Event detection: Difference between revisions
Expand Nanopore Wiki source page with structured educational content |
Expand Nanopore Wiki source page with structured educational content |
||
| (4 intermediate revisions by the same user not shown) | |||
| Line 2: | Line 2: | ||
<translate> | <translate> | ||
<!--T:1--> | <!--T:1--> | ||
[[File: | [[File:Patch_clamp_schematic_1.png|thumb|right|320px|Event detection identifies signal deviations from a stable baseline.]] | ||
'''Event detection''' is part of the Nanopore Wiki learning path. Event detection identifies real blockade or signal-change events in nanopore recordings and extracts comparable features from them. | '''Event detection''' is part of the Nanopore Wiki learning path. Event detection identifies real blockade or signal-change events in nanopore recordings and extracts comparable features from them. | ||
| Line 13: | Line 13: | ||
<!--T:3--> | <!--T:3--> | ||
== Scientific basis == | |||
* Detection thresholds should be tied to baseline and local noise. | |||
* Event boundaries affect dwell time, area, and shape metrics. | |||
* Interpretations should be tied to measurable quantities, experimental controls, and documented analysis settings. | |||
* Claims should distinguish direct observations from downstream inference. | |||
<!--T:4--> | |||
== Event detection workflow == | == Event detection workflow == | ||
{| class="wikitable np-knowledge-table" | {| class="wikitable np-knowledge-table" | ||
| Line 34: | Line 41: | ||
|} | |} | ||
<!--T: | <!--T:5--> | ||
== Typical applications == | == Typical applications == | ||
* Quantifying blockade depth and dwell time. | * Quantifying blockade depth and dwell time. | ||
| Line 40: | Line 47: | ||
* Building training data for classification models. | * Building training data for classification models. | ||
<!--T: | <!--T:6--> | ||
== | == Quality checks == | ||
* Record platform, pore or flow-cell type, chemistry, voltage or acquisition settings, and software versions. | |||
* Use negative controls, positive controls, and replicate measurements where the claim depends on specificity. | |||
* Report limitations such as noise, bias, sample degradation, pore instability, or model-dependent analysis. | |||
* Keep laboratory metadata linked to the raw signal or sequencing run before interpreting results. | |||
<!--T:7--> | |||
== Interpretation limits == | |||
* Using one threshold for recordings with strong drift. | * Using one threshold for recordings with strong drift. | ||
* Failing to test false positives on blank controls. | * Failing to test false positives on blank controls. | ||
* | * A single result should not be generalized beyond the sample, protocol, instrument state, and analysis pipeline used. | ||
* Medical, legal, regulatory, or commercial decisions require domain-specific validation beyond this educational summary. | |||
<!--T: | <!--T:8--> | ||
== Related pages == | == Related pages == | ||
* [[Data analysis]] | * [[Data analysis]] | ||
| Line 53: | Line 68: | ||
* [[Troubleshooting]] | * [[Troubleshooting]] | ||
<!--T: | <!--T:9--> | ||
== References == | == References == | ||
* [https://en.wikipedia.org/wiki/Detection_theory Signal detection theory] | * [https://en.wikipedia.org/wiki/Detection_theory Signal detection theory] | ||
Latest revision as of 09:25, 15 July 2026
Event detection is part of the Nanopore Wiki learning path. Event detection identifies real blockade or signal-change events in nanopore recordings and extracts comparable features from them.
Core concepts
- Detection thresholds should be tied to baseline and local noise.
- Event boundaries affect dwell time, area, and shape metrics.
- Weak, overlapping, or multi-level events need careful validation.
- False positives can arise from noise, drift, bubbles, clogging, and analysis settings.
Scientific basis
- Detection thresholds should be tied to baseline and local noise.
- Event boundaries affect dwell time, area, and shape metrics.
- Interpretations should be tied to measurable quantities, experimental controls, and documented analysis settings.
- Claims should distinguish direct observations from downstream inference.
Event detection workflow
| Step | Description |
|---|---|
| 1 | Annotate experimental phases and sample addition times. |
| 2 | Estimate local baseline and noise. |
| 3 | Set candidate thresholds and minimum duration rules. |
| 4 | Merge, split, or reject events using documented criteria. |
| 5 | Export event tables and plots for review. |
Typical applications
- Quantifying blockade depth and dwell time.
- Comparing samples or pore variants.
- Building training data for classification models.
Quality checks
- Record platform, pore or flow-cell type, chemistry, voltage or acquisition settings, and software versions.
- Use negative controls, positive controls, and replicate measurements where the claim depends on specificity.
- Report limitations such as noise, bias, sample degradation, pore instability, or model-dependent analysis.
- Keep laboratory metadata linked to the raw signal or sequencing run before interpreting results.
Interpretation limits
- Using one threshold for recordings with strong drift.
- Failing to test false positives on blank controls.
- A single result should not be generalized beyond the sample, protocol, instrument state, and analysis pipeline used.
- Medical, legal, regulatory, or commercial decisions require domain-specific validation beyond this educational summary.