Michael Almstetter

COSMO s im : Bioisosteric Similarity Based on COSMO-RS sigma Profiles.

A novel approach for the quantification of drug similarity is proposed, which makes use of the surface polarities, that is, conductor surface polarization charge densities sigma, as defined in the quantum chemically based conductor-like screening model for realistic solvation(COSMO-RS). The histogram of these surface polarities, the so-called sigma profiles, have been proven to be the key for the calculation of all kinds of partition and adsorption coefficients and, therefore, of relevant absorption, distribution, metabolism, and excretion parameters such as solubility, pKa, log BB, and many others. They also carry a large part of the information required for the estimation of desolvation and binding processes responsible for receptor binding and enzyme inhibition of drug molecules. Thus, a large degree of similarity with respect to the sigma profiles appears to be a necessary condition for drugs of similar physiological action. Driven by this insight, we propose a sigma-profile-based drug similarity measure COSMOsim for the detection of new bioisosteric drug candidates. In several examples, we demonstrate its statistical and pharmaceutical plausibility, its practicability for real drug research projects, and its unique independence from the chemical structure, which enables scaffold hopping in a natural way.

Nomen Est Omen: Quantitative Prediction of Molecular Properties Directly from IUPAC Names

The International Union of Pure and Applied Chemistry (IUPAC) was formed in 1919 by chemists from indus- try and academia (1). Over nearly nine decades the Union has succeeded in fostering worldwide communications in the chemical sciences and in uniting chemistry - academic, industrial and government - in a common language. As one of the results of the Union, IUPAC names nowadays serve as a commonly agreed text representation of chemical structures in patents, publications and databases. In public databases of chemical compounds, like PubChem with more than 12 million entries, chemical structures are identified by default using their IUPAC names (2). We report a very fast linguistic method to extract the implicit information contained in IUPAC names to statistically predict pharmacologically relevant proper- ties. This provides an efficient annotation tool that can be used to assess the likelihood of a given compound as a drug candidate and renders the entire chemical literature a searchable database for virtual screening experiments and data mining.

Diversity in Very Large Libraries

Combinatorial chemistry methods can be used, in principle, for the synthesis of very large compound libraries. However, these very large libraries are so large that the enumeration of all individual members of a library may not be practicable. We discuss here how one may increase the chances of finding compounds with desired properties from very large libraries by using combinatorial optimisation methods. Neuronal networks, evolutionary programming and especially genetic algorithms are heuristic optimisation methods that can be used implicitly to discover the relation between the structure of molecules and their properties. Genetic algorithms are derived from principles that are used by nature to find optimal solutions. Genetic algorithms have now been adapted and applied with success to problems in combinatorial chemistry. The optimisation behaviour of genetic algorithms was investigated using a library of molecules with known biological activities. From these studies, one can derive methods to estimate the diversity and structure property relationships without the need to enumerate and calculate the properties of the whole search space of these very large libraries.