Difference API/command line tetrahedralstereoisomerscount

Hi,

I'm encountering the following problem:

When calculating the number of tetrahedralstereoisomers via the API, I don't have the same results than via the command line for some molecules (attached).

Here is the command line:

tetrahedralstereoisomercount -T true Problems.smi

And here is the part of the code where I'm invoking the API (via jython):

for molecule in molecules:
        counter+=1
        plugin = StereoisomerPlugin()
        plugin.setMolecule(molecule)
        plugin.setStereoisomerismType(StereoisomerPlugin.TETRAHEDRAL)
        plugin.setProtectDoubleBondStereo(1)
        plugin.setProtectTetrahedralStereo(1)
        plugin.run()
        stereoisomerCount = plugin.getStereoisomerCount()
        print "%s %s" %(counter, stereoisomerCount)

I think the command line gives better results (Problems_command_line_count.txt) than the API (Problems_API_count.txt).

Am I forgetting something when using the API, or is it a bug?

Florent.

P.S: I'm adding a note that the structures of 32, 34 and 35 should have only 1
stereoform (Is the "bridged" section of these molecules difficult
to compute ?)

Hi petronfl,

I moved the topic to the calculator plugins area. You will get a reply soon.

Hi,

I get the same results whether I use cxcalc or the Java API (see the attached Java code and the output). I don't know why jython API returns different results.

Does the attached Java code return the correct results for you?

Zsolt

I tried the java version, same wrong results (8, 8, 16 etc).

I think I must have deprecated Jar files or something.

I'll take a look at that.

Thanks for the help, dzatonyi & Zsolt !

Edit : That was it (I was using 5.8 jars). The problem persists for the 2.6.0.v0133 ChemAxon/Infocom Marvin Extensions Feature (KNIME) though. (That's where I was having the problem first, and I checked using the API)