Models and Code Samples
The models on this page can be regarded as open source and are free to all. These models were created with the program Molecular Modeling Pro, version 8.1 and can be modified and improved by anyone owning the program. Place the models in the /mmp/models subdirectory and access them through the Calculate/user models menu item to predict property values. This page will be frequently updated.
Substructure Addition Model Date Download Link Database used
r^2
n
std dev
Author
Comments and notes
Log Kow Jan 18 2018 Model of Log_Kow.txt chemelectrica.mdb
0.961
4202
0.56
James Quinn
Amalgamated, no restrictions, no inclusions, no cross-products
Specific gravity Mar 26 2018 Model of Lit_density.txt chemelectrica.mdb
0.998
1914
0.059
James Quinn
Amalgamated, no restrictions, no inclusions, cross-products and division, one pass
Specific gravity - organic only Jan 24 2018 Model_of_Lit_density_organic_only.txt chemelectrica.mdb
0.980
1624
0.056
James Quinn
Amalgamated, restricted to organic molecules, no cross-products.
Boiling Point Dec 23 2017 Model of Lit_Boiling_Point__C_.txt chemelectrica.mdb
0.997
1850
24 C
James Quinn
Amalgamated, no restrictions, no inclusions, with cross-products; all molecules with values in the database
HLB Mar 19 2018 Model_of_Literature_HLB.txt
industrials.mdb
0.984
218
0.86
James Quinn
Surfactants only; amalgamated, no includions, cross-products, divisions; 2 iterations
If you have models you would like to share, please submit them for inclusion here to info@norgwyn.com. Note: Models created from the data in ChemicaElectrica with no restrictions and inclusions have the following kinds of molecules: household products, gasses, solvents, surfactants (ionic and non-ionic), pharmaceuticals, dyes, pesticides, flavors and fragrances, biochemically interesting molecules, and a limited number of inorganic molecules and polymers. Outliers in the analysis were not removed. Data was collected from many, many sources.
Free code samples. This code can be used to create the substructure variables from structure and is similar to a subroutine used by MMP to create the terms used in the substructure models above. Also included is code to transform the data and then revert it back to the original. These samples are Open Source (free to use, use at your own risk.) If you modify the Make_Substructures.txt file to create different output and would like to submit the new code for inclusion here and in MMP, e-mail it with a description to info@norgwyn.com. Similarily, if you code in a different language and would like to make the code available here, e-mail it to us.
Code Link Description Date Author
Make_Substructures.txt Input structural variables (number of atoms, molecular number, connections, bonds) and output substructure variables. Accepts any atom type. Amalgamates H, OH, =O, halogen, NH2, #N, NO2, CH3, =S with the atom it is attached to. Looks for PEG and PPG, urea, carbamate, carbon chains and a few other specific groups, but will create substructure variables for both organic and inorganic compounds. Jan 20, 2018 James Quinn
Transform_Data.txt Outputs y, y^2, sqrt(y), logit y, log10(y) and inverse of all these; multiplies or divides by molecular weight and/or volume Jan 20, 2018 James Quinn
Antitransform_Data.txt Takes output of Tranform_Data.bas and reverts it to the original value. Jan 20, 2018 James Quinn
Go to the Molecular Modeling Pro page on norgwyn.com