cyclization modelling with SMILES

[astar]

Dear All;

We are a academic group working on a bioinformatics application written in Java, that involves certain chemistry solutions. That's the part we need some help solving. We are storing our molecules in SMILES format, and at the end our goal is to turn linear structures into cyclisized.

FIRST QUESTION: are SMILES the best choice for database entries and computation
of new structres???

SECOND QUESTION: if you model spontaneous cyclization from a linear compound, how can you determine which product out of all possible ones is most likely to be the real thing

THIRD QUESTION: is there a software that can help us solve "second question", and could you recomend us some(it should cooperate with SMILES "Daylight toolkit" if possible)

P.S. are there any Java libraries for this stuff???

Thanks for all suggestions!

[alexceb]

There is a type of reaction called Electrocyclic. It is of the pericyclic reactions, known for their conserted mechinism. An electrocyclic reaction can form a ring or break a ring. Can also make a ring into a double ring.

These reactions are stereospecific- the Woodward-Hoffman rules predict which products are "allowed".

Though these reactions disallow certain products, stereoisomers can still form. That prefferece can be inferred form substituent effects - for that, read some papers by K.N. Houk. He, along with others, have studied these reactions (and more) using computational chemistry tools.