To: fis@listas.unizar.es
Subject: [Fis] molecular structures (was: ON MOLECULAR BIONETWORKS (II))
Dear FISers,
Dear Jerry,
I would like to give my opinion about some of your excellent remarks about
chemical structure:
> One conclusion was that the networks of Kirchoff could not be used to =20=
> describe chemical structures. Nonetheless, the "drawings" of
=20
> Kirchoff's networks and the "drawings" of chemical structures
look =20
> very very similar. If we call such drawings by the formal =20
> mathematical name of "graphs", then the Graphs of Kirchoff
morphed =20
> into groups and became closely associated with differential equations
=20=
> and factoring of products (polynominals), linear algebra. Thus, =20
> chemical theory separated from mathematical theory.
The chemical structure remains to be defined. Look in the CAS Registry Number
book, and count how many RN's the water has received: clearly, there is a
need to specify a particular use of the structure concept before selecting
a model for structure.
Actually, many chemical data bases include a graph (nodes and edges are colored)
to describe the structural formula. Of course, this is not convenient for
all chemists (need of 3D information, quantum data, etc.).
> Secondly, the optical isomer problem of tartaric acid initiated the =20
> general problem of isomers - substances of the same molecular formula
=20=
> but different physical - chemical properties
> So, what is the logic of a chemical structure? This is a crucial =20
> question.
Sure. CAS is experienced with chemical structures (may be roughly 40 millions
are known), and noticed the extreme importance of this question. It was one
of the reasons which lead them to define the RN (Registry Number).
We can understand that saying just "water" does not suffice to define a
chemical structure: is it vapor ? ice ? agregates in the liquid ?
isolated water molecule in the space ? the liquid itself ? deuterated water ?
unusual temperature/pressure conditions (critical point) ?
presence of electrical charges ? excited state of some conformer ? etc..
There many questions and situations which are difficult to imagine
without having the experience of chemical data banks.
> It is important to note that the physical community generally =20
> presupposes that "quantum mechanics" or "symmetry" or
=20
> "thermodynamics" provides the logic to solve these problems. I
=20
> invite individuals to address the relations between physical theories
=20=
> and these two questions - it would be a healthy opening for further =20
> discussions.
An example here: there are various file formats used to store chemical
structures on computer (mol, mol2, pdb, cas, csd, etc..), and many
users are asking for an universal format. BABEL do conversions, but
developpers would like to programme themselves the input/output modules
reading/writing molecular structures. There are tasks groups working
on this topic, but I am not optimistic: as a consequence of the multiple
definitions of "structure", it is impossible to define a file format
handling all situations for all users: it would be an enormous dataset,
so much complicated that nobody would use it.
Michel Petitjean, Email: petitjean@itodys.jussieu.fr
ITODYS (CNRS, UMR 7086) ptitjean@ccr.jussieu.fr
1 rue Guy de la Brosse Phone: +33 (0)1 44 27 48 57
75005 Paris, France. FAX : +33 (0)1 44 27 68 14
http://petitjeanmichel.free.fr/itoweb.petitjean.graphs.html
_______________________________________________
fis mailing list
fis@listas.unizar.es
http://webmail.unizar.es/mailman/listinfo/fis
Received on Tue Nov 22 12:15:54 2005
