ambit-users Mailing List for AMBIT:Chemical Structure DB&Web Service

Hello Nina,
Hello Nick,

I will do so. I was afraid I would have to write two different SMIRKS as a
solution, as this might significantly increase the total number of SMIRKS
in my system.
Again, thanks a lot.

Best regards,

Yannick

On Tue, Sep 13, 2016 at 2:09 AM, Nikolay Kochev <ni...@un...>
wrote:

> Dear Yannick
> If you need to transform/remove/add H atoms in Ambit-SMART it can be done
> only treating them as explicit H atoms. The SMIRKS must define explicitly
> what happens with each H atom.
>
> The indicated implicit H atoms are used only for defining logical
> expression in order to match the substructure (target fragment) that is to
> be transformed according to the SMIRKS.
>
> Here I suggest to you a working solution based on explicit H atoms:
>
> (1) First you need two different SMIRKS for both cases since the H atom
> configuration is different (It is impossible with explicit H managing to
> handle both cases with a single SMIRKS):
>
> (2) Secondly your target molecule must be set with explicit H atoms
>
> (3) here are the SMIRKS for both cases:
>
> [#6;A;H2X4:7]([H:99])([H])!@-[#6;A;H2X4:6]!@-[#8;X2:5][P;X4:
> 4]([#8;X2:2])([#8;A;X2:3][H])=[O;X1:1]>>[#8;X2:2][P;X4:4]1(=
> [O;X1:1])[#8;A;X:3][#6;A;H1X4:7]([H:99])!@-[#6;A;H2X4:6]!@-[#8;X2:5]1
>
> [#6;A;H2X4:7]([H:99])([H])!@-[#6;A;H2X4:6]!@-[#8;X2:5][P;X4:
> 4]([#8;X2:2])([#8;A;X1-:3])=[O;X1:1]>>[#8;X2:2][P;X4:4]1(=[
> O;X1:1])[#8;A;X:3][#6;A;H1X4:7]([H:99])!@-[#6;A;H2X4:6]!@-[#8;X2:5]1
>
> Notice that also the explicit H atoms are treated on atom 7
> [#6;A;H2X4:7]([H:99])([H])  because when you close the ring one of the H
> atoms must be removed (i.e. this is the unmapped H atom) the other H atom
> stays (i.e. this is the mapped H atom: [H:99])
>
>
> With best regards
> Nick
>
>
>
>
>
> On 9/12/16 22:29, Yannick .Djoumbou wrote:
>
>
>
> On Sun, Jul 10, 2016 at 7:57 PM, Yannick .Djoumbou <y.d...@gm...>
> wrote:
>
>> Hi all,
>>
>> I have some issues with my SMIRKS manager.
>> have written the SMIKRS below that I am applying on one compounds:
>>
>> [#6;A;H2X4:7]!@-[#6;A;H2X4:6]!@-[#8;X2:5][P;X4:4]([#8;X2:2])
>> ([#8;A;X2H1,X1-:3])=[O;X1:1]>>[#8;X2:2][P;X4:4]1(=[O;X1:1])[
>> #8;A;X:3][#6;A;H1X4:7]!@-[#6;A;H2X4:6]!@-[#8;X2:5]1
>>
>>
>> I am trying to join two atoms here to form a cycle, The C7 and the O3.  In
>>  the reactant, O3 can be linked to a Hydrogen atom (case 1)  or can be
>> negatively charged (case 2). The problem is that in the product, the O3 has
>> a valence of 3 (with an implicit H) (case 1), or a valence of 2 (case 2).
>>
>> The examples are illustrated in the attached file.
>>
>>
>> Could anyone help?
>>
>>
>> Thanks.
>>
>>
>> Regards,
>>
>
>
>

Dear Yannick

If you need to transform/remove/add H atoms in Ambit-SMART it can be 
done only treating them as explicit H atoms. The SMIRKS must define 
explicitly what happens with each H atom.

The indicated implicit H atoms are used only for defining logical 
expression in order to match the substructure (target fragment) that is 
to be transformed according to the SMIRKS.

Here I suggest to you a working solution based on explicit H atoms:

(1) First you need two different SMIRKS for both cases since the H atom 
configuration is different (It is impossible with explicit H managing to 
handle both cases with a single SMIRKS):

(2) Secondly your target molecule must be set with explicit H atoms

(3) here are the SMIRKS for both cases:

[#6;A;H2X4:7]([H:99])([H])!@-[#6;A;H2X4:6]!@-[#8;X2:5][P;X4:4]([#8;X2:2])([#8;A;X2:3][H])=[O;X1:1]>>[#8;X2:2][P;X4:4]1(=[O;X1:1])[#8;A;X:3][#6;A;H1X4:7]([H:99])!@-[#6;A;H2X4:6]!@-[#8;X2:5]1

