Heavy atom count

Is it possible to do a heavy atom count using cxcalc?

I see atomcount, counts of aliphatic or aromatic atoms but not heavy atoms.

No it isn't but it can be done with Evaluator.

Example:

Subtract the Hydrogen atom count from atom count:

$ evaluate -e "atomCount() - atomCount('1')" "NCCCO"
5

For more see https://www.chemaxon.com/marvin/help/chemicalterms/EvaluatorFunctions.html#atomcount

Zsolt

Many thanks for this,

That will do nicely.

It might still be worth adding HAC to the cxcalc options.

At the moment I'm using

cxcalc input_file  -o  output_file logp logd -H 7.4 pka -b 1 -a 1 mass acceptorcount donorcount polarsurfacearea rotatablebondcount

To give a results.tab containing the data, what would be the best way to add the HAC to the file?

If it's OK to get a semicolon separated list as output instead of a tab separated list, then Chemical Terms Evaluator will do the job:

evaluate nci10.smiles -e "logp(); logd('7.4'); apka('1'); bpka('1'); mass(); acceptorcount(); donorcount(); psa(); rotatablebondcount(); atomCount()-atomCount('1')"
1.42;1.42;;-7.61;122.12;2;0;34.14;0;9
6.22;6.22;;0.77;332.49;2;0;25.78;3;20
2.15;0.28;2.72;-8.56;218.55;5;1;111.87;2;14
0.71;0.61;7.99;5.83;145.14;4;2;81.7;1;9
2.09;2.09;;2.04;223.23;3;1;60.16;0;17
4.61;0.13;3.61;-5.43;490.1;5;2;83.83;1;27
1.54;1.54;;-0.73;235.67;3;0;37.38;1;16
3.37;3.37;;-7.42;267.24;4;0;79.96;1;20
0.41;0.29;7.89;1.11;116.12;4;2;65.18;1;8
5.11;5.11;;-8.08;262.29;0;0;0;3;19

cxcalc cannot perform mathematical operations on calculations, so it cannot subtract hydrogen count from atom count.

I hope this helps,

Zsolt

Many thanks

Chris

Hi,

evaluate myfile.sdf  -e name(); logp(); logd('7.4'); apka('1'); bpka('1'); atomCount(); mass(); acceptorcount(); donorcount(); psa(); rotatablebondcount(); atomCount()-atomCount('1') -o output.txt

Running this on a file of 30,000 structures I get this error

error "/Applications/ChemAxon/MarvinBeans/bin/evaluate: line 60: [: too many arguments

/Applications/ChemAxon/MarvinBeans/bin/evaluate: line 63: [: too many arguments

Exception in thread \"main\" chemaxon.nfunk.jep.ParseException: Error while evaluating expression:

name(); logp(); logd('7.4'); apka('1'); bpka('1'); atomCount(); mass(); acceptorcount(); donorcount(); psa(); rotatablebondcount(); atomCount()-atomCount('1')

    chemaxon.marvin.io.MolExportException: Name generation failed: java.lang.ArrayIndexOutOfBoundsException: 18

at chemaxon.nfunk.jep.JEP.getValueAsObject(Unknown Source)

at chemaxon.jep.ChemJEP.evaluate(Unknown Source)

at chemaxon.jep.Evaluator.main(Unknown Source)

Caused by:

chemaxon.nfunk.jep.ParseException: chemaxon.marvin.io.MolExportException: Name generation failed: java.lang.ArrayIndexOutOfBoundsException: 18

at chemaxon.jep.function.Plugin.run(Unknown Source)

at chemaxon.nfunk.jep.EvaluatorVisitor.visit(Unknown Source)

at chemaxon.nfunk.jep.ASTFunNode.jjtAccept(Unknown Source)

at chemaxon.nfunk.jep.EvaluatorVisitor.getValue(Unknown Source)

at chemaxon.nfunk.jep.JEP.getValueAsObject(Unknown Source)

at chemaxon.jep.ChemJEP.evaluate(Unknown Source)

at chemaxon.jep.Evaluator.main(Unknown Source)

