Non-Linear Curve Fit

General model

DataView can fit parameters to general non-linear models defined by the user with the built-in expression parser. Up to 8 parameters can be fitted.

The data have been fitted to a model comprising an offset cosine wave. The data were generated as an offset noisy sine wave from within DataView, and saved as a user-defined file.

Membrane time constant

A specialised version of non-linear curve fitting estimates membrane time constants using either exponential, double exponential, or error-function models.

The falling phase of the voltage response to an injected pulse of negative current is fitted to a double exponential function. The final value of the voltage has been constrained in the model (with data derived by measuring the average voltage over the last 40 msec), and the initial voltage, time constants and weights are estimated by curve fitting. The current pulse was injected through a bridge circuit, so the initial voltage value is not certain. The same data could be fitted to a single exponential model appropriate for cell body recording or an error function appropriate to an infinite cable, and yield different values for initial voltage and time constant.