What the chart shows
The x-axis represents ordered runs or time, and the y-axis shows a selected control or method quantity. A center line represents a baseline mean; additional lines commonly represent one, two, and three baseline standard deviations.
The display makes shifts, trends, isolated extremes, and increased variability visible. It does not explain cause or establish that the underlying distribution and baseline are appropriate.
Choose one interpretable quantity per chart
Possible bioassay quantities include reference EC50, control recovery, relative potency for a stable control sample, slope, fitted asymptotes, replicate CV, or signal window. Each monitors a different part of the method.
Keep concentration units, transformation, sample/control lot, method version, instrument context, and calculation definition stable. Log potency-like quantities may be more appropriate when ratios are scientifically meaningful.
Version the baseline and limits
Select a period of routine, stable performance with enough independent runs and relevant operating variation. Investigate known special causes before estimating a baseline, but do not delete inconvenient points merely to narrow the limits.
Record included run identifiers, date range, method version, transformation, center-line calculation, dispersion calculation, and effective date. Rebaselining should create a new version rather than rewriting historical limits.
Worked example: turn a baseline into chart lines
If a stable control has a baseline mean of 100% recovery and a standard deviation of 4 percentage points, the +1 SD and −1 SD lines are 104% and 96%. The ±2 SD lines are 108% and 92%, and the ±3 SD lines are 112% and 88%.
These numbers describe the illustrative baseline; they are not universal assay limits. A 109% result is above +2 SD in this example, but its meaning depends on the selected rule, prior points, method context, and investigation procedure.
| Chart line | Recovery |
|---|---|
| +3 SD | 112% |
| +2 SD | 108% |
| +1 SD | 104% |
| Mean | 100% |
| −1 SD | 96% |
| −2 SD | 92% |
| −3 SD | 88% |
Keep every signal connected to run context
Annotate reagent and control lots, analyst, instrument, maintenance, method changes, known deviations, and investigations. Clicking a point should open the source analysis and report.
A signal should trigger the response defined by a procedure. It is not automatically a failed assay, and a point within limits is not proof the run meets all suitability criteria.
Frequently asked questions
How many runs are needed to establish limits?
There is no universal count. It depends on stability, independence, distribution, method variation, and intended use. Use statistical expertise and document the selected baseline.
Should acceptance limits and control limits be the same?
Not necessarily. Acceptance limits define a method or product decision; control limits summarize expected process behavior. Display and label both when both matter.