Mean of the first N readings in the time window (after blank subtraction). SD is the standard deviation of those N readings.
Fmax / Tmax
Maximum of the (smoothed) curve and the time it occurs; ΔF = Fmax − F0, fold = Fmax / F0.
Threshold & call
A curve crosses the threshold when n consecutive readings are ≥ the threshold; the crossing time is interpolated linearly between neighbouring points. Wells that cross before the cut-off time with ΔF > 0 are positive. “Negative-control mean + k×SD” uses all readings of the negative-control group.
Lag time
Threshold = time the threshold is crossed (time to threshold); amplitude % = first time F0 + p%·ΔF is reached; tangent = the max-slope tangent meeting F = F0, t_lag = t_ms − (F_ms − F0) / k_max; sigmoid fit F = A1 + (A2 − A1) / (1 + e^((t½ − t)/τ)) (Levenberg–Marquardt), lag time = t½ − 2τ, apparent rate k = 1/τ.
1/lag time
Reciprocal of the lag time, usable as a rate (called RAF in some papers).
T50
First time the curve reaches F0 + ½ΔF (or ½ Fmax), interpolated linearly.
Time to plateau
Time the curve enters its plateau. Default = first sustained time at F0 + 90%·ΔF; tangent = the max-slope tangent meeting F = Fmax, t = t_ms + (Fmax − F_ms) / k_max; fit = t½ + 2τ. The instant of the maximum reading jumps around on a noisy plateau and does not mark plateau entry; it is hidden by default (tick it under “Columns…”).
Max slope
dF/dt from a sliding linear regression over w points of the smoothed curve; the maximum is reported with its time and signal.
Group positive
A group (sample) is positive when the share of positive replicates is ≥ the setting (default 50%, e.g. ≥ 2 of 4 wells).
AUC
Trapezoidal integral over the time window; with baseline removal it integrates F − F0. Units: signal × time.