Hi John,
I took a look at the CDK source. Sure enough, the pattern generator just
uses the element symbol to pull the generic isotopes from the atom (see
lines 139-141 here
https://github.com/cdk/cdk/blob/ce36b8af886e08e6015d15a2724001193f71be76/tool/formula/src/main/java/org/openscience/cdk/formula/IsotopePatternGenerator.java#L135
).
That will always overwrite any custom isotopes you have.
On Tue, Mar 22, 2022 at 3:59 PM Rob Smith <2robsm...@gmail.com> wrote:
> I'm happy to. Just to clear up that I'm not doing something stupid, here's
> a gist. Let me know and I can still put it up on github.
>
> I've chopped my code down to a minimal example below. The formula
> generator search below will return hits that include 16C and 17C (though
> this snippet just calls getNextFormula once, so it returns just one). The
> masses of these hits will be in the correct range (626.75-626.9). But if
> you print the isotope pattern generated, you will get masses outside of
> that range, because they correspond to the same molecular formula (e.g.,
> O38 or O39) with the normal C isotopes (12C, 13C, etc), and therefore
> incorrect masses for each of the isotope pattern masses. In other words, if
> you were to create a vanilla O38 or O39 molecule and generate the isotope
> pattern (which, presumably uses the naturally-occuring ratio of 12C : 13C :
> 14C,
> 98.9 : 1.1 : 0.0001) you will get the same result as you do below when you
> generate the isotope patterns for nonstandard isotopes (here, O38 is 16C :
> 17C, 36:3). This suggests to me that the IsotopePatternGenerator does not
> actually use the isotope objects in the IMolecularFormula provided, but
> rather goes off the natural abundances of the atoms in the formula,
> irrespective of the isotopes set in the IMolecularFormula.
>
> public class FormulaGenerator {
> public static void main(String[] args){
> IChemObjectBuilder builder = SilentChemObjectBuilder.getInstance();
> MolecularFormulaRange mfRange = new MolecularFormulaRange();
>
> try{
> IsotopeFactory isotopeFactory = Isotopes.getInstance();
>
> IIsotope i = isotopeFactory.getIsotope("O", "15.99491462", 1.0);
> mfRange.addIsotope(i, isotope.min, isotope.max);
>
> i = isotopeFactory.getIsotope("O", "16.9991317", 1.0);
> mfRange.addIsotope(i, isotope.min, isotope.max);
> } catch (IOException e){
> System.exit(0);
> }
>
> MolecularFormulaGenerator generator = new
> MolecularFormulaGenerator(builder, 626.75, 626.9, mfRange);
> IMolecularFormulaSet formulaSet = new MolecularFormulaSet();
>
> IMolecularFormula formula = generator.getNextFormula();
>
> IsotopePatternGenerator isotope_pattern_generator = new
> IsotopePatternGenerator(0.01);
> IsotopePattern isotope_pattern =
> isotope_pattern_generator.getIsotopes(formula);
> for (IsotopeContainer isotope_container :
> isotope_pattern.getIsotopes()) {
> System.out.println("Isotope: mass:" + isotope_container.getMass() +
> " intensity:" + isotope_container.getIntensity());
> }
> }
> }
>
> On Tue, Mar 22, 2022 at 3:19 PM John Mayfield <john.wilkinson...@gmail.com>
> wrote:
>
>> Sounds like a possible bug, but not completely clear what you're trying
>> to do could you post an issue with example code on GitHub?
>>
>> Thanks,
>> John
>>
>> On Tue, 22 Mar 2022 at 21:06, Rob Smith <2robsm...@gmail.com> wrote:
>>
>>> Hello,
>>> I've built a cdk workflow to generate IMolecularFormulas from a mass.
>>> The workflow allows ranges of counts of different isotopes for different
>>> atoms. So far, so good.
>>>
>>> The next step of the workflow is to generate IsotopePatterns for each
>>> generated IMolecularFormula.
>>>
>>> I'm probably missing something, but I have spent quite some time writing
>>> different approaches and verifications of what is happening.
>>>
>>> I have confirmed that the IMolecularFormulas outputted from the formula
>>> generator step are producing non-natural isotope abundances as expected to
>>> fit the range criteria. However, the IsotopePatternGenerator seems to
>>> ignore these isotopes and print the pattern as if only natural abundances
>>> are being used.
>>>
>>> Does the IsotopePatternGenerator not support non-standard isotope
>>> abundances? Or am I missing a step here?
>>>
>>> I'd appreciate any insight anyone may have, as I've exhausted all code
>>> permutations I can think of.
>>>
>>> Thank you for your time.
>>> -Rob
>>> _______________________________________________
>>> Cdk-user mailing list
>>> Cdk-user@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/cdk-user
>>>
>>
_______________________________________________
Cdk-user mailing list
Cdk-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/cdk-user
