Dear Dmitry,
Excerpts from Druzhilovsky's message of Thu Dec 23 13:51:05 +0100 2010:
> Dear Christoph,
> 
> There are no smart reasons to use SMARTS notation instead of MNA.

As we are aiming in OpenTox for interoperable webservices we need a
common standard for representing substructures. We already had a quite
lengthy discussion on this topic and came to the conclusion that
substructure features should be identified by a ot:smarts entry (as in
the example I have sent before).

Best regards,
Christoph

> 1. There is no reason to use 15 characters of SMARTS notation
> "[H]C([H])([H])N" instead of only five of MNA "CHHHN".
> 2. As I understand your example:
>   H
>   |
> H-C-N
>   |
>   H
>  
> which would translate to the SMARTS string "[H]C([H])([H])N"
>  
> the three another SMARTS strings can be generated:
>   H
>   |
> N-C-H       NC([H])([H])[H]
>   |
>   H
>  
>   N
>   |
> H-C-H       [H]C(N)([H])[H]
>   |
>   H
>  
>   H
>   |
> H-C-H       [H]C([H])(N)[H]
>   |
>   N
>  
> But MNA is unique - it is "CHHHN" in all cases.
> 3. In general, the MNA descriptors are determined as (D. Filimonov, V.
> Poroikov, Probabilistic Approaches in Activity Prediction, in
> Chemoinformatics Approaches to Virtual Screening, A. Varnek and A. Tropsha,
> eds., Cambridge (UK): RSC Publishing, 2008, pp. 182-216):
> The 2D structural formulae of compounds were chosen as the basis for
> description of chemical structure because this is the only information
> available in the early stage of research. The Multilevel Neighbourhoods of
> Atoms (MNA) descriptors are based on the molecular structure representation,
> which includes the hydrogens according to the valences and partial charges
> of other atoms and does not specify the types of bonds. MNA descriptors are
> generated as recursively defined sequence:
> .    zero-level MNA descriptor for each atom is the mark A of the atom
> itself;
> .    any next-level MNA descriptor for the atom is the sub-structure
> notation A(D1D2...Di...),
> where Di is the previous-level MNA descriptor for i-th immediate neighbour's
> of the atom A.
> The mark of atom may include not only the atomic type but also any
> additional information about the atom. In particular, if the atom is not
> included into the ring, it is marked by "-". The neighbour descriptors
> D1D2...Di... are arranged in unique manner, e.g., in lexicographic order.
> Iterative process of MNA descriptors generation can be continued covering
> first, second, etc. neighbourhoods of each atom. MNA descriptors have more
> general background than the descriptors, which look like MNA.
> The molecular structure is represented by the set of unique MNA descriptors
> of the 1st and 2nd levels.
>  
> So, if You wish use SMARTS then you can use this notation, but we use for
> SAR/QSAR MNA about fifteen years and we concluded that MNA descriptors are
> very smart, powerful and universal descriptors.
> 
> Best regards
> 
> Dmitry S. Druzhilovsky
> 
> Laboratory of Structure-Function Based Drug Design
> 119121, Russia, Moscow, Pogodinskaya street, 10 
> Phone: +7 499 255-30-29
> Fax: +7 499 245-08-57
> 
> 
> D> -----Original Message-----
> D> From: development-bounces at opentox.org [mailto:development-
> D> bounces at opentox.org] On Behalf Of Christoph Helma
> D> Sent: Wednesday, December 22, 2010 3:43 PM
> D> To: development
> D> Subject: Re: [OTDev] IBMC QNA\MNA services
> D> 
> D> Dear Dimitry,
> D> 
> D> Excerpts from Druzhilovsky's message of Wed Dec 22 12:09:52 +0100
> D> 2010:
> D> >
> D> > Dear Christoph,
> D> >
> D> > You have written:
> D> > "I have ... suspect that some brackets around explicit hydrogens are
> D> misssing."
> D> >
> D> > Explanation.
> D> > The MNA descriptors:
> D> >
> D> > HO
> D> 
> D> I was suggesting to provide MNA descriptors in SMARTS notation. I
> D> guess
> D> that "CHHHN" stands for
> D> 
> D>   H
> D>   |
> D> H-C-N
> D>   |
> D>   H
> D> 
> D> which would translate to the SMARTS string "[H]C([H])([H])N".
> D> 
> D> Best regards,
> D> Christoph
> D> _______________________________________________
> D> Development mailing list
> D> Development at opentox.org
> D> http://www.opentox.org/mailman/listinfo/development
>