Skip to content
Buy Spikeling units

Fluorescence imaging simulation

This page explains the fluorescence imaging simulation in the Spikeling GUI: what it represents, how it is computed from membrane potential (Vm), and how to use it for teaching.

Spikeling’s imaging simulation is not a camera and not a microscope pipeline. It is a conceptual bridge between: - fast electrical activity (Vm and spikes), and - slower optical readouts (calcium and fluorescence) that students commonly see in modern systems neuroscience.

See also: - Concepts (why Vm can be transformed into fluorescence) - GUI overview (where the Imaging screen lives) - Teaching hub → Lab 4 — Imaging (structured activity)

**What this simulation is **

  • a teaching model that converts Vm / spiking activity into: 1) a simulated intracellular calcium trace, and
    2) a simulated fluorescence trace
  • a way to explain why fluorescence signals are:
  • delayed relative to spikes
  • smoother (low-pass filtered)
  • often non-linear and baseline-dependent
  • a tool for demonstrating common analysis concepts such as ΔF/F

The conceptual pipeline: Vm → calcium → fluorescence

The simulation is built around three simple ideas that match how students should reason about imaging data:

1) Spikes drive calcium influx

Spikes (or spike-like events inferred from Vm) are treated as the main driver of calcium entry.

Teaching translation:

“A spike happens quickly, but calcium rises and decays more slowly.”

2) Calcium is slow compared to Vm

The calcium trace is a filtered representation of spiking: it rises with activity and decays with a characteristic time constant.

Teaching translation:

“Even if Vm returns to baseline, calcium can stay elevated.”

3) Fluorescence is a transformation of calcium

The fluorescence trace is a further transformation of calcium into an optical-like signal. In real imaging, this relationship can be non-linear and depends on indicator kinetics and baseline.

Teaching translation:

“Fluorescence is an indirect proxy for spikes, not a direct measurement of Vm.”

Where to find it in the GUI

Go to:

Imaging → Imaging Simulation

The imaging screen is designed to look and behave like the main oscilloscope page, but for calcium/fluorescence.

How to use the Imaging Simulation screen

Step 1 — Choose a source

Spikeling imaging can be driven by either:

  • Spikeling hardware (live Vm stream from the device), or
  • GUI emulator (no hardware required; best for demonstrations)

Use the “source” selector (e.g., from GUI emulator). If you are teaching a class without enough devices, the emulator is often the cleanest option.

Step 2 — Choose which traces to display

Typical display options include:

  • Vm (Spikeling and optional Synapse channels)
  • Calcium (Spikeling and optional Synapse channels)
  • Fluorescence (Spikeling and optional Synapse channels)
  • Stimulus (so students can align cause and effect)

Step 3 — (Optional) Enable ΔF/F

ΔF/F is a common normalised fluorescence metric.

  • F is a baseline fluorescence level (often estimated from a low-activity period)
  • ΔF is the change relative to baseline
  • ΔF/F makes it easier to compare recordings with different baseline brightness

Teaching translation:

“ΔF/F is a normalised measure of relative activity rather than absolute brightness.”

Step 4 — Axis scaling

If available, Auto Range Y axis is helpful in early demos. For comparative labs (same stimulus, different mode), it is often better to keep axis ranges fixed so students can compare amplitudes honestly.

Step 5 — Record imaging data

The imaging screen includes a recording section similar to the Neuron Interface:

  1. Choose a directory
  2. Enter a filename
  3. Press Record

This produces an export students can analyse later (e.g., comparing Vm vs fluorescence timing).

What students should learn from it

Key lesson 1: fluorescence is delayed

A spike occurs in milliseconds. Calcium and fluorescence typically peak later. Students should be able to answer: - “Why does fluorescence peak after spikes?” - “Why can fluorescence remain elevated after spiking stops?”

Key lesson 2: fluorescence is smoother

Vm can contain sharp spikes and fast oscillations. Fluorescence is slower and smoother. Students should be able to explain: - why fast voltage features are lost in fluorescence - why individual spikes can merge into a single broad transient at high firing rates

Key lesson 3: the mapping is not one-to-one

At low firing rates, individual spikes may produce distinct transients. At higher rates, transients overlap and it becomes difficult to infer exact spike timing.

This supports a core modern neuroscience idea:

Imaging is excellent for population activity patterns, but it is not a direct substitute for electrophysiology when precise spike timing is required.

Suggested demonstrations (work well in teaching)

Demo A — one stimulus, two neuron modes

  1. Keep the stimulus the same (e.g., square-wave or a step protocol).
  2. Switch neuron mode (tonic spiking vs adapting / phasic).
  3. Observe:
  4. Vm spikes differ strongly across modes
  5. calcium/fluorescence patterns differ (peak height, decay, steady-state)

Teaching question:

“Does fluorescence reflect ‘number of spikes’, ‘burstiness’, or both?”

Demo B — frequency sweep (chirp / ZAP-like input)

  1. Apply a chirp stimulus (linear or exponential sweep).
  2. Observe Vm resonance-like behaviours (if present) and how fluorescence averages over them.

Teaching question:

“Which features survive the Vm → fluorescence transformation?”

Demo C — near-threshold noise

  1. Hold Vm just below threshold (DC injection / patch clamp).
  2. Add noise until occasional spikes occur.
  3. Observe that fluorescence becomes an intermittent transient signal.

Teaching question:

“Why is imaging often analysed statistically rather than spike-by-spike?”

Common pitfalls and how to explain them

“Fluorescence does not look like Vm”

Correct: it should not. Fluorescence is downstream of calcium kinetics and indicator dynamics.

“My fluorescence amplitude changed when I changed baseline”

That is expected. Baseline and normalisation (ΔF/F) matter a lot in optical measurements.

“I cannot infer exact spike timing from fluorescence”

Also expected, especially at high firing rates where transients overlap.