, Christina Thaller, Mei Xu, Sarah L Pallas, Michael C 1.282 Crair, Joe Warren, Wah Chiu, and Gregor Eichele. A digital atlas to characterize the mouse 1.283 brain transcriptome. PLoS Comput Biol, 1(4):e41, 2005. 1.284 - [6] Mark H. Chin, Alex B. Geng, Arshad H. Khan, Wei-Jun Qian, Vladislav A. Petyuk, Jyl Boline, 1.285 + [7] Mark H. Chin, Alex B. Geng, Arshad H. Khan, Wei-Jun Qian, Vladislav A. Petyuk, Jyl Boline, 1.286 Shawn Levy, Arthur W. Toga, Richard D. Smith, Richard M. Leahy, and Desmond J. Smith. 1.287 A genome-scale map of expression for a mouse brain section obtained using voxelation. 1.288 Physiol. Genomics, 30(3):313–321, August 2007. 1.289 - [7] D C Van Essen, H A Drury, J Dickson, J Harwell, D Hanlon, and C H Anderson. An integrated 1.290 + [8] D C Van Essen, H A Drury, J Dickson, J Harwell, D Hanlon, and C H Anderson. An integrated 1.291 software suite for surface-based analyses of cerebral cortex. Journal of the American Medical 1.292 Informatics Association: JAMIA, 8(5):443–59, 2001. PMID: 11522765. 1.293 - [8] Shiaoching Gong, Chen Zheng, Martin L. Doughty, Kasia Losos, Nicholas Didkovsky, Uta B. 1.294 + [9] Shiaoching Gong, Chen Zheng, Martin L. Doughty, Kasia Losos, Nicholas Didkovsky, Uta B. 1.295 Schambra, Norma J. Nowak, Alexandra Joyner, Gabrielle Leblanc, Mary E. Hatten, and 1.296 Nathaniel Heintz. A gene expression atlas of the central nervous system based on bacte- 1.297 rial artificial chromosomes. Nature, 425(6961):917–925, October 2003. 1.298 - [9] Trevor Hastie, Robert Tibshirani, Michael Eisen, Ash Alizadeh, Ronald Levy, Louis Staudt, 1.299 +[10] Trevor Hastie, Robert Tibshirani, Michael Eisen, Ash Alizadeh, Ronald Levy, Louis Staudt, 1.300 Wing Chan, David Botstein, and Patrick Brown. ’Gene shaving’ as a method for identifying dis- 1.301 tinct sets of genes with similar expression patterns. Genome Biology, 1(2):research0003.1– 1.302 research0003.21, 2000. 1.303 -[10] Jano Hemert and Richard Baldock. Matching Spatial Regions with Combinations of Interact- 1.304 - ing Gene Expression Patterns, volume 13 of Communications in Computer and Information 1.305 - Science, pages 347–361. Springer Berlin Heidelberg, 2008. 1.306 -[11] C Kemp, JB Tenenbaum, TL Griffiths, T Yamada, and N Ueda. Learning systems of concepts 1.307 +[11] Jano Hemert and Richard Baldock. Matching spatial regions with combinations of interact- 1.308 + ing gene expression patterns. In Bioinformatics Research and Development, volume 13 of 1.309 + Communications in Computer and Information Science, pages 347–361. Springer Berlin Hei- 1.310 + delberg, 2008. 1.311 + 16 1.312 + 1.313 +[12] C Kemp, JB Tenenbaum, TL Griffiths, T Yamada, and N Ueda. Learning systems of concepts 1.314 with an infinite relational model. In AAAI, 2006. 1.315 -[12] F. Kruggel, M. K. Brckner, Th. Arendt, C. J. Wiggins, and D. Y. von Cramon. Analyzing the 1.316 +[13] F. Kruggel, M. K. Brckner, Th. Arendt, C. J. Wiggins, and D. Y. von Cramon. Analyzing the 1.317 neocortical fine-structure. Medical Image Analysis, 7(3):251–264, September 2003. 