[#6;A;H2X4:7]([H:99])([H])!@-[#6;A;H2X4:6]!@-[#8;X2:5][P;X4:4]([#8;X2:2])([#8;A;X1-:3])=[O;X1:1]>>[#8;X2:2][P;X4:4]1(=[O;X1:1])[#8;A;X:3][#6;A;H1X4:7]([H:99])!@-[#6;A;H2X4:6]!@-[#8;X2:5]1

Notice that also the explicit H atoms are treated on atom 7 
[#6;A;H2X4:7]([H:99])([H])  because when you close the ring one of the H 
atoms must be removed (i.e. this is the unmapped H atom) the other H 
atom stays (i.e. this is the mapped H atom: [H:99])


With best regards
Nick




On 9/12/16 22:29, Yannick .Djoumbou wrote:
>
>
> On Sun, Jul 10, 2016 at 7:57 PM, Yannick .Djoumbou 
> <y.d...@gm... <mailto:y.d...@gm...>> wrote:
>
>     Hi all,
>
>     I have some issues with my SMIRKS manager.
>     have written the SMIKRS below that I am applying on one compounds:
>
>     [#6;A;H2X4:7]!@-[#6;A;H2X4:6]!@-[#8;X2:5][P;X4:4]([#8;X2:2])([#8;A;X2H1,X1-:3])=[O;X1:1]>>[#8;X2:2][P;X4:4]1(=[O;X1:1])[#8;A;X:3][#6;A;H1X4:7]!@-[#6;A;H2X4:6]!@-[#8;X2:5]1
>
>
>     I am trying to join two atoms here to form a cycle, The C7 and the
>     O3. In  the reactant,O3 can be linked to a Hydrogen atom (case 1)
>      or can be negatively charged (case 2). The problem is that in the
>     product, the O3 has a valence of 3 (with an implicit H) (case 1),
>     or a valence of 2 (case 2).
>
>     The examples are illustrated in the attached file.
>
>
>     Could anyone help?
>
>
>     Thanks.
>
>
>     Regards,
>
>

Yannick,

Could you please file an issue https://sourceforge.net/p/ambit/bugs/ if not
already.

SMIRKS processing has number of updates and cleanups since this spring, do
try to use the version in trunk 3.0.3-SNAPSHOT.

Best regards,
Nina

On 12 September 2016 at 22:29, Yannick .Djoumbou <y.d...@gm...>
wrote:

>
>
> On Sun, Jul 10, 2016 at 7:57 PM, Yannick .Djoumbou <y.d...@gm...>
> wrote:
>
>> Hi all,
>>
>> I have some issues with my SMIRKS manager.
>> have written the SMIKRS below that I am applying on one compounds:
>>
>> [#6;A;H2X4:7]!@-[#6;A;H2X4:6]!@-[#8;X2:5][P;X4:4]([#8;X2:2])
>> ([#8;A;X2H1,X1-:3])=[O;X1:1]>>[#8;X2:2][P;X4:4]1(=[O;X1:1])[
>> #8;A;X:3][#6;A;H1X4:7]!@-[#6;A;H2X4:6]!@-[#8;X2:5]1
>>
>>
>> I am trying to join two atoms here to form a cycle, The C7 and the O3.  In
>>  the reactant, O3 can be linked to a Hydrogen atom (case 1)  or can be
>> negatively charged (case 2). The problem is that in the product, the O3 has
>> a valence of 3 (with an implicit H) (case 1), or a valence of 2 (case 2).
>>
>> The examples are illustrated in the attached file.
>>
>>
>> Could anyone help?
>>
>>
>> Thanks.
>>
>>
>> Regards,
>>
>
>

On Sun, Jul 10, 2016 at 7:57 PM, Yannick .Djoumbou <y.d...@gm...>
wrote:

> Hi all,
>
> I have some issues with my SMIRKS manager.
> have written the SMIKRS below that I am applying on one compounds:
>
> [#6;A;H2X4:7]!@-[#6;A;H2X4:6]!@-[#8;X2:5][P;X4:4]([#8;X2:2])
> ([#8;A;X2H1,X1-:3])=[O;X1:1]>>[#8;X2:2][P;X4:4]1(=[O;X1:1])[
> #8;A;X:3][#6;A;H1X4:7]!@-[#6;A;H2X4:6]!@-[#8;X2:5]1
>
>
> I am trying to join two atoms here to form a cycle, The C7 and the O3.  In
>  the reactant, O3 can be linked to a Hydrogen atom (case 1)  or can be
> negatively charged (case 2). The problem is that in the product, the O3 has
> a valence of 3 (with an implicit H) (case 1), or a valence of 2 (case 2).
>
> The examples are illustrated in the attached file.
>
>
> Could anyone help?
>
>
> Thanks.
>
>
> Regards,
>