pK_a calculation training

If you think your experimental data improved the accuracy of the default pK_a calculator, you can take advantage of the supervised pK_a learning method that is built into the pK_a calculator. Special structural parts can have an effect on the pK_a values calculated by the built-in method, so your correction library based on your experimental data helps the pK_a calculator increase the prediction accuracy.

Training set

Inaccurately predicted ionization centers need to be identified and experimental data for them have to be collected. Since the learning algorithm is based on linear regression analysis, therefore you need to collect as many experimental pK_a data as possible otherwise there won't be correlation. There is no strict rule on the number of the experimental datapoints, if your purpose is to create a local model, only for a certain type of chemical of the ionization centers, then it may be enough to collect a few representative structures. A robust model, however, requires as many diverse structures and pK_a values as possible.

The experimental data should be collected in an sdf file, then the training algorithm has to be run which creates a correction library. This will be stored on your local computer, in your user folder. Finally, this correction library via MarvinSketch, cxcalc, Chemical Terms can be applied.

• Structure of the molecule
• pK_a value 1 (field name: pKa1)
• ID of the atom which has the pKa1 value (field name: ID1). It can be viewed by checking the Atom number option in MarvinView (menu: View > Misc).
• Additional fields of pK_a values are optional (recommended for handling multiprotic compounds). For example pK_a value 2 (pKa2), ID2, etc.
• Definition of only one pK_a value is enough to apply the training data, but more values in case of multiprotic compounds will enhance the reliability of the pK_a training.

A sample of a typical training set is shown in the picture (pKa_trainingset.sdf). ID1 is the index of the atom with the experimental pK_a1 value.

Generate the correction library

cxtrain pka -i [library name] [training file] 

cxtrain pka -i mypka mydata.sdf

Application of the correction library

MarvinSketch

1. Select MarvinSketch menu:Tools > Protonation > pKa.
2. Set the 'Use correction library' box to activate the training option (see figure below).
3. If you have created multiple training sets, choose the most accurate one from the dropdown list below the checkbox.



The next figure shows the results with (I) and without (II) applying the correction library.

I. pKa calculation with training data	II. pKa calculation without training data

Cxcalc

To include your correction library in the pK_a calculation use the parameter --correctionlibrary or its short form: -L.

cxcalc pKa  --correctionlibrary  [library name] [input file/string]

Example

$ cxcalc pKa --correctionlibrary mypka "CSC1=NC2=C(N1)C=NC(O)=N2"

Result

 id      apKa1   apKa2   bpKa1   bpKa2   atoms

 1       11.19   16.01   2.34    -2.59   7,11,9,4


If you use cxcalc pKa calculation without the correction library, the results will be calculated with the built-in dataset.
Example


$ cxcalc pKa "CSC1=NC2=C(N1)C=NC(O)=N2"

Result

 id      apKa1   apKa2   bpKa1   bpKa2   atoms

 1       8.34   16.01   2.34    -2.59   7,11,9,4

For more options see this page.

Chemical Terms

Chemical Terms are available either from command line, evaluator or from Instant JChem.

Command line, evaluator

Evaluator is designed to evaluate mathematical expressions on molecules. Your correction library can be applied as follows:


evaluate -e "pKa('correctionlibrary:[library name]')" "[input file/string]"

Example

evaluate -e "pKa('correctionlibrary:mypka')" "CSC1=NC2=C(N1)C=NC(O)=N2"

or Result


;;;-2,59;;;11,19;;2,34;;16,01;

For more evaluator functions on pKa training see this page.

Instant JChem

Choose the 'New Chemical Terms Field icon' and type the chemical term into the window, use the correctionlibrary:[library name] parameter. Do not forget to adjust the Name, the Type and the DB Column Name.
Example
The following figure demonstrates the usage of pK_a training in the 'New Chemical terms' window. The expression pKa ('correctionlibrary:mypKa type:acidic','1') defines that the plugin use the correction library named mypKa, and it will calculate the strongest acidic pK_a of the molecule(s).



The results of this calculation are shown in the figure below. You can see the difference between the untrained (column 5., Strongest acidic pKa) and trained (column 6., Trained strongest acidic pKa) pK_a values.