Peak tissue ΔT—Numerical grid maximum · not junction T
Synchronized experiment notebook
Readouts & controls
Solid ink: soma/light trial. Blue: selected compartment. Dashed brass: identical-drive dark control. Shading: illumination. Click a trace to inspect time.
Cell
Opsin
Soma mW/mm²
Events / dark
In-light / dark
Reduction
Thermal field & energy ledger · unresolved hotspots
Enable heat to calculate a temperature field.
Numerical diagnostics & interpretation limits
Review & reproduce
Designer / cleanroom bridge
Bring a device into the laboratory
Import a native optoelectrode file or a Neural Interface Designer layout. Electrode footprints are not LEDs until explicitly converted. Cleanroom recipes are retained as provenance only.
Drop a JSON file here · maximum 1 MB
No import pending.
Conversion uses 470 nm, 0.02 mW/site, 6.5% efficiency, +Z normal and z=−75 µm. These explicit assumptions remain editable and are retained in the exported metadata.
100 automated numerical/import tests passed, including all 46 retained core tests. Three real Chromium flows passed 65 assertions covering calculations, replay, cancellation, imports, exports, desktop and 390-pixel mobile layout. Physical iPhone and Safari have not been tested.
The reduced photocycle implementation matches 48 independent SciPy matrix-exponential cases to about 1 × 10⁻¹⁴ in state probability. Nine independent chloride integrations agree to about 6 × 10⁻¹⁴ mM. These checks verify the chosen equations, not biological calibration.
The included short inhibition refinement (0.01 → 0.005 ms) changed somatic voltage by at most 0.0024 mV. The inherited morphology-refinement limit remains. Thermal hotspots are unresolved: the microLED example peaks at 0.557 °C on 80 µm voxels and 1.012 °C on 48 µm voxels. No safety threshold is asserted.
Portable contracts & source evidence
Accepted: bts-opto-experiment/3, bts-opto-device/1, bts-neural-interface version 1 with explicit µm units, and Microfabrication Laboratory recipe metadata. Maximum 16 emitters, 256 display contacts, 8 cells. No URLs, scripts, STL/GLB meshes, fabricated efficiency or electrical continuity are inferred from a layout.
Scope: the following ledger describes the retained single-neuron ChR2/fiber core. New chloride, reduced opsin, microLED, population and thermal assumptions are documented in the Expanded Laboratory and its exported diagnostics.
A reconstructed human cell
Allen specimen 571654895, a human middle temporal gyrus layer-2 spiny cell, with published fitted active membrane parameters. The 40 µm mesh has 211 membrane compartments and 75 zero-area junctions. A 20 µm mesh is available for refinement.
Dendritic geometry comes from the source reconstruction. The active axon is the source-prescribed 60 µm synthetic stub. The morphology does not establish a full biological axon or its initiation site.
ChR2(H134R) follows the Williams et al. four-state model with its activation variable, voltage dependence, temperature factors, and current–voltage fit. Each expressing membrane compartment has its own state. Conductance density is an assumed experimental input, not measured expression in this specimen.
The source fit was characterized at approximately −80 to −10 mV. This cell rests near −92 mV after conditioning and crosses positive voltages during a spike, so both regions use the mathematical extrapolation. All neuron experiments use 34 °C, matching the native channel model.
Uniform illumination provides a controlled local-exposure experiment. The fiber uses a seeded Monte Carlo field for a 100 µm core, NA 0.22, 470 nm source in homogeneous bulk tissue. Power scales the field; its orientation and position determine compartment exposure.
There is no tissue surface, implant obstruction, blood-vessel geometry, or thermal solver. The displayed cone indicates source direction; the light overlay samples the computed field.
Native membrane channels and opsin current enter the compartmental cable solve. The neuron conditions for 100 ms before the displayed experiment. Somatic events are detected at the solver timestep; the replay uses saved samples.
Each export includes the full experiment configuration. The CSV contains computed traces and local exposure; it is not an experimental recording. Channel assay is one ideal isopotential membrane patch, not a perfect voltage clamp of the branched cell.
The fiber kernel is scalar fluence rate, integrated over photon directions, in mW/mm² per mW emitted. It is not directional irradiance at a flat detector. Uniform light is a prescribed local exposure. Passing a diffuse fluence rate to an opsin model characterized with irradiance assumes a matching effective photon exposure; that calibration is not established for this human specimen.
The original 1.3 photon-loss factor in the Williams fit is retained even with tissue attenuation. Its transfer from channel characterization to tissue fluence is an explicit coupling assumption, not a measured tissue correction. Optical energy and integrated fluence are not tissue temperature or safety estimates.
The current fit is evaluated as ḡ(O₁ + 0.1 O₂)[10.6408 − 14.6408 exp(V / −42.7671)]. This removes the division by V in the published form. The fitted current is not forced to zero at V = 0; positive-voltage and resting-voltage use is extrapolation.
This is a computational research and learning model. Numerical checks and source attribution are distinct from experimental validation of the combined optical–neuron prediction. The Expanded Laboratory adds exploratory chloride, opsin, microLED, independent-population and diagnostic heating models. Connected networks and experimental validation remain outside this review.
Attribution and bundled licenses
Human neuron data and channel mechanisms: Allen Institute / Mosher, Wei et al.; source commit 750e7840cc1a9eb32492c924e79609a85a893b8a. Cable implementation recovered from the earlier BuildTheSimulation review snapshot (98ae75a826a536293158637360b173460f877602). Allen material has a noncommercial redistribution condition. ChR2 equations: Williams et al. (2013), independently implemented here. Installed C059 / Century Schoolbook and URW Gothic are used when available, with system fallbacks; no font binaries are included. The package contains source, notices and executable numerical tests.
Full Allen Institute license
Allen Institute Software License - This software license is the 2-clause BSD
license plus a third clause that prohibits redistribution for commercial
purposes without further permission.
Copyright 2020. Allen Institute. All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
3. Redistributions for commercial purposes are not permitted without the
Allen Institute's written permission.
For purposes of this license, commercial purposes is the incorporation of the
Allen Institute's software into anything for which you will charge fees or
other compensation. Contact [email protected] for commercial licensing
opportunities.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
Single-neuron response surface
When does a pulse produce one spike?
Scan local light and pulse width. Select a completed cell to inspect its actual compartment recordings.
Uses the current cell, expression, mesh and timestep. Each trial starts from the same dark-conditioned state. Uniform light, one pulse at 20 ms, at least 50 ms after pulse offset; no trial noise or spike-probability claim. Maximum 16 trials.
No response map yet. The main neuron setup supplies expression and numerical settings.
0 · no event1 · one event2+ · multiple eventsNumbers count post-onset somatic threshold crossings in the recorded window.