DynaFit Scripting Manual

  Index

Experimental error

Initial reaction velocities are always affected by finite experimental error. The magnitude and type of experimental uncertainty associated with the measurements of initial velocities is expressed in chapter 7.4 discussing reaction progress curves.

For example, the following script assigns a constant error to all initial velocity data points. The standard deviation is equal to 1.5% of the maximum data value found across all four data sets (files, f1 through f4).

[velocity]
   error percent 1.5
   directory ./test/data
   extension txt
   variable I
   file f1 | concentration S = 20
   file f2 | concentration S = 40
   file f3 | concentration S = 80
   file f4 | concentration S = 160
[end]

It is also possible to specify a nonconstant error variance, as is described in section 7.4.3, or a constant error specified in its absolute value (section 7.4.1).

Standard errors of measurements

In addition to defining the standard error of measurements by using a linear, quadratic, or cubic ``error function'', described in the preceding paragraph, we can also define the standard errors of measurements directly in the data file. This is accomplished by using the keyword combination error data, shown in the example below:

[velocity]
   error data
   directory ./test/data
   extension txt
   variable I

   ; Text files 'f1.txt' through 'f4.txt'
   ; must contain THREE columns:
   ;
   ;    column 1: concentration [I]
   ;    column 2: initial velocity 'v'
   ;    column 3: standard error of 'v'

   file f1 | concentration S = 20
   file f2 | concentration S = 40
   file f3 | concentration S = 80
   file f4 | concentration S = 160
[end]

In this case the program will assume that all data files contain three columns of ASCII text. The third column must contain the standard deviation of initial velocities. One example of such data file is shown below:

; data file "020710-2SD-8nM"
; active site titration
;
[I] v         error
--- --------  --------
0   0.7047    0.009512
1   0.4729    0.009351
2   0.3226    0.007252
4   0.1426    0.00873
6   0.06624   0.008714
8   0.0377    0.00869

If and when the data files contain only two columns, the program will ignore the error data instruction and assign equal weights to all data points.

  Index

biokin.com/dynafit/scripting/html/node51.html
Petr Kuzmic | Jul 12 2005