ultimi comportamenti

2026-03-27 12:21:45 +01:00
parent 65915a5463
commit 2dbb8eeb3a
1 changed files with 77 additions and 2 deletions
@@ -318,17 +318,92 @@ Limita rilascio NT: Dipende da POST che tende a bloccare rialascio di NT se non
 #### CaTrace concentration
 The biological meaning is that a synapse that has just been through a burst is primed for fast recovery — the molecular machinery for vesicle docking is already engaged, calcium-dependent priming factors are still elevated, and the system is in a ready state. A synapse that has been silent for several seconds has cooled down and replenishes slowly.
 Serve a dare la velocita' al trasporto di vesicles da RP a RRP. Ha un decadimento proprio il che dice alla Presinapsi di accellerare se da poco c'e' stata una spike, altrimenti di andare piu' piano. So after one second of silence Tr_Ca has fallen to ~37% of its peak value, after two seconds to ~14%, after three seconds to ~5%. It asymptotes toward zero but never exactly reaches it. Between spikes, Ca2+ falls toward zero as the pumps clear it.
 spike → Ca_micro rises → Tr_Ca rises slowly
                       → Tr_Ca stays elevated for ~1-2 s after burst
                       → recruitment speed elevated during that window
                       → RRP refills faster
                       → more vesicles available for the next burst
 Tr_Ca = Tr_Ca + (Ca_micro - Tr_Ca / tau_Tr_Ca) * dt
 This is a leaky integrator — it has two competing forces at every step. The Ca_micro term pushes Tr_Ca upward toward the current calcium level. The - Tr_Ca / tau_Tr_Ca term pulls it back down toward zero at a rate proportional to how high it already is. The balance between these two forces means Tr_Ca is always chasing Ca_micro but never quite reaching it, and always decaying when Ca_micro is low.
 During a spike, Ca_micro jumps sharply. Tr_Ca rises but more slowly — it integrates rather than jumps. Between spikes, Ca_micro falls back toward zero as the pumps clear it. Tr_Ca also falls, but much more slowly because tau_Tr_Ca = 1000 ms — it remembers the spike for roughly a second after it happened. Over a burst of many spikes, Tr_Ca climbs steadily as each spike adds to the residual trace before the previous one has fully decayed. A long silence after a burst allows Tr_Ca to decay exponentially back toward zero.
 The result is that Tr_Ca encodes not the instantaneous calcium level but the recent history of calcium activity — a smoothed, time-averaged measure of how active the synapse has been over the past one to two seconds.
 #### RP->RRP shuttling
-Dipende da?
+How RP is moved to RRP
 This happens in the seconds loop, once per second. The recruitment step reads Tr_Ca and moves a fraction of the available RP into the RRP:
 current_recruitment_rate = map_trace_to_speed(Tr_Ca)   # /s
 refill_amount = current_recruitment_rate *dt_slow_s* N_RP * (Max_RRP - N_RRP) / Max_RRP
 refill_amount = max(0.0, refill_amount)
 refill_amount = min(refill_amount, N_RP)
 N_RRP = min(N_RRP + refill_amount, Max_RRP)
 N_RP  = max(0.0, N_RP - refill_amount)
 Three factors multiply together to determine how many vesicles move:
 current_recruitment_rate — set by Tr_Ca history. High recent activity means faster recruitment. This is the only place Tr_Ca has any effect in the model.
 N_RP — how many vesicles are available in the reserve. As N_RP depletes, fewer vesicles are available to move regardless of how fast the rate is. This is the depletion bottleneck — a synapse that has been firing for minutes may have a high Tr_Ca driving fast recruitment, but if N_RP is nearly empty there is nothing left to recruit.
 (Max_RRP - N_RRP) / Max_RRP — the headroom in the RRP, normalised. When the RRP is nearly full, recruitment slows automatically because there is little room to fill. When the RRP is empty after a burst, headroom is maximal and recruitment runs at full speed. This prevents overfilling and makes the system self-regulating — recruitment is fastest precisely when it is most needed.
 The two guard clauses ensure the arithmetic stays physical: refill_amount cannot be negative, and cannot exceed what N_RP actually contains.
 ### min: behavior
 #### Refill RP from Glutamine
-Dipende da Glutamine messa a disposizione dall'Astrocyte
+How RP is replenished
 This happens in the minutes loop, once per minute, via the glutamine shuttle from the astrocyte. It is a two-step process across two cells.
 Step 1 — astrocyte side
 The astrocyte has been accumulating cleared glutamate from the cleft since the last minutes-loop execution. Its glutamine synthetase enzyme converts that glutamate into glutamine, filling the Glutamine_pool. The fraction successfully converted per cycle is conversion_efficiency, which is set by glucose availability and boosted temporarily if the astrocyte calcium wave fired during the preceding seconds:
 refill_RP = Glutamine_pool * conversion_efficiency
 Glutamine_pool = max(0.0, Glutamine_pool - refill_RP)
 Step 2 — presynapse side
 The glutamine crosses into the presynapse, where glutaminase converts it back into glutamate. That glutamate is immediately repackaged into vesicles and added to N_RP:
 N_RP = min(Max_RP, N_RP + refill_RP)
 The ceiling guard `min(Max_RP, ...)` prevents N_RP from exceeding its physical maximum.
 ---
 **The full chain from release to re-availability**:
 ms:       NT released from RRP  →  N_RRP falls
                                →  NT_cleft rises
                                →  NT_released_this_window accumulates
 seconds:  EAATs clear NT_cleft  →  glutamate enters astrocyte
          Tr_Ca gates speed     →  N_RP moves into N_RRP
                                →  (partially restores what was just depleted)
 mins:     glutamate → glutamine →  Glutamine_pool filled
          glutamine → N_RP     →  reserve restocked for next seconds cycle
 ---
 **The asymmetry that makes depletion possible**:
 The chain reveals why sustained high-frequency firing eventually depletes the synapse even with all replenishment mechanisms running.
 The RRP holds at most `Max_RRP = 20` vesicles. At 20 Hz with strong Ca²⁺, release can draw 2-4 vesicles per spike — potentially exhausting the RRP in under a second. The seconds loop can move vesicles from RP to RRP at a maximum rate of `k_rec_fast = 5 /s`, meaning at most 5 vesicles per second under ideal conditions. Release outpaces recruitment by roughly an order of magnitude during a burst.
 The RP holds up to `Max_RP = 200` vesicles — ten times the RRP. At sustained 20 Hz the RP can sustain firing for tens of seconds even after the RRP is repeatedly emptied, as long as recruitment keeps pace. But the minutes loop only refills N_RP once per minute at a rate limited by `Glutamine_pool * conversion_efficiency`. If glucose is low or the astrocyte wave has not fired, this replenishment may add only a fraction of what was consumed.
 The result is a three-tier buffer with mismatched timescales:
 RRP  — depletes in seconds,   refilled in seconds   (fast but shallow)
 RP   — depletes in minutes,   refilled in minutes   (deep but slow)
 Gln  — depletes over bursts,  refilled by glucose   (slowest, astrocyte-dependent)
 Each tier buys time for the one below it to respond. When all three are depleted simultaneously — which only happens under prolonged high-frequency firing with insufficient glucose — the synapse has no remaining buffer and goes silent until the minutes loop restores the Glutamine_pool.
 ## BEH-PRE-VGCC: Container