Any non-linear least-squares regression software can be used for this purpose, such as SigmaPlot, Grafit, XLfit (an Excel add-in) as long as it has the right equation or can be programmed with it.

The usual equation for a simple 1:1 binding is y=a[L]/(Kd + [L]) + b, where y is the measure of binding, a+b is the maximal signal at saturation, b is the background signal, and [L] is the ligand concentration. SigmaPlot and XLfit have this equation built in.

However, a poor fit to this equation will occur if the experiment is done in the tight binding regime because this equation assumes that the free ligand concentration is equal to the total ligand concentration, which isn't true when the Kd is close to the receptor concentration. If that is the case, you should use this equation:

[RL] = {Kd + RT + LT - [(Kd + RT + LT)2-4RTLT]1/2}/2

where RT and LT are the total receptor and ligand concentrations and [RL] is the concentration of the complex. I find Grafit to be convenient for fitting this type of equation because it has a simple way of entering the equation and the data.

Thank you Adam

OriginLab  is another software 

Thank u

Following.