property calculation with different molecular formats

User 8feadbc205

21-04-2014 06:31:03

Hello,


 


I just have some questions about chemaxon calculating molecular properties of the same molecule at different formats, and seems like they give different results. Take ATP for example. I downloaded the mole file directly from KEGG database and calculate its water accessible surface area (ASA)


It gives me this


id   Atom count    Formal charge     ASA       ASA+      ASA-      ASA_H     ASA_P       (mol file)


1       47                         0                 528.45     271.24    257.21      213.18     315.27


 


When I converted the mole file to smiles string and inchi strings and calculated the same properties again. I got this:


 


NC1=NC=NC2=C1N=CN2[C@@H]1O[C@H](COP(O)(=O)OP(O)(=O)OP(O)(O)=O)[C@@H](O)[C@H]1O (smiles string chemaxon)


id Atom count     Formal charge   ASA      ASA+      ASA-     ASA_H    ASA_P


1          47                       0                536.16    274.93   261.23   217.84      318.32


 


InChI=1S/C10H16N5O13P3/c11-8-5-9(13-2-12-8)15(3-14-5)10-7(17)6(16)4(26-10)1-25-30(21,22)28-31(23,24)27-29(18,19)20/h2-4,6-7,10,16-17H,1H2,(H,21,22)(H,23,24)(H2,11,12,13)(H2,18,19,20)/t4-,6-,7-,10-/m1/s1  (inchi string chemaxon)


id      Atom count Formal charge       ASA    ASA+   ASA-   ASA_H   ASA_P


1               47         0     611.34   334.02   277.32   229.53   381.81


 


I also have another chemistry toolbox called open babel and they have different way of writing smiles string and inchi string than chemaxon, the result is again quite different.


 


C([C@@H]1[C@H]([C@H]([C@H](n2cnc3c(N)ncnc23)O1)O)O)OP(=O)(O)OP(=O)(O)OP(=O)(O)O    (smiles string open babel)


id Atom count   Formal charge    ASA    ASA+   ASA-   ASA_H   ASA_P


1   47   0   530.93   273.18   257.75   214.46   316.47


 


InChI=1/C10H16N5O13P3/c11-8-5-9(13-2-12-8)15(3-14-5)10-7(17)6(16)4(26-10)1-25-30(21,22)28-31(23,24)27-29(18,19)20/h2-4,6-7,10,16-17H,1H2,(H,21,22)(H,23,24)(H2,11,12,13)(H2,18,19,20)/t4-,6-,7-,10-/m1/s1/f/h18-19,21,23H,11H2   (inchi string open babel)


id   Atom count   Formal charge   ASA   ASA+   ASA-   ASA_H   ASA_P


1   47   0   611.34   334.02   277.32   229.53   381.81


 


The difference is actually quite significant. Could you give me some explanations for such discrepancy and any possible work around for this? Is the way of writing smiles and inchi strings correct in both chemaxon and open babel? 


 


 


Thank you for your time and help!


 


Bin

User 851ac690a0

23-04-2014 14:56:17

Hi,


 


It seems that the difference is caused by  the discrepancy of the generated 3D conformers which are taken into account in  the ASA calculation.  


 This bug will be fixed  in the subsequent release (6.4)


At his moment I  would propose you to catch the 3D structure from an external source like Corina for example  OR try to use our conformer generator to obtain  a low energy  3D structure and input that for the ASA calculation.


 


Thanks for reporting this bug.


 


Jozsi