ambit-users Mailing List for AMBIT:Chemical Structure DB&Web Service (Page 4)

SourceForge Headquarters 225 Broadway Suite 1600 San Diego, CA 92101 +1 (858) 422-6466

Hi Yannick,

Nina has given you good advices concerning the compilation and 
integration problems (I will not add to these).

I would like to add few comments on the test cases mentioned in your 
first mail.
I tested all these SMIRKS in pure java approach and they work fine.
The first test case 
([*;#6:1][C:2]#[C:3][*;#6:4]>>[*:1]/[C:2](/[H])=[*:3](/[H])[C:4]) 
generates parser error. I clarified this below after the java code. The 
second test case in my mail is the first test corrected in order to be 
parsed without problems.

Here is the testing code for my tests (it is quite analogous to the one 
shown in the examples on the Ambit web page)

public void testSMIRKS(String smirks, String targetSmiles)throws 
Exception
	{
		System.out.println("Testing SMIRKS: " + smirks);
		SMIRKSManager smrkMan = new 
SMIRKSManager(SilentChemObjectBuilder.getInstance());
		SMIRKSReaction reaction = smrkMan.parse(smirks);
		if (!smrkMan.getErrors().equals(""))
		{
			System.out.println(smrkMan.getErrors());
			 return;
		 }

		if (targetSmiles.equals(""))
			 return;

		IAtomContainer target = 
SmartsHelper.getMoleculeFromSmiles(targetSmiles);
		smrkMan.applyTransformation(target, reaction);
		String transformedSmiles = SmartsHelper.moleculeToSMILES(target);

		System.out.println("Reaction application: " + targetSmiles + "  -->  " 
+ transformedSmiles);
	}

...

testSMIRKS("[*;#6:1][C:2]#[C:3][*;#6:4]>>[*:1]/[C:2](/[H])=[*:3](/[H])[C:4]","COCCCC#CCCC");
testSMIRKS("[*:1][C:2]#[C:3][*;#6:4]>>[*:1]/[C:2](/[H])=[C:3](/[H])[C:4]","COCCCC#CCCC");  //corercted one
testSMIRKS("[*;#6:1][C:2]=[O:3]>>[*;#6:1][C:2]-[O:3][*;#6]","CCCC=O");
testSMIRKS("[C:1]=O>>[C:1]1OCCO1","CCCCCCC=O");
testSMIRKS("[*:1]C(=O)O>>[*:1]C(N)=O","CCCCCCC(=O)O");

And here is the result output from the code above:

Testing SMIRKS: 
[*;#6:1][C:2]#[C:3][*;#6:4]>>[*:1]/[C:2](/[H])=[*:3](/[H])[C:4]
Map 1 atom types are inconsistent!
Map 3 atom types are inconsistent!

Testing SMIRKS: 
[*:1][C:2]#[C:3][*;#6:4]>>[*:1]/[C:2](/[H])=[C:3](/[H])[C:4]
Reaction application: COCCCC#CCCC  -->  [H]C(=C([H])CCCOC)CCC

Testing SMIRKS: [*;#6:1][C:2]=[O:3]>>[*;#6:1][C:2]-[O:3][*;#6]
Reaction application: CCCC=O  -->  COCCCC

Testing SMIRKS: [C:1]=O>>[C:1]1OCCO1
Reaction application: CCCCCCC=O  -->  C1COC(O1)CCCCCC

Testing SMIRKS: [*:1]C(=O)O>>[*:1]C(N)=O
Reaction application: CCCCCCC(=O)O  -->  NC(=O)CCCCCC

Just few comments for the first test case:
Ambit SMARTS/SMIRKS parser has STRICT rules for the mapped atom types. 
The atom types corresponding to the same mapping index on the left 
(reactant) and on the right (products) must be of the same type 
otherwise parser interprets this as an error.

For the following SMIRKS  
[*;#6:1][C:2]#[C:3][*;#6:4]>>[*:1]/[C:2](/[H])=[*:3](/[H])[C:4]
we have
[*;#6:1]  >>  [*:1]
i.e. the left is carbon and right is any atom (wild card only)
Also bear in mind that [*;#6:1] is equivalent to [#6:1] so the first 
token '*' is not needed.

Analogously parser produces error message for:
[*;#6:4] >> [*:3]

This requirement for same atoms types is not applied for NOT-MAPPED 
atoms in the SMIRKS expression.
Actually not mapped atoms are interpreted as atoms(fragments) been 
removed on the left and atoms (fragments) been added on the right. That 
is why it is not required to be of the same type.

