Strange metabolizer output when used as command line tool

User 25aae95547

14-01-2015 08:51:01

Dear all,


i am using the metabolizer command line tool within KNIME to predict transformation products (TPs) for several hundred input molecules over night. However, this morning i realized that an error occured during the execution, because one of the generated output files containing the smiles of the TPs differed from the usual structure.


 
































































































































































































































#SMILES name Synthesis Code Generation Phase Formation Degradation Production Accumulation Route Exact Mass
OC1CC2=CC(OCCCCC3=NN=NN3C3CCCCC3)=CC=C2NC1=O
1 0 0 0 63270 100.00% < 0.001%
385.2113897
O=C1NC2=CC=C(OCCCCC3=NN=NN3C3CCCCC3)C=C2C=C1
Dehydration(1):1 1 0 16807 190121 26.56% < 0.001% Dehydration 367.2008251
O=C1CCCCC1
Aromatic N-Dealkylation(1):1/2 1 0 2401 134458 3.80% < 0.001% Aromatic N-Dealkylation 98.07316494
OC(CCCOC1=CC=C2NC(=O)C(O)CC2=C1)C1=NN=NN1C1CCCCC1
BenzylicHydroxylation(1):1 1 0 2401 176508 3.80% < 0.001% $BenzylicHydroxylation 401.2063044 TRUE
OC1C(O)C2=CC(OCCCCC3=NN=NN3C3CCCCC3)=CC=C2NC1=O
BenzylicHydroxylation(1):2 1 0 2401 293472 3.80% < 0.001% $BenzylicHydroxylation 401.2063044 TRUE
O=C1CC2=CC(OCCCCC3=NN=NN3C3CCCCC3)=CC=C2NC1=O
SecondaryAlcoholOxidation(1):1 1 0 2401 173414 3.80% < 0.001% SecondaryAlcoholOxidation 383.1957397



As you can see the 4th and 5th prediction contain an additional column without header at the end which contains TRUE as value. Furthermore the transformation route is given as $BenzylicHydroxylation, while benzylic hydroxylations of other input molecules did not contain the additional $-sign in the route description and also not the additional column. Unfortunately, this is messing up my workflow as i cannot define a fixed number of input cells when using the R command read.table to read in the generated output files. I think i could come up with a work-around but i would prefer to understand why this happens for certain molecules and not for all and what the TRUE value in the last column actually represents.


You should be able to reproduce the output when typing this command into the command line: metabolize OC1CC2=CC(OCCCCC3=NN=NN3C3CCCCC3)=CC=C2NC1=O and in case it matters i have the version 15.1.5 of  Metabolizer installed on my computer.


Best regards,


MG









ChemAxon d76e6e95eb

15-01-2015 13:59:58

Regarding the dollar sign in front of a transformation name:
Some molecules have multiple sites for a certain transformation type, such as benzylic hydroxylation. In this case, some of the sites may be preferred to others. This is indicated by the dollar sign. So in your case, the transformations marked with the $ sign are unpreferred sites, the benzylic hydroxlation is more likely on another position.


The trailing true is a printout issue, please ignore it.