I'm trying to fit a binding data with a user defined function (derived from single site binding equation) "y=Fm*((P+x+K)-sqrt((P+x+K)^2-4*P*x))/(2*P)". This equation has been worked for others and all my data except one set (an important one unfortunately). Here is a picture of my problematic fitting.
The picture is not great but you can get the idea. The red one is fitting with Hill equation and green is fitting with my equation. The green one is apparently not fitting to the data points! Anyone know how to fix this problem?
I digitized your data in 8.6 and fit with your function and it was mostly within 0.01 (in Y) of the Hill function. Almost indistinguishable on the graph.
What Origin version are you using? Is there some data not shown on the graph that may be skewing the fit?
This is all the data but I can send you the original data if you would like to take a look. I have 8 on one of my computer and 8.5 on the other one. Both are giving me the same weird fitting though.Should I try a newer version?
Non linear curve fitting problem
Posted - 10/30/2012 : 10:12:49 PM