1.318 - 16 1.319 - 1.320 -[13] Ed S. Lein, Michael J. Hawrylycz, Nancy Ao, Mikael Ayres, Amy Bensinger, Amy Bernard, 1.321 +[14] Ed S. Lein, Michael J. Hawrylycz, Nancy Ao, Mikael Ayres, Amy Bensinger, Amy Bernard, 1.322 Andrew F. Boe, Mark S. Boguski, Kevin S. Brockway, Emi J. Byrnes, Lin Chen, Li Chen, 1.323 Tsuey-Ming Chen, Mei Chi Chin, Jimmy Chong, Brian E. Crook, Aneta Czaplinska, Chinh N. 1.324 Dang, Suvro Datta, Nick R. Dee, Aimee L. Desaki, Tsega Desta, Ellen Diep, Tim A. Dolbeare, 1.325 @@ -806,49 +808,49 @@ 1.326 Yaylaoglu, Rob C. Young, Brian L. Youngstrom, Xu Feng Yuan, Bin Zhang, Theresa A. Zwing- 1.327 man, and Allan R. Jones. Genome-wide atlas of gene expression in the adult mouse brain. 1.328 Nature, 445(7124):168–176, 2007. 1.329 -[14] Susan Magdaleno, Patricia Jensen, Craig L. Brumwell, Anna Seal, Karen Lehman, Andrew 1.330 +[15] Susan Magdaleno, Patricia Jensen, Craig L. Brumwell, Anna Seal, Karen Lehman, Andrew 1.331 Asbury, Tony Cheung, Tommie Cornelius, Diana M. Batten, Christopher Eden, Shannon M. 1.332 Norland, Dennis S. Rice, Nilesh Dosooye, Sundeep Shakya, Perdeep Mehta, and Tom Cur- 1.333 ran. BGEM: an in situ hybridization database of gene expression in the embryonic and adult 1.334 mouse nervous system. PLoS Biology, 4(4):e86 EP –, April 2006. 1.335 -[15] Lydia Ng, Amy Bernard, Chris Lau, Caroline C Overly, Hong-Wei Dong, Chihchau Kuan, 1.336 +[16] Lydia Ng, Amy Bernard, Chris Lau, Caroline C Overly, Hong-Wei Dong, Chihchau Kuan, 1.337 Sayan Pathak, Susan M Sunkin, Chinh Dang, Jason W Bohland, Hemant Bokil, Partha P 1.338 Mitra, Luis Puelles, John Hohmann, David J Anderson, Ed S Lein, Allan R Jones, and Michael 1.339 Hawrylycz. An anatomic gene expression atlas of the adult mouse brain. Nat Neurosci, 1.340 12(3):356–362, March 2009. 1.341 -[16] George Paxinos and Keith B.J. Franklin. The Mouse Brain in Stereotaxic Coordinates. Aca- 1.342 +[17] George Paxinos and Keith B.J. Franklin. The Mouse Brain in Stereotaxic Coordinates. Aca- 1.343 demic Press, 2 edition, July 2001. 1.344 -[17] A. Schleicher, N. Palomero-Gallagher, P. Morosan, S. Eickhoff, T. Kowalski, K. Vos, 1.345 +[18] A. Schleicher, N. Palomero-Gallagher, P. Morosan, S. Eickhoff, T. Kowalski, K. Vos, 1.346 K. Amunts, and K. Zilles. Quantitative architectural analysis: a new approach to cortical 1.347 mapping. Anatomy and Embryology, 210(5):373–386, December 2005. 1.348 -[18] Oliver Schmitt, Lars Hmke, and Lutz Dmbgen. Detection of cortical transition regions utilizing 1.349 + 17 1.350 + 1.351 +[19] Oliver Schmitt, Lars Hmke, and Lutz Dmbgen. Detection of cortical transition regions utilizing 1.352 statistical analyses of excess masses. NeuroImage, 19(1):42–63, May 2003. 1.353 -[19] S.B. Serpico and L. Bruzzone. A new search algorithm for feature selection in hyperspec- 1.354 +[20] S.B. Serpico and L. Bruzzone. A new search algorithm for feature selection in hyperspec- 1.355 tral remote sensing images. Geoscience and Remote Sensing, IEEE Transactions on, 1.356 39(7):1360–1367, 2001. 1.357 - 17 1.358 - 1.359 -[20] Constance M. Smith, Jacqueline H. Finger, Terry F. Hayamizu, Ingeborg J. McCright, Janan T. 1.360 +[21] Constance M. Smith, Jacqueline H. Finger, Terry F. Hayamizu, Ingeborg J. McCright, Janan T. 1.361 Eppig, James A. Kadin, Joel E. Richardson, and Martin Ringwald. The mouse gene expres- 1.362 sion database (GXD): 2007 update. Nucl. Acids Res., 35(suppl_1):D618–623, 2007. 1.363 -[21] Larry Swanson. Brain Maps: Structure of the Rat Brain. Academic Press, 3 edition, November 1.364 +[22] Larry Swanson. Brain Maps: Structure of the Rat Brain. Academic Press, 3 edition, November 1.365 2003. 