Caused by:

chemaxon.marvin.plugin.PluginException: chemaxon.marvin.io.MolExportException: Name generation failed: java.lang.ArrayIndexOutOfBoundsException: 18

at chemaxon.marvin.calculations.IUPACNamingPlugin.getPreferredIUPACName(Unknown Source)

at chemaxon.marvin.calculations.IUPACNamingPlugin.run(Unknown Source)

at chemaxon.jep.function.Plugin.run(Unknown Source)

at chemaxon.nfunk.jep.EvaluatorVisitor.visit(Unknown Source)

at chemaxon.nfunk.jep.ASTFunNode.jjtAccept(Unknown Source)

at chemaxon.nfunk.jep.EvaluatorVisitor.getValue(Unknown Source)

at chemaxon.nfunk.jep.JEP.getValueAsObject(Unknown Source)

at chemaxon.jep.ChemJEP.evaluate(Unknown Source)

at chemaxon.jep.Evaluator.main(Unknown Source)

Caused by: chemaxon.marvin.io.MolExportException: Name generation failed: java.lang.ArrayIndexOutOfBoundsException: 18

at chemaxon.marvin.io.formats.name.NameExport.convert(Unknown Source)

at chemaxon.marvin.calculations.IUPACNamingPlugin.getName(Unknown Source)

... 9 more

Caused by: chemaxon.marvin.io.formats.name.nameexport.IUPACNamer$Error: Name generation failed: java.lang.ArrayIndexOutOfBoundsException: 18

at chemaxon.marvin.io.formats.name.nameexport.NamingCentral.getName(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.IUPACNamerThread.run(Unknown Source)

Caused by: java.lang.ArrayIndexOutOfBoundsException: 18

at chemaxon.marvin.io.formats.name.nameexport.GeneralFusedRingSystem.recognize(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.GeneralSpiroNamer.recognize(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.GeneralSpiroNamer.nameRingSystem(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.UnbranchedSpiroNamer.getAllNames(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.UnbranchedSpiroNamer.computePartNames(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.GeneralSpiroNamer.getPartNames(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.GeneralSpiroNamer.getModifiers(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.Part.suffixCount(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.Part.getAnionCount(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.Part.getSeniorityClass(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.ParentFinder.compare(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.ParentFinder.computeParent(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.ParentFinder.parent(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.TopologyAnalyser.getNextFragment(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.TopologyAnalyser.analyse(Unknown Source)

at chemaxon.marvin.io.formats.name.nameexport.NamingCentral.analyse(Unknown Source)

Is this a Java memory issue?

If I remove the name() function it runs OK

IUPAC name generator has some limitations, for example it cannot generate names for complex fused structures. In this case the "name()" CT function throws an exception. To ignore the error, and continue the processing with next structure use the

-g, --ignore-error                  continue with next molecule on error

command line option.

Examples:

evaluate -g -e "logP(); name()" mols.smiles

evaluate --ignore-error -e "logP(); name()" mols.smiles 2>errormessages.txt

The "2>errormessages.txt" works only on unix systems, and sends the error messages to file errormessages.txt.

Zsolt

Many thanks for the rapid response :-)

I did wonder if that was the issue.

The only reason I'm using it is to provide an easy way for me to check the results for a particular record, is there any way to include the compound ID from the sdf file, or maybe a SMILES representation?

Thanks

Chris

Yes, there is.

Example:

$ evaluate -e "field('id'); molString('smiles'); name()" demo10.sdf
1;CC1=CC(=O)C=CC1=O;2-methylcyclohexa-2,5-diene-1,4-dione
2;S(SC1=NC2=C(S1)C=CC=C2)C1=NC2=CC=CC=C2S1;2-(1,3-benzothiazol-2-yldisulfanyl)-1,3-benzothiazole
3;OC1=C(Cl)C=C(C=C1[N+]([O-])=O)[N+]([O-])=O;2-chloro-4,6-dinitrophenol
4;[O-][N+](=O)C1=CNC(=N)S1;5-nitro-2,3-dihydro-1,3-thiazol-2-imine
5;NC1=CC2=C(C=C1)C(=O)C1=C(C=CC=C1)C2=O;2-amino-9,10-dihydroanthracene-9,10-dione
6;OC(=O)C1=C(C=CC=C1)C1=C2C=CC(=O)C(Br)=C2OC2=C1C=CC(O)=C2Br;2-(4,5-dibromo-6-hydroxy-3-oxo-3H-xanthen-9-yl)benzoic acid
7;CN(C)C1=C(Cl)C(=O)C2=C(C=CC=C2)C1=O;2-chloro-3-(dimethylamino)-1,4-dihydronaphthalene-1,4-dione
8;CC1=C(C2=C(C=C1)C(=O)C1=CC=CC=C1C2=O)[N+]([O-])=O;2-methyl-1-nitro-9,10-dihydroanthracene-9,10-dione
9;CC(=NO)C(C)=NO;N-[3-(hydroxyimino)butan-2-ylidene]hydroxylamine
10;C1=CC=C(C=C1)P(C1=CC=CC=C1)C1=CC=CC=C1;triphenylphosphane

