A database is useless unless you have a means of putting the data in. We have not yet completed the "submission" interfaces but we know what things we want to be able to do. The first is we must be able to use the anatomy to label our gene-expression patterns, that is fairly easy. We must also be able to enter patterns directly on the atlas and for this we are going to use two methods. The first is simple "painting" of the gene-expression pattern onto the atlas using something like the paint program, the second is to use morphing which bends a picture of the experimental data so that it matches the atlas.