Structure-Based Classification of Chemical Reactions without Assignment of Reaction Centers

Abstract

The automatic classification of chemical reactions is of high importance for the analysis of reaction databases, reaction retrieval, reaction prediction, or synthesis planning. In this work, the classification of photochemical reactions was investigated with no explicit assignment of the reacting centers. Classifications were explored with Random Forests or Kohonen neural networks in three different situations, using different levels of information: (a) pairs of reactants were classified according to the type of reaction they produce, (b) products were classified according to the type of reaction from which they can be synthesized, and (c) reactions were classified from the difference between the descriptors of the product and the descriptors of the reactants. In all cases molecular maps of atom-level properties (MOLMAPs) were used as descriptors. They are generated by a self-organizing map and encode physicochemical properties of the bonds available in a molecule. Correct classification could be achieved for approximately 90% of the 78 reactions in an independent test set.

Sesquiterpene Lactones with Anti-Hepatitis C Virus Activity Using Molecular Descriptors

“…The difficulty to explain the biological activity of analogues (12)(13)(14)(15)(16)(17)(18) with the analyzed descriptors corroborate with the results observed previously by Hwang et al [20] who have already demonstrated that the same analogues were not recovered from the media where the anti-HCV activity was analyzed.…”

“…Nevertheless, it is observed that the HATS3v descriptor does not permit to explain exactly the activity of the parthenolide (1) analogues (12)(13)(14)(15)(16)(17)(18), which have amino substituent groups in the 13position. These groups collaborate to acquire a linear periphery, instead of a spherical one, that would be much more influent to HATS3v.…”

“…First of all, it is important to note that all active compounds analyzed by equation 1 have the -metilene--lactone moiety (1)(2)(3)(4)(5) or are substituted analogues that, somehow, recover this portion (11)(12)(13)(14)(15)(16)(17)(18).…”

“…(1). Molecular structures of SLs (1-8), exo-methylene lactones (9,10) and parthenolide analogues (11)(12)(13)(14)(15)(16)(17)(18)(19) analyzed for the anti-HCV activity.…”

“…The most used ANNs architecture for pattern recognition and classification is the Self-organizing maps (SOM). SOM is a powerful visualization tool able to reduce dimensions of projections and displays similarities among objects, which has already been successfully used in several applications in chemical database [15][16][17][18][19].…”