Re: [Rdkit-discuss] useQueryQueryMatches with recursive SMARTS

[Ivan Tubert-Brohman]

Hi Greg,

Thank you. After more thought I decided that query-to-query matching is not
the best way to go for my task, regardless of what it does with recursive
SMARTS, so don't lose sleep about it on my behalf. :-) Still, it's good to
know about this behavior for future reference.

Best,
Ivan


On Fri, Mar 9, 2018 at 12:13 AM, Greg Landrum 
wrote:

> Hi Ivan,
>
> On Wed, Mar 7, 2018 at 8:58 PM, Ivan Tubert-Brohman  schrodinger.com> wrote:
>
>>
>> Is it reasonable to expect that a SMARTS should match itself when
>> useQueryQueryMatches=True?
>>
>
> Not my favorite kind of answer to give, but... "it depends"
>
>
>>   query = Chem.MolFromSmarts('[C;!$(C=O)]Cl')
>>   query.HasSubstructMatch(query, useQueryQueryMatches=True)
>>
>> The above returns False. Without useQueryQueryMatches, it returns True,
>> but I think I need useQueryQueryMatches for my application because
>> queries are not generally valid molecules, and in some cases
>> HasSubstructMatch ends up raising exceptions (for example when the
>> SMARTS specifies hydrogen counts).
>>
>> If I remove the recursive part of the SMARTS above, I get a match, so I
>> suspect that the query-query search doesn't play well with recursive SMARTS.
>>
>
> That is the expected, though not particularly nice, behavior. The code is
> currently quite conservative about when it says that two queries match each
> other: if it sees a recursive query it just says "no".
> There may be a straightforward fix for this, but in general making the
> query-query matcher more flexible is one of those "it would be so cool, but
> it's s much work" projects, so I'm not going to make any promises.
>
> Best,
> -greg
>
>
>
>
>> Best,
>> Ivan
>>
>> 
>> --
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
>> ___
>> Rdkit-discuss mailing list
>> Rdkit-discuss@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>>
>>
>
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

Re: [Rdkit-discuss] useQueryQueryMatches with recursive SMARTS

[Greg Landrum]

Hi Ivan,

On Wed, Mar 7, 2018 at 8:58 PM, Ivan Tubert-Brohman <
ivan.tubert-broh...@schrodinger.com> wrote:

>
> Is it reasonable to expect that a SMARTS should match itself when
> useQueryQueryMatches=True?
>

Not my favorite kind of answer to give, but... "it depends"


>   query = Chem.MolFromSmarts('[C;!$(C=O)]Cl')
>   query.HasSubstructMatch(query, useQueryQueryMatches=True)
>
> The above returns False. Without useQueryQueryMatches, it returns True,
> but I think I need useQueryQueryMatches for my application because
> queries are not generally valid molecules, and in some cases
> HasSubstructMatch ends up raising exceptions (for example when the SMARTS
> specifies hydrogen counts).
>
> If I remove the recursive part of the SMARTS above, I get a match, so I
> suspect that the query-query search doesn't play well with recursive SMARTS.
>

That is the expected, though not particularly nice, behavior. The code is
currently quite conservative about when it says that two queries match each
other: if it sees a recursive query it just says "no".
There may be a straightforward fix for this, but in general making the
query-query matcher more flexible is one of those "it would be so cool, but
it's s much work" projects, so I'm not going to make any promises.

Best,
-greg




> Best,
> Ivan
>
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> Rdkit-discuss mailing list
> Rdkit-discuss@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/rdkit-discuss
>
>
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss

[Rdkit-discuss] useQueryQueryMatches with recursive SMARTS

[Ivan Tubert-Brohman]

Hi,

Is it reasonable to expect that a SMARTS should match itself when
useQueryQueryMatches=True?

  query = Chem.MolFromSmarts('[C;!$(C=O)]Cl')
  query.HasSubstructMatch(query, useQueryQueryMatches=True)

The above returns False. Without useQueryQueryMatches, it returns True, but
I think I need useQueryQueryMatches for my application because queries are
not generally valid molecules, and in some cases HasSubstructMatch ends up
raising exceptions (for example when the SMARTS specifies hydrogen counts).

If I remove the recursive part of the SMARTS above, I get a match, so I
suspect that the query-query search doesn't play well with recursive SMARTS.

Best,
Ivan
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot___
Rdkit-discuss mailing list
Rdkit-discuss@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rdkit-discuss