For example: this SMIRKS is OK
[C:1]N>>[C:1]O

I hope my comments would be helpful as well.
Best regards
Nick

On 2014-03-04 20:39, Yannick .Djoumbou wrote:
> Hi Nina,
> 
> According to your last email and to the information on this file [4],
> there is a separate ambit-smirks-2.5.4- module, which is an extension
> of ambit-smarts. Well, I could not find a downloadable jar file on the
> web. Looking at the file I attached, is it OK to have those require
> libraries, or do i explicitly require the ambit-smirks library beside
> ambit-core, ambit-smarts, and ambit-core?
> 
> Cheers,
> 
> Yannick
> 
> On Mon, Mar 3, 2014 at 11:54 PM, Yannick <y.d...@gm...> wrote:
> 
>> Hi Nina,
>> 
>> Thanks for your reply. I changed the cdm version to 1.4.11. However,
>> I still have the same error messages.
>> 
>> Thanks in advance for your help.
>> 
>> Yannick
>> 
>> On 2014-03-03, at 11:30 PM, Nina Jeliazkova
>> <jel...@gm...> wrote:
>> 
>> Dear Yannick,
>> 
>> Thanks for your interest in ambit-smirks.
>> 
>>  (cc: to Nick Kochev, who is the author of the package).
>> 
>> I am not familiar with Ruby/JRuby, but the NoMethodError
>> newInstance  means some of the dependencies is missing and the
>> corresponding class could not be instantiated.
>> 
>> ambit-smirks 2.5.4-x depends on CDK 1.4.11 and has not been tested
>> with CDK 1.4.19 - quite probably this might be the reason. Could you
>> try changing the CDK dependency to 1.4.11 .
>> 
>> Best regards,
>> Nina
>> 
>> On 3 March 2014 23:39, Yannick .Djoumbou <y.d...@gm...>
>> wrote:
>> 
>> Hello,
>> 
>> I am a new user of AMBIT. I would like to use SMIRKS and I have
>> browsed through some examples on this web page [1].
>> I have tried some examples with several SMIRKS but I have add
>> several issues.
>> In the attached file, I show a prototype of the JRuby code I use.
>> All the libraries have been imported properly.
>> Please have a look at the examples provided in the file above.
>> 
>> Would you please let me know what is the cause for this? I have
>> looked at the source code but I could find a solution.
>> 
>> Cheers,
>> 
>> Yannick
>> 
> ------------------------------------------------------------------------------
>> Subversion Kills Productivity. Get off Subversion & Make the Move
>> to Perforce.
>> With Perforce, you get hassle-free workflows. Merge that actually
>> works.
>> Faster operations. Version large binaries.  Built-in WAN
>> optimization and the
>> freedom to use Git, Perforce or both. Make the move to Perforce.
>> 
> http://pubads.g.doubleclick.net/gampad/clk?id=122218951&iu=/4140/ostg.clktrk
>> [2]
>> _______________________________________________
>> Ambit-users mailing list
>> Amb...@li...
>> https://lists.sourceforge.net/lists/listinfo/ambit-users [3]
> 
> 
> 
> Links:
> ------
> [1] 
> http://ambit.sourceforge.net/AMBIT2-LIBS/ambit2-smarts/examples.html#SMIRKS
> [2]
> http://pubads.g.doubleclick.net/gampad/clk?id=122218951&amp;iu=/4140/ostg.clktrk
> [3] https://lists.sourceforge.net/lists/listinfo/ambit-users
> [4]
> file:///Users/yandj/Programming/Projects/drug_metabolism/opentoxeuro2013_submission_20.pdf

2014	Jan	Feb	Mar (6)	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
2015	Jan (4)	Feb	Mar	Apr (34)	May (4)	Jun (1)	Jul (3)	Aug (9)	Sep	Oct	Nov (1)	Dec
2016	Jan	Feb (2)	Mar	Apr	May	Jun	Jul (2)	Aug (1)	Sep (4)	Oct	Nov (1)	Dec
2017	Jan	Feb (1)	Mar (2)	Apr	May	Jun (4)	Jul	Aug (1)	Sep	Oct	Nov	Dec
2019	Jan	Feb	Mar	Apr	May	Jun	Jul (1)	Aug	Sep	Oct	Nov	Dec

ambit-users Mailing List for AMBIT:Chemical Structure DB&Web Service (Page 4)

Chemical structures database & machine learning with web services API

ambit-users — Ambit users