Table 4.
Markram–Tsodyks model best-fit parameters for different conditions.
| Cond | A | U | F(ms) | D (ms) | A' | U' | A'' | U'' | %Err | Conf. |
|---|---|---|---|---|---|---|---|---|---|---|
| [Ca2+]o 2 →4 | 3.0285 | 0.3422 | 19.8 | 128.4 | 2.5821 | 0.5057 | 2.7 | 0.0003 | ||
| [Ca2+]o 2 →4 | 2.7921 | 0.3676 | 152.7 | 125.4 | 2.7921 | 0.4694 | 3.9 | 0.0002 | ||
| HP: 0.1 →10.1 [Ca2+]o: 4 | 2.7824 | 0.4676 | 160.7 | 137.4 | 1.5361 | 0.4324 | 2.1 | 0.0005 | ||
| HP: 0.1 →10.1 [Ca2+]o: 4 | 2.8584 | 0.4511 | 25000 | 138.4 | 1.5202 | 0.4511 | 2.5 | 0.0001 | ||
| HP: 5.1 →10.1 [Ca2+]o: 2 →4 | 2.0318 | 0.3201 | 1400 | 105.8 | 1.4027 | 0.4731 | 3.3 | 0.0004 | ||
| HP: 0.1 →10.1 [Ca2+]o: 2 →2 →4 | 3.0940 | 0.3087 | 30000 | 97.0 | 1.5708 | 0.3173 | 1.3482 | 0.4912 | 3.9 | 0.009 |
The table contains the best-fit parameters, accounting for the experimental data with the minimal number of parameters affected by the condition. Only U and A were found to be significantly affected by the changes in [Ca2+]o and pressure, respectively while the other parameters remained fixed. The rightmost columns report for each case the fit error on the absolute response amplitudes and the best-fit confidence. For the experiments involving two conditions (i.e., before and after; the first five rows of the table), 2 × 5 fEPSPs were available to be fitted by the model. Therefore, five (e.g., A, U, F, D, A’) or six (e.g., A, U, F, D, A’, U’) free parameters were identified simultaneously. In bold we outlined the parameters that were allowed to change and, as expected, the lesser degrees or freedom, the higher fit confidence (underlined).
For the other experiments, 3 × 5 fEPSPs were available to be fitted by the model, therefore requiring identifying eight (e.g., A, U, F, D, A’, U’, A’’, U’’) parameters.