Simulations¶

With a micron sink SWC and synpts in hand, instantiate the Arbor cell (TSModel / TSRecipe), restore loops with gap junctions at # CYCLE_BREAK / # MULTI_NECK sites, generate synaptic event streams, and record voltages.

Each spine with synpts has a package under simulations/. Shared analysis lives in simulations/common/; you normally edit only the per-spine files.

Run from the repository root as modules.

uv run python -m simulations.ts1.axons
uv run python -m simulations.ts1.axons_dash
uv run python -m simulations.ts76.axons_dash
uv run python -m simulations.ts2.ts2_integration
uv run python -m simulations.ts2.ts2_integration --verbose

Build the NMODL catalogue first if toric_spines_sim/mechanisms/custom-catalogue.so is missing or stale (see Getting started). Confirm with check_catalogue():

from toric_spines_sim.model import check_catalogue
check_catalogue()

If that raises, rebuild from the repository root:

uv run bash scripts/make_custom_catalogue.sh

Precomputed morphologies already in data/ are data/swc/microns/TS{id}_wsink_r10um.swc (also r5um). Synapse coordinates are data/pointsets/microns/TS{id}_synpts.txt. Expected inputs for an axon study also include data/ts_axons/ts{id}_axons.txt. PDF / HTML outputs go under simulations/ts{id}/axons/results/.

simulations/
  README.md
  common/                 shared axon PDF / Dash / animation helpers
  ts1/                    (same pattern for ts2, ts3, ts4, ts48, ts67, ts76)
    params.py             returns make_icx_parameter_bank_invivo(); switch factory or set .value to tweak
    inputs.py             SWC / synpts / axon map + scenario options
    axons.py              PDF study
    axons_dash.py         3D Dash dashboard
    axons_animation.py    HTML voltage animation
    neurosignature/       TS1-only descriptor pipeline
    hypergrid/            TS1-only parameter sweep
  ts2/
    ts2_integration.py    extra Poisson integration + PDF
    ts2_integration_viz.py

Tweak biophysics in params.py. Each spine returns make_icx_parameter_bank_invivo(); uncomment make_default_parameter_bank() or make_icx_parameter_bank_invitro(), or set .value on sampled keys, for a per-spine change. Tutorials use make_default_parameter_bank(). Tweak which axons fire, rates, and pulse timing in inputs.py. Leave common/ alone unless a change should apply to every spine.

Shared vs per-model¶

Path	Role
`simulations/common/utils.py`	Axon colors, pulse times, topology load (`n_axons` from the assignment file)
`simulations/common/inputs.py`	`InputScenario` builders (pulse / periodic / Poisson)
`simulations/common/axons_report.py`	Pairwise / sequential / synchrony PDF study
`simulations/common/axons_dash.py`, `axons_animation.py`	Dash and HTML animation helpers
`simulations/ts{id}/params.py`	Returns `make_icx_parameter_bank_invivo()`; uncomment another factory or set `.value` to tweak
`simulations/ts{id}/inputs.py`	Axon map path, scenario options (`ACTIVE_AXONS`, rates, sequential order)
`simulations/ts{id}/axons.py`	Thin `run_axons_study(MODEL)` entry point

Axon maps¶

Format (data/ts_axons/ts{id}_axons.txt): one axon per line, comma-separated 1-based synapse indices covering 1..N uniquely (N = lines in the matching synpts file). Empty lines are ignored; comments are not supported.

ts1_axons.txt is the original TS1 map (10 axons, 25 synapses). The other files are placeholders (round-robin into min(10, N) axons) so the shared pipeline can run; replace them when real axon assignments are known.

File	Synapses	Axons	Notes
`ts1_axons.txt`	25	10	original map
`ts2_axons.txt`	6	6	placeholder
`ts3_axons.txt`	46	10	placeholder
`ts4_axons.txt`	23	10	placeholder
`ts48_axons.txt`	18	10	placeholder
`ts67_axons.txt`	8	8	placeholder
`ts76_axons.txt`	5	5	placeholder

TS21 and TS24 have sink SWCs but no TS*_synpts.txt, so they are not wired up yet.

Active runners¶

Path	Role
`simulations/ts{id}/axons.py` (ts1, ts2, ts3, ts4, ts48, ts67, ts76)	Axon PDF study (`python -m simulations.ts{id}.axons`)
`ts{id}/axons_dash.py`, `axons_animation.py`	Dash dashboard and HTML voltage animation
`ts1/neurosignature/ts1_neurosignature.py`	Neurosignature descriptor pipeline (`uv sync --group neurosignature`; design)
`ts1/hypergrid/ts1_hypergrid.py`	TS1 HyperGrid parameter sweep
`ts2/ts2_integration.py`	TS2 Poisson integration + PDF
`ts2/ts2_integration_viz.py`	TS2 HTML frusta animation

All of the above use TSSimulator and toric_spines_sim.paths.

Archived¶

Superseded scripts live under archive/simulations/ (ts1/original/, standalone ts2_sim.py, empty ts3/ts4 placeholders). Do not run them.