In the example 1st column contains the content of the "id" SDfile field, 2nd column contains the SMILES representation of the molecule, and 3rd column contains the IUPAC name. The input file is attached.

Zsolt

Many thanks

$ evaluate -e "field('id'); molString('smiles'); name()" demo10.sdf
1;CC1=CC(=O)C=CC1=O;2-methylcyclohexa-2,5-diene-1,4-dione
2;S(SC1=NC2=C(S1)C=CC=C2)C1=NC2=CC=CC=C2S1;2-(1,3-benzothiazol-2-yldisulfanyl)-1,3-benzothiazole
3;OC1=C(Cl)C=C(C=C1[N+]([O-])=O)[N+]([O-])=O;2-chloro-4,6-dinitrophenol
4;[O-][N+](=O)C1=CNC(=N)S1;5-nitro-2,3-dihydro-1,3-thiazol-2-imine
5;NC1=CC2=C(C=C1)C(=O)C1=C(C=CC=C1)C2=O;2-amino-9,10-dihydroanthracene-9,10-dione
6;OC(=O)C1=C(C=CC=C1)C1=C2C=CC(=O)C(Br)=C2OC2=C1C=CC(O)=C2Br;2-(4,5-dibromo-6-hydroxy-3-oxo-3H-xanthen-9-yl)benzoic acid
7;CN(C)C1=C(Cl)C(=O)C2=C(C=CC=C2)C1=O;2-chloro-3-(dimethylamino)-1,4-dihydronaphthalene-1,4-dione
8;CC1=C(C2=C(C=C1)C(=O)C1=CC=CC=C1C2=O)[N+]([O-])=O;2-methyl-1-nitro-9,10-dihydroanthracene-9,10-dione
9;CC(=NO)C(C)=NO;N-[3-(hydroxyimino)butan-2-ylidene]hydroxylamine
10;C1=CC=C(C=C1)P(C1=CC=CC=C1)C1=CC=CC=C1;triphenylphosphane

If the 'ID' is "10E-593" it seems to be treated as a number?

It should be treated as string. Could you give an example?

Zsolt

Chris, you are right, if the content of the ID field can be read as a number, then it's converted to a number.

We will provide a method to read the content of a field as string (around Marvin 5.4.1).

Zsolt

Thanks for confirming it so quickly.

I'm trying to find definitions of all the terms. If I use 

evaluate -l

It provides a link to http://www.chemaxon.com/marvin/help/chemicalterms/EvaluatorTables.html which seems not to exist?

Dear Chris, 

we have fixed this issue. The correct link pointing to the function tables is :

https://docs.chemaxon.com/display/chemicalterms/Functions+by+Categories

The newer versions of our product will contain the fix. 

Best wishes, 

Daniel

Hi,

Thanks for that, very useful.

I have an historical script that calls psa() I'm guessing this is now topologicalPolarSurfaceArea()?

psa() works too. Some functions have short and long names, in this case psa() is the short and topologicalPolarSurfaceArea() is the long.

Thanks

Chris

I was trying to calculate the basic pka of 

1-Benzyl-4-methylpiperazine

http://www.chemicalize.org/structure/#!mol=CN1CCN%28CC1%29Cc2ccccc2&source=fp

Shouldn't the nitrogen bearing the methyl group be the most basic?

Hi,

Yes, the "N-methyl" nitrogen should be a little more basic than the "N-benzyl" nitrogen. 

The micro ionization constant (pKa=8.39) of the "N-benzyl" nitrogen is  a little over-predicted as it is shown on the attached figure. It should be below of the micro pKa = 8.14 of the "N-methyl" group. 

The two macro pKa values 8.59 and 2.9 are also shown on the figure.

Thanks for reporting this bug ,it will be fixed asap.

Jozsi