Re: Release candidates available for testing

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Release candidates available for testing

Douglas Houston
Hi,

According to this page:

http://baoilleach.webfactional.com/site_media/ob-docs/Features/Fingerprints.html#fingerprints

I can use the -s option with the name of a file, e.g.:

babel mymols.sdf -sSubstructure.sdf mymols_filtered.sdf

will write out all the compounds in mymols.sdf that contain the group  
in Substructure.sdf. However, what I'd like is the opposite, i.e. to  
write out everything that doesn't match. I tried using -v instead of  
-s but this didn't work.

It would also be nice if I could use a file that contained more than  
one entry, i.e. the file Substructure.sdf could define multiple  
fragments that I don't want  in my molecules, rather than having to  
run Babel on each one separately.

Would this be doable before the final release?

cheers,

Doug




> ----- Forwarded message from [hidden email] -----
>     Date: Sat, 21 Aug 2010 20:03:08 +0100
>     From: Noel O'Boyle <[hidden email]>
> Reply-To: Noel O'Boyle <[hidden email]>
>  Subject: [Open Babel] Release candidates available for testing
>       To: [hidden email]
>
> Hello all,
>
> The release of Open Babel 2.3.0 is coming up soon. There have been
> many changes so it would be great to get some feedback on whether the
> current release candidates are working.
>
> These are available at
> http://sourceforge.net/projects/openbabel/files/openbabel-snapshots/2010-08-03/openbabel-2.3.0b2-20100803-r3942.tar.gz/download
> for Linux and
> http://sourceforge.net/projects/openbabel/files/openbabel-snapshots/2010-08-18/OpenBabel2.3.0rc2_Windows_Installer.exe/download
> for Windows.
>
> I'm responsible for the new compiling documentation, currently
> available at
> http://baoilleach.webfactional.com/site_media/ob-docs/Installation/install.html#compiling-open-babel.
> If there are any problems with these, please let us know also.
>
> Regards,
>    Noel
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> OpenBabel-discuss mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>



_____________________________________________________
Dr. Douglas R. Houston
Scientific Programme Coordinator / Research Associate
Room 3.23
Institute of Structural and Molecular Biology
Michael Swann Building
King's Buildings
University of Edinburgh
Edinburgh, EH9 3JR, UK
Tel. 0131 650 7358

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.



------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:

Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
Reply | Threaded
Open this post in threaded view
|

Re: Release candidates available for testing

Chris Morley-3
On 08/09/2010 13:17, Douglas Houston wrote:

> Hi,
>
> According to this page:
>
> http://baoilleach.webfactional.com/site_media/ob-docs/Features/Fingerprints.html#fingerprints
>
> I can use the -s option with the name of a file, e.g.:
>
> babel mymols.sdf -sSubstructure.sdf mymols_filtered.sdf
>
> will write out all the compounds in mymols.sdf that contain the group
> in Substructure.sdf. However, what I'd like is the opposite, i.e. to
> write out everything that doesn't match. I tried using -v instead of
> -s but this didn't work.

I think the following will do what you want
babel mymols.sdf -s ~Substructure.sdf mymols_filtered.sdf
>
> It would also be nice if I could use a file that contained more than
> one entry, i.e. the file Substructure.sdf could define multiple
> fragments that I don't want  in my molecules, rather than having to
> run Babel on each one separately.

It would have to be decided whether the combination was AND or OR.
What you want is NOT(frag1 OR frag2 OR...)
In the related --filter option there is an implicit AND between tests,
but I think for both the -s filter and its inverse the OR behaviour is
more intuitive.
>
> Would this be doable before the final release?

It may be, but I am about to go on holiday, so don't hold your breath.

Thanks for this suggestion. Other suggestions for useful features are
very welcome.

You could look at the Sieve program, http://www.silicos.be/sieve.html 
which is an industrial-strength filtering program, now using OpenBabel.

