| Author |
 Topic  |
|
|
geofabien
United Kingdom
1 Posts |
 Posted - 12/01/2020 : 08:24:36 AM
|
Hi all,
I'm trying to use a modified form of the gompertz equation to model a bacterial growth curve with a sigmoidal fit of absorbance vs time data. I tried to edit the original SGompertz function to obatin the modified version in the fitting tab.
original:y = a*exp(-exp(-k*(x-xc)))
modified: y = A*exp(-exp(umax*e/A*(L-t)+1))
where umax is the maximum specific growth (linear part of curve)
e is exp(1)
A is the absorbance (on y axis)
L is the lag time, in this case the tangent line through the inflection point
t is time (on x axis)
I have tried to create this function on origin labs but it doesn't fit the data at all.
I have limited maths and coding knowledge, could anyone help me create this function properly in the fitting function builder?
This is what I have as the initialisation code
sort( x_y_curve );
//smooth( x_y_curve, 2 );
a = max(y_data);
x_y_curve = ln( x_y_curve );
x_y_curve = ln( ln(a) - x_y_curve );
double coeff[2];
fitpoly( x_y_curve, 1, coeff );
xc = -coeff[0] / coeff[1];
k = -coeff[1];
I think this is what I need to change to make it fit properly.
Here are some picture of how I've tried to set it up and the fit it gave me.
 |
|
|
Chris D
428 Posts |
Posted - 12/01/2020 : 09:22:49 AM
|
Hi,
If you will send the following to tech@originlab.com, we'll take a look:
1. Your Origin version.
2. The last 7 digits of your serial number
3. An example dataset (as an Origin OPJU file).
4. The FDF file for the fitting function you worked on. It will be in your User Files folder in the fitfunc subfolder.
Thanks,
Chris Drozdowski
Originlab Technical Support
|
 |
|
|
YimingChen
1647 Posts |
Posted - 12/01/2020 : 1:39:12 PM
|
In the Parameter and Variable definition page, you should set t as independent variable and A as fitting parameter.
James |
Edited by - YimingChen on 12/01/2020 1:39:35 PM |
|
|
| |
Topic |
|
|
|
Modified Gompertz Equation - help with function