1.1 --- a/grant.html Mon Apr 20 16:21:13 2009 -0700 1.2 +++ b/grant.html Mon Apr 20 16:57:54 2009 -0700 1.3 @@ -581,15 +581,27 @@ 1.4 but as a sample, PCA, NNMF, and landmark 1.5 Isomap are shown in the first, second, and third 1.6 rows of Figure 6. 1.7 -After applying the dimensionality reduction, we ran clustering algorithms on the reduced data. To date we have tried 1.8 -k-means and spectral clustering. The results of k-means after PCA, NNMF, and landmark Isomap are shown in the last 1.9 -row of Figure 6. To compare, the leftmost picture on the bottom row of Figure 6 shows some of the major subdivisions of 1.10 -cortex. These results clearly show that different dimensionality reduction techniques capture different aspects of the data 1.11 -and lead to different clusterings, indicating the utility of our proposal to produce a detailed comparion of these techniques 1.12 -as applied to the domain of genomic anatomy. 1.13 -Many areas are captured by clusters of genes 1.14 -todo 1.15 -todo 1.16 + 1.17 +Figure 7: Prototypes corresponding to sample gene clusters, clustered by 1.18 +gradient similarity. Region boundaries for the region that most matches 1.19 +each prototype are overlayed. After applying the dimensionality reduc- 1.20 + tion, we ran clustering algorithms on the re- 1.21 + duced data. To date we have tried k-means and 1.22 + spectral clustering. The results of k-means after 1.23 + PCA, NNMF, and landmark Isomap are shown 1.24 + in the last row of Figure 6. To compare, the 1.25 + leftmost picture on the bottom row of Figure 1.26 + 6 shows some of the major subdivisions of cor- 1.27 + tex. These results clearly show that different di- 1.28 + mensionality reduction techniques capture dif- 1.29 + ferent aspects of the data and lead to differ- 1.30 + ent clusterings, indicating the utility of our pro- 1.31 + posal to produce a detailed comparion of these 1.32 + techniques as applied to the domain of genomic 1.33 + anatomy. 1.34 +Many areas are captured by clusters of genes We also clustered the genes using gradient similarity to see if the 1.35 +spatial regions defined by any clusters matched known anatomical regions. Figure 7 shows, for ten sample gene clusters, 1.36 +each cluster’s average expression pattern, compared to a known anatomical boundary. 1.37 _________________________________________ 1.38 195-fold cross-validation. 1.39 Research plan