Chris

------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:

Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
Reply | Threaded
Open this post in threaded view
|

Re: Release candidates available for testing

Douglas Houston
Thanks Chris,

The line you gave seemed to do the trick - after I escaped out the  
tilde. Is the usage of the tilde to reverse the sense of the filter  
documented? I would have thought the -v option would be logical.

Sieve seems to do what I want - almost. It looks like it filters using  
SMARTS - but all my undesirable groups are in SDF format. How do I  
convert to SMARTS? OpenBabel doesn't seem to do it (although it can  
filter using them, strangely).

I realise I could potentially use SMILES as input but this doesn't  
work in some cases, e.g. 'C1=CC=CC=C1' is a benzene ring in SMILES  
format but according to the following link will not match benzene if  
it's used as a SMARTS string.

http://www.daylight.com/dayhtml/doc/theory/theory.smarts.html

cheers,

Doug



Quoting Chris Morley <[hidden email]>:

> On 08/09/2010 13:17, Douglas Houston wrote:
>> Hi,
>>
>> According to this page:
>>
>> http://baoilleach.webfactional.com/site_media/ob-docs/Features/Fingerprints.html#fingerprints
>>
>> I can use the -s option with the name of a file, e.g.:
>>
>> babel mymols.sdf -sSubstructure.sdf mymols_filtered.sdf
>>
>> will write out all the compounds in mymols.sdf that contain the group
>> in Substructure.sdf. However, what I'd like is the opposite, i.e. to
>> write out everything that doesn't match. I tried using -v instead of
>> -s but this didn't work.
>
> I think the following will do what you want
> babel mymols.sdf -s ~Substructure.sdf mymols_filtered.sdf
>>
>> It would also be nice if I could use a file that contained more than
>> one entry, i.e. the file Substructure.sdf could define multiple
>> fragments that I don't want  in my molecules, rather than having to
>> run Babel on each one separately.
>
> It would have to be decided whether the combination was AND or OR.
> What you want is NOT(frag1 OR frag2 OR...)
> In the related --filter option there is an implicit AND between tests,
> but I think for both the -s filter and its inverse the OR behaviour is
> more intuitive.
>>
>> Would this be doable before the final release?
>
> It may be, but I am about to go on holiday, so don't hold your breath.
>
> Thanks for this suggestion. Other suggestions for useful features are
> very welcome.
>
> You could look at the Sieve program, http://www.silicos.be/sieve.html
> which is an industrial-strength filtering program, now using OpenBabel.
>
> Chris
>
> ------------------------------------------------------------------------------
> This SF.net Dev2Dev email is sponsored by:
>
> Show off your parallel programming skills.
> Enter the Intel(R) Threading Challenge 2010.
> http://p.sf.net/sfu/intel-thread-sfd
> _______________________________________________
> OpenBabel-discuss mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>
>




_____________________________________________________
Dr. Douglas R. Houston
Scientific Programme Coordinator / Research Associate
Room 3.23
Institute of Structural and Molecular Biology
Michael Swann Building
King's Buildings
University of Edinburgh
Edinburgh, EH9 3JR, UK
Tel. 0131 650 7358

--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.



------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:

Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
Reply | Threaded
Open this post in threaded view
|

Re: Release candidates available for testing

Craig James-2
On 9/8/10 9:32 AM, Douglas Houston wrote:

> I realise I could potentially use SMILES as input but this doesn't
> work in some cases, e.g. 'C1=CC=CC=C1' is a benzene ring in SMILES
> format but according to the following link will not match benzene if
> it's used as a SMARTS string.
>
> http://www.daylight.com/dayhtml/doc/theory/theory.smarts.html

If you canonicalize a SMILES (babel ... -o can), it usually works correctly as a SMARTS.  That's one of the features of the canonical algorithm.  In the case you cite, it would convert to "c1ccccc1", which is the SMARTS that you want.

Craig

------------------------------------------------------------------------------
This SF.net Dev2Dev email is sponsored by:

Show off your parallel programming skills.
Enter the Intel(R) Threading Challenge 2010.
http://p.sf.net/sfu/intel-thread-sfd
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
Reply | Threaded
Open this post in threaded view
|

Re: Release candidates available for testing

Chris Morley-3
In reply to this post by Douglas Houston
In the development code, -v now can take a file name. For -v and -s
the files can contain multiple pattern molecules (which are OR-ed).

