Table 3.
List of parameters used in simulation
| gCa | 1.1 mS.cm−2 | C | 1 μF cm−2 | ka | 0.1 μM |
|---|---|---|---|---|---|
| gK | 2 mS.cm−2 | υ | 0.2 m s−1 | kr | 0.4 μM |
| gL | 0.46 mS.cm−2 | I bgm | 27 μA | β | 5 μM s−1 |
| V Ca | 100 mV | I bgs | 0 μA | m | 4 |
| VK | −70 mV | u low | 0.2 | n | 2 |
| VL | −65 mV | u up | 0.2 | Ip | 0.11 μM s−1 |
| V 1 | −1 mV | τnu | 50 ms | γ | 50 μM s−1 |
| V 2 | 15 mV | τr | 600 ms | [Ca2+]o | 2 mM |
| V 3 | 0 mV | τau | 250 ms | V th | 10 mV |
| V 4 | 30 mV | τg | 30 s | w ee | 4.0 |
| Vr | 0 mV | ξave | 0.02 | w ei | 4.0 |
| Vi | −90 mV | ηmax | 0.32 ms−1 | b | 0 |
Notes: With the exceptions of τau and τg, all the other parameter values used in this work were obtained from previous works (Tsodyks et al. 2000; Volman et al. 2007; Huang et al. 2017b). Throughout all the simulation runs, the leaking channel conductance parameter (gL) was set slightly lower (0.46 mS) than the previous works (0.5 mS) to induce more spiking activities. The value of release constant for residual calcium (γ) was used as 33 μM s−1 (Huang et al. 2017b), whereas (Volman et al. 2007) used 80 μM s−1. In our simulations γ was tuned to 50 μM s−1. After fixing the astrocytic uptake time constant τau equal to the experimentally obtained τdecay value (see Fig. 3C), glutamate recovery to the Z state by the neuronal uptake mechanism τnu was tuned to 50 ms to match with the experimental firing and bursting patterns. Glutamate retrieval from the Z state to the ready to release active zone X was set (600 ms) within the range of “kiss-n-run” mode of vesicle retrieval τr (Gandhi and Stevens 2003). All the other remaining parameters were identical to the previous works (Tsodyks et al. 2000; Volman et al. 2007; Huang et al. 2017b).