Hi Nina,
I didn't mean to offend but i was a bit vague. My comment about UIT: It's
with using a prescreen using fingerprints (extended fingerprinter) in memory
before actually using UIT.
Anyway once i get to it I will create a list of structures for the according
query and dataset that are not found using the ported chemkit algorithm.
that might be useful.
Best Regards,
Joos
2011/9/8 Nina Jeliazkova <[email protected]>
>
>
> On 8 September 2011 12:24, Joos Kiener <[email protected]> wrote:
>
>> Hi all,
>>
>> first off i guess you will be hearing more from me rather sooner than
>> later but now to the actually subject. Please see:
>>
>>
>> http://chembioinfo.wordpress.com/2011/03/15/benchmarking-substructure-search/
>>
>> for the context of this message.
>>
>> Currently I'm playing around with substructure search ( I have a certain
>> goal in mind, more on that in later messages). Anyway UIT isn't exactly fast
>> especially compared to commercial products like ChemFinder or InstantJChem
>> were searches seem almost instantaneous.
>>
>>
> Just to note a comparison between a library method (as UIT) to a database
> search is not quite fair, as database search systems usually employ lot of
> pre-screening and other optimization techniques.
>
> e.g. this online search does use CDK ( but not UIT )
>
>
> http://apps.ideaconsult.net:8080/ambit2/query/smarts?type=smiles&search=CCC%28C%28CC%28C%28C%29C%29C%29C%29C&text=&page=0&pagesize=100
>
>
> Best regards,
> Nina
>
>
>> I was comparing UIT and the above referenced code ported from chemkit.
>> First the difference in real world usage seems much less extreme than in
>> that benchmark (for small molecules) or I'm misinterpreting the chart.
>> Anyway in my case it takes about 60% of the time compared to UIT.
>>
>> Now to the subject of the message. I think there is an issue in the ported
>> version. Following query returns 44 hits with chemkit and 106 with UIT.
>> ChemFinder also gives 106 hits so I'm inclined to believe 106 is correct.
>>
>> Here the Query Molecule:
>>
>> CCC(C(CC(C(C)C)C)C)C
>>
>> Did not find or check for any other inconsistencies.
>>
>> Best Regards,
>>
>> Joos Kiener
>>
>>
>> ------------------------------------------------------------------------------
>> Doing More with Less: The Next Generation Virtual Desktop
>> What are the key obstacles that have prevented many mid-market businesses
>> from deploying virtual desktops? How do next-generation virtual desktops
>> provide companies an easier-to-deploy, easier-to-manage and more
>> affordable
>> virtual desktop model.http://www.accelacomm.com/jaw/sfnl/114/51426474/
>> _______________________________________________
>> Cdk-user mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/cdk-user
>>
>>
>
------------------------------------------------------------------------------
Doing More with Less: The Next Generation Virtual Desktop
What are the key obstacles that have prevented many mid-market businesses
from deploying virtual desktops? How do next-generation virtual desktops
provide companies an easier-to-deploy, easier-to-manage and more affordable
virtual desktop model.http://www.accelacomm.com/jaw/sfnl/114/51426474/
_______________________________________________
Cdk-user mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/cdk-user
