This manual gives you a walk-through on how to train the log D Plugin
If you think your experimental data could improve the performance of the default log D calculator, you can take advantage of the supervised log D learning method that is built into the calculator.
The log D method can be trained by applying existing p K a and log P training libraries. For detailed information on training p K a and log P plugins, see the corresponding p K a and log P training manuals.
To apply the pre-generated training library in MarvinSketch, see the following steps:
Choose Calculations > Partitioning > logD in MarvinSketch.
Select the User defined training method.
If you have many log P training sets, you can select the one you want to use for training from the logP training ID dropdown list.
If you have many p K a correction libraries, you can select the one that you want to use for training by enabling the Use pKa correction library option, and choosing the library from the dropdown list.
Fig. 1 Applying log P and pKa training libraries for log D training
Trained value : pH logD 7.40 -0.34 Untrained value : pH logD 7.40 -0.08
To apply your p K a correction library to train the log D method with cxcalc, use the --pkacorrectionlibrary option :
cxcalc logd `--method [method] --pkacorrectionlibrary` [library name] [input file/string]
To apply your log P dataset to train the log D method, use the --method main option, combined with the --logptrainingid secondary option :
cxcalc logd `--method [method] --logptrainingid` [library name] [input file/string]
cxcalc logd --method user --pkacorrectionlibrary mypka_1 --logptrainingid mylogp_1 --pH 7.4 "CC1=NC2=C(N1)C(O)=NC(N)=N2"
cxcalc logd --pH 7.4 "CC1=NC2=C(N1)C(O)=NC(N)=N2"
Trained value : id logD[pH=7.4] 1 -0.34 Untrained value : id logD[pH=7.4] 1 -0.08
Chemical Terms functions can be evaluated using the Chemical Terms Evaluator command line tool. They are also available in e.g. JChem.
The pkacorrectionlibrary and logptrainingid parameters can be applied as Chemical Terms parameters as well. For example:
evaluate -e "logd('method:[method] pkacorrectionlibrary:[library name] logptrainingid:[id]')" [input file/string]
evaluate -e "logd('method:user pkacorrectionlibrary:mypkalib_1 logptrainingid:mylogp_1 pH:7.4')" "CC1=NC2=C(N1)C(O)=NC(N)=N2" (trained)
evaluate -e "logd('pH:7.4')" "CC1=NC2=C(N1)C(O)=NC(N)=N2"
Trained value : 7.4;-0.34 Untrained value : 7.4;-0.08