Hello Greg,
I develop my own algorithm which is based on machine learning, therefore I
am looking for detailed shape descriptors to incorporate. The prerequisite
is that the descriptor must be expressed as an array of numbers. I don't
have something in mind specifically, but from the literature research that
I did I think the simplest approach is to work with inter-atomic distances
and distribution moments. USR can be easily expanded to include more
numbers (e.g. inclusion of moments higher than the 3rd and incorporation of
more inter-atomic distances). I guess the patent will not apply in this
case. So before I start writing something of my own, I would like to know
if there is something similar already available.
Can the 3D descriptors from DRAGON be expressed as arrays of numbers?
best
Thomas
On 20 May 2017 at 07:56, Greg Landrum <greg.land...@gmail.com> wrote:
> Hi Thomas,
>
> There isn't currently anything there.
> The RDKit had a USR (and USRCAT) implementation a few years ago, but we
> removed it because of the patent on USR. Now that the patent has lapsed,
> there's an active PR to re-integrate those descriptors.
>
> There's also a PR from Guillaume Godin that implements a large number of
> 3D descriptors that are known from the DRAGON software. I'm still reviewing
> that one, but it should be integrated in the not-too-distant future.
>
> Is there any particular descriptor that you're looking for?
>
> -greg
>
> On Fri, May 19, 2017 at 12:40 PM, Thomas Evangelidis <teva...@gmail.com>
> wrote:
>
>>
>> Greetings,
>>
>> Is there any shape descriptor available in RDKit that can be expressed as
>> an array of numbers (e.g. like the 2D similarity fingerprints)?
>> Alternatively is anyone aware of any other implementation of such a
>> descriptor? The only one I know is the Ultrafast Shape Recognition (USR)
>> algorithm which encodes shape information into 12 floating point numbers,
>> but the shape information that it provides is rather poor since it was
>> adapted for screening millions of compounds in short time scales.
>>
>> I would appreciate any advice.
>>
>> best,
>> Thomas
>>
>>
>> --
>>
>> ======================================================================
>>
>> Dr Thomas Evangelidis
>>
>> Post-doctoral Researcher
>> CEITEC - Central European Institute of Technology
>> Masaryk University
>> Kamenice 5/A35/2S049,
>> 62500 Brno, Czech Republic
>>
>> email: tev...@pharm.uoa.gr
>>
>> teva...@gmail.com
>>
>>
>> website: https://sites.google.com/site/thomasevangelidishomepage/
>>
>>
>> ------------------------------------------------------------
>> ------------------
>> 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
>>
>>
>
--
======================================================================
Dr Thomas Evangelidis
Post-doctoral Researcher
CEITEC - Central European Institute of Technology
Masaryk University
Kamenice 5/A35/2S049,
62500 Brno, Czech Republic
email: tev...@pharm.uoa.gr
teva...@gmail.com
website: https://sites.google.com/site/thomasevangelidishomepage/
------------------------------------------------------------------------------
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
