Manuscript outline

[aarmey]

I know it's very early, but think it still could be useful to think about what an initial manuscript would look like.

Figures

Figure 1

• Initial plot of proteomic data (clustergram?) - see #23
• Data-driven selection of network nodes from OHSU

Figure 2

• Schematic of autoencoder structure
• Toy problem?

Figure 3

• Training against actual data
• Description of fit model

Figure 4

• Cell line perturbation
• Description of that data

Figure 5

• Model/validation comparison

Figure 6

• Could have even a cartoon description of other encoder structures
• Point of this would be to speak to the generality of the approach
• This could be left out with the idea of addressing it in a review or some such later

[aarmey]

From call: Move a schematic of the overall model up to the top of the figures.

@aarmey will move this outline to the manubot files.

[aarmey]

Moved to the manuscript.