1.366 -[22] Carol L. Thompson, Sayan D. Pathak, Andreas Jeromin, Lydia L. Ng, Cameron R. MacPher- 1.367 +[23] Carol L. Thompson, Sayan D. Pathak, Andreas Jeromin, Lydia L. Ng, Cameron R. MacPher- 1.368 son, Marty T. Mortrud, Allison Cusick, Zackery L. Riley, Susan M. Sunkin, Amy Bernard, 1.369 Ralph B. Puchalski, Fred H. Gage, Allan R. Jones, Vladimir B. Bajic, Michael J. Hawrylycz, 1.370 and Ed S. Lein. Genomic anatomy of the hippocampus. Neuron, 60(6):1010–1021, Decem- 1.371 ber 2008. 1.372 -[23] Pavel Tomancak, Amy Beaton, Richard Weiszmann, Elaine Kwan, ShengQiang Shu, 1.373 +[24] Pavel Tomancak, Amy Beaton, Richard Weiszmann, Elaine Kwan, ShengQiang Shu, 1.374 Suzanna E Lewis, Stephen Richards, Michael Ashburner, Volker Hartenstein, Susan E Cel- 1.375 niker, and Gerald M Rubin. Systematic determination of patterns of gene expression during 1.376 drosophila embryogenesis. Genome Biology, 3(12):research008818814, 2002. PMC151190. 1.377 -[24] Shanmugasundaram Venkataraman, Peter Stevenson, Yiya Yang, Lorna Richardson, 1.378 +[25] Shanmugasundaram Venkataraman, Peter Stevenson, Yiya Yang, Lorna Richardson, 1.379 Nicholas Burton, Thomas P. Perry, Paul Smith, Richard A. Baldock, Duncan R. Davidson, 1.380 and Jeffrey H. Christiansen. EMAGE edinburgh mouse atlas of gene expression: 2008 up- 1.381 date. Nucl. Acids Res., 36(suppl_1):D860–865, 2008. 1.382 -[25] Axel Visel, Christina Thaller, and Gregor Eichele. GenePaint.org: an atlas of gene expression 1.383 +[26] Axel Visel, Christina Thaller, and Gregor Eichele. GenePaint.org: an atlas of gene expression 1.384 patterns in the mouse embryo. Nucl. Acids Res., 32(suppl_1):D552–556, 2004. 1.385 -[26] Robert H Waterston, Kerstin Lindblad-Toh, Ewan Birney, Jane Rogers, Josep F Abril, Pankaj 1.386 +[27] Robert H Waterston, Kerstin Lindblad-Toh, Ewan Birney, Jane Rogers, Josep F Abril, Pankaj 1.387 Agarwal, Richa Agarwala, Rachel Ainscough, Marina Alexandersson, Peter An, Stylianos E 1.388 Antonarakis, John Attwood, Robert Baertsch, Jonathon Bailey, Karen Barlow, Stephan Beck, 1.389 Eric Berry, Bruce Birren, Toby Bloom, Peer Bork, Marc Botcherby, Nicolas Bray, Michael R 1.390 @@ -863,6 +865,8 @@ 1.391 cinda A Fulton, Robert S Fulton, Terrence S Furey, Diane Gage, Richard A Gibbs, Gustavo 1.392 Glusman, Sante Gnerre, Nick Goldman, Leo Goodstadt, Darren Grafham, Tina A Graves, 1.393 Eric D Green, Simon Gregory, Roderic Guig, Mark Guyer, Ross C Hardison, David Haussler, 1.394 + 18 1.395 + 1.396 Yoshihide Hayashizaki, LaDeana W Hillier, Angela Hinrichs, Wratko Hlavina, Timothy Holzer, 1.397 Fan Hsu, Axin Hua, Tim Hubbard, Adrienne Hunt, Ian Jackson, David B Jaffe, L Steven John- 1.398 son, Matthew Jones, Thomas A Jones, Ann Joy, Michael Kamal, Elinor K Karlsson, Donna 1.399 @@ -870,8 +874,6 @@ 1.400 drew Kirby, Diana L Kolbe, Ian Korf, Raju S Kucherlapati, Edward J Kulbokas, David Kulp, 1.401 Tom Landers, J P Leger, Steven Leonard, Ivica Letunic, Rosie Levine, Jia Li, Ming Li, Chris- 1.402 tine Lloyd, Susan Lucas, Bin Ma, Donna R Maglott, Elaine R Mardis, Lucy Matthews, Evan 1.403 - 18 1.404 - 1.405 Mauceli, John H Mayer, Megan McCarthy, W Richard McCombie, Stuart McLaren, Kirsten 1.406 McLay, John D McPherson, Jim Meldrim, Beverley Meredith, Jill P Mesirov, Webb Miller, Tra- 1.407 cie L Miner, Emmanuel Mongin, Kate T Montgomery, Michael Morgan, Richard Mott, James C