Chris

On 08/09/2010 17:32, Douglas Houston wrote:

> Thanks Chris,
>
> The line you gave seemed to do the trick - after I escaped out the
> tilde. Is the usage of the tilde to reverse the sense of the filter
> documented? I would have thought the -v option would be logical.
>
> Sieve seems to do what I want - almost. It looks like it filters using
> SMARTS - but all my undesirable groups are in SDF format. How do I
> convert to SMARTS? OpenBabel doesn't seem to do it (although it can
> filter using them, strangely).
>
> I realise I could potentially use SMILES as input but this doesn't
> work in some cases, e.g. 'C1=CC=CC=C1' is a benzene ring in SMILES
> format but according to the following link will not match benzene if
> it's used as a SMARTS string.
>
> http://www.daylight.com/dayhtml/doc/theory/theory.smarts.html
>
> cheers,
>
> Doug
>
>
>
> Quoting Chris Morley<[hidden email]>:
>
>> On 08/09/2010 13:17, Douglas Houston wrote:
>>> Hi,
>>>
>>> According to this page:
>>>
>>> http://baoilleach.webfactional.com/site_media/ob-docs/Features/Fingerprints.html#fingerprints
>>>
>>> I can use the -s option with the name of a file, e.g.:
>>>
>>> babel mymols.sdf -sSubstructure.sdf mymols_filtered.sdf
>>>
>>> will write out all the compounds in mymols.sdf that contain the group
>>> in Substructure.sdf. However, what I'd like is the opposite, i.e. to
>>> write out everything that doesn't match. I tried using -v instead of
>>> -s but this didn't work.
>>
>> I think the following will do what you want
>> babel mymols.sdf -s ~Substructure.sdf mymols_filtered.sdf
>>>
>>> It would also be nice if I could use a file that contained more than
>>> one entry, i.e. the file Substructure.sdf could define multiple
>>> fragments that I don't want  in my molecules, rather than having to
>>> run Babel on each one separately.
>>
>> It would have to be decided whether the combination was AND or OR.
>> What you want is NOT(frag1 OR frag2 OR...)
>> In the related --filter option there is an implicit AND between tests,
>> but I think for both the -s filter and its inverse the OR behaviour is
>> more intuitive.
>>>
>>> Would this be doable before the final release?
>>
>> It may be, but I am about to go on holiday, so don't hold your breath.
>>
>> Thanks for this suggestion. Other suggestions for useful features are
>> very welcome.
>>
>> You could look at the Sieve program, http://www.silicos.be/sieve.html
>> which is an industrial-strength filtering program, now using OpenBabel.
>>
>> Chris
>>
>> ------------------------------------------------------------------------------
>> This SF.net Dev2Dev email is sponsored by:
>>
>> Show off your parallel programming skills.
>> Enter the Intel(R) Threading Challenge 2010.
>> http://p.sf.net/sfu/intel-thread-sfd
>> _______________________________________________
>> OpenBabel-discuss mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
>>
>>
>
>
>
>
> _____________________________________________________
> Dr. Douglas R. Houston
> Scientific Programme Coordinator / Research Associate
> Room 3.23
> Institute of Structural and Molecular Biology
> Michael Swann Building
> King's Buildings
> University of Edinburgh
> Edinburgh, EH9 3JR, UK
> Tel. 0131 650 7358
>
>
>
>
>
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 9.0.851 / Virus Database: 271.1.1/3121 - Release Date: 09/08/10 07:07:00
>


------------------------------------------------------------------------------
Start uncovering the many advantages of virtual appliances
and start using them to simplify application deployment and
accelerate your shift to cloud computing.
http://p.sf.net/sfu/novell-sfdev2dev
_______________________________________________
OpenBabel-discuss mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/openbabel-discuss