Home
Results
Download

Mouse Brain data: More complex patterns

General notes applying to all analysis:

  • Results images are filtered to remove genes with mostly negative expression values.
  • Images show data that has been adjusted to mean zero, variance one, and clipped to increase contrast. More on reading images here.
  • The image column headings refer to "subA"; this is an artifact of how we set this up, and both subA and subB arrays were used for the analysis.
  • Clicking on an image will show you the detailed results for that gene.
  • There are also HTML versions of the data. At this point the genes which have mostly negative values are still in those tables. A warning will be shown when you encounter one which would be filtered.
  • You can access Sandberg's raw data file lines corresponding to a probe or accession number by entering it in the form; similarly you can get this from the HTML table versions of the data.
  • In each figure, genes with p values <0.001 are shown. The leftmost column shows a black tick if Sandberg listed the gene; the second column is the average expression level of the gene (clipped at 1000). The rest of the columns show the data, adjusted to a mean of zero and variance of one, and clipped to the range -1 to 1 (this increases the contrast). The genes are listed in order of increasing p value.
  • Annotations are not necessarily complete as given here.

These pictures represent some experimentation with complex patterns. The first image is what you get if you look for genes which are high in the cerebellum of the 129 strain but low in the black 6 strain, or vice versa. We actually manage to pull out one or two genes. The second image is an easier target: genes which are high/low in the cerebellum of black 6 only.

High in 129 Cb, low in black6 Cb

High in black 6 Cb only

References

--