Table 1.
The upper part of the table lists the different model assumptions made. A tick indicates that the model (for a given column) contains an assumption (given by the row) while a cross indicates that it does not. The lower part (highlighted in gray) summarizes the model output. The first row indicates whether each model is capable of producing a linear relationship between the fitness protein and the selection pressure. The second row shows whether we can observe a unimodal shift to the right (increase in the production) of the fitness protein in response to stress. The next row shows whether an increase in the mean mRNA level is observed in response to selection pressure. Finally, the maximum mean ratio of the fitness protein to the reference protein for each model is displayed in the last row. The only model that captures the observed data (unimodal shift, mRNA increase, and Ohm’s Law) is model 8, which also worked when tested for the regulated case and which we adopt in the main paper.
| Model | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| mRNA | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Protein | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Biased partitioning | ✗ | ✗ | ✗ | ✗ | ✓ | ✗ | ✗ | ✗ | ✓ | ✗ |
| Global transcription feedback | ✗ | ✗ | ✗ | ✓ | ✗ | ✓ | ✗ | ✓ | ✗ | ✗ |
| Global translation feedback | ✗ | ✗ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | ✗ | ✗ |
| Constitutive | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ | ✗ |
| Regulated | ✗ | ✗ | ✗ | ✓ | ✗ | ✗ | ✗ | ✗ | ✗ | ✓ |
| No. of parameters | 2 | 3 | 3 | 4 | 3 | 4 | 4 | 5 | 4 | 5 |
| Ohm’s Law | ✗ | ✗ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | ✗ | ✓ |
| Unimodal shift | ✗ | ✗ | ✗ | ✗ | ✗ | ✗ | ✓ | ✓ | ✗ | ✗ |
| mRNA increase | ✗ | ✗ | ✗ | ✓ | ✗ | ✗ | ✗ | ✓ | ✗ | ✓ |
| Max ratio | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.75 | 1.8 | 1.0 | 2.1 |