read smiles and return smiles of canonical tautomer

Dear support team,

I am struggling to add the neccesary assemblies and create a code for reading smiles and returning a SMILES for canonical tautomer.

TautomerizationPlugin plugin = new TautomerizationPlugin();

plugin.setProtectAllTetrahedralStereoCenters(true);

plugin.setProtectAromaticity(true);

plugin.setProtectDoubleBondStereo(true);

plugin.setProtectEsterGroups(true);

plugin.setProtectLabeledTetrahedralStereoCenters(true);

plugin.setTakeCanonicalForm(true);

plugin.setMolecule(??

 

Ken

Hi Karen,

in the license forum post I mentioned that you need to use the ChemAxon.IKVM.marvinbeans.dll.

However it will require dependencies as well. You had better add all the dlls from that folder JChem.NET API.

Namespace is:

using chemaxon.marvin.calculations;

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using chemaxon.marvin.calculations;

namespace TautomerPluginTest {
    class Program {
        static void Main(string[] args) {

            TautomerizationPlugin plugin = new TautomerizationPlugin();
            plugin.setProtectAllTetrahedralStereoCenters(true);
            plugin.setProtectAromaticity(true);
            plugin.setProtectDoubleBondStereo(true);
            plugin.setProtectEsterGroups(true);
            plugin.setProtectLabeledTetrahedralStereoCenters(true);
            plugin.setTakeCanonicalForm(true);
            //plugin.setMolecule(??
        }
    }
}

This code compiles on my machine after having added all dlls. The dll hell wil be resolved from 5.12 when you will be able to use 4 dlls ChemAxon .NET libraries and CalculationServices ..

Now for the time being the java code can used directly through these massive number of dlls.

 

Viktor

Thanks, Victor. I'll add all DLLs.

So how do I pass smiles to plugin.setMolecule? And then get smiles out of plugin?

Basically this is java code you may find in the java documentation as well:

Please try out this:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using chemaxon.marvin.calculations;
using chemaxon.struc;
using chemaxon.formats;

namespace TautomerPluginTest {
    class Program {
        static void Main(string[] args) {

            TautomerizationPlugin plugin = new TautomerizationPlugin();
            plugin.setProtectAllTetrahedralStereoCenters(true);
            plugin.setProtectAromaticity(true);
            plugin.setProtectDoubleBondStereo(true);
            plugin.setProtectEsterGroups(true);
            plugin.setProtectLabeledTetrahedralStereoCenters(true);
            plugin.setTakeCanonicalForm(true);
            Molecule m = MolImporter.importMol("C=C");
            plugin.setMolecule(m);
            var ret =  plugin.run();
            Molecule[] molecules = plugin.getResultMolecules();
        }
    }
}

Viktor

If it does nto work, I will forward this whole topic to the Calc team to review.

Viktor

Hmm.

plugin.getStructureCount() return 1, but plugin.getResultMolecules() returns null.

 

Here is my smiles

"CC(C)C[C@H](NC(=O)[C@@H](COC(C)(C)C)NC(=O)[C@H](CC1=CC=C(O)C=C1)NC(=O)[C@H](CO)NC(=O)[C@H](CC1=CNC2=CC=CC=C12)NC(=O)[C@H](CC1=CN=CN1)NC(=O)[C@@H]1CCC(=O)N1)C(=O)N[C@@H](CCCN=C(N)N)C(=O)N1CCC[C@H]1C(=O)NNC(N)=O"

Calcualtor team, please check this issue.

Thanks,

Viktor

Hi Karen,

the plugin.getResultMolecules() method is generally used to display plugin results on the GUI, and can't be used this way. You can use plugin.getStructure(int index) to get the index-th result structure. When you would like to generate the canonical tautomer (which is always one molecule), you can do it like:

Molecule canonicalForm = plugin.getStructure(0);

Rita

Hi,

 

It took me 8 hours to generate canonical tautomers for 1000 records. Some of them taking ~40 minutes. Is there any way to speed up the process? Here is my plugin settings

 

plugin = new TautomerizationPlugin();

                plugin.setProtectAllTetrahedralStereoCenters(true);

                plugin.setProtectAromaticity(true);

                plugin.setProtectDoubleBondStereo(true);

                plugin.setProtectEsterGroups(true);

                plugin.setProtectLabeledTetrahedralStereoCenters(true);

                plugin.setTakeCanonicalForm(true);

                plugin.setExcludeAntiAromaticCompounds(true);

                plugin.setMaximumTautomerizationPathLength(0);

                plugin.setMaxStructureCount(1000);

                plugin.setRingChainTautomerizationAllowed(true);

 

According to your experience what would be a reasonbale path length so I can cover majority of fused aromatic heterocycles?

Also how expensive is setRingChainTautomerizationAllowed and does it involve paths through CH2 and CH3?

We have 20+ millions of chemicals, ti is important for us to cover the majority of tautomeric space.

Hi,

 

You can increase the calculation speed of the canonic tautomer generation by setting of the "soft" mode of the tautomer  handling :

"plugin.setRationalTautomerGenerationMode(true);" Please insert this line into your code.

 

You can not change the optional length of the canonic tauotmer generation. So this line "plugin.setMaximumTautomerizationPathLength(0);" has no effect on the final result. 

 

The ring-chain tautomerization is not taken into account in the canonic tauotmer generation.

 

 

Jozsi