新永資訊有限公司


軟體資產管理家- 本公司可代為訂購國內外軟硬體
Tel:02-2597-1006   Fax:02-2597-1007
友站連結

COSMOsim

生化統計分析軟體
Biochemistry Software

歡迎來電洽詢
TEL: 02-2597-1006

需要產品報價?

COSMOsim  生命科學軟體

The COSMO-RS method has proven the σ-profiles as the crucial information for most ADME properties as solubility, blood-brain-partition coefficients, and intestinal absorption, and even for many adsorption phenomena. Considering this fundamental importance of the σ-profiles for surface interactions of molecules in liquid states, they most likely also carry a large part of information required for the estimation of desolvation and binding processes, which are responsible for the inhibition of enzyme receptors by drug molecules. Thus a high similarity with respect to the σ-profiles appears to be a necessary condition for drugs of similar physiological action.

Driven by this insight, we have developed a σ-profile based drug similarity measure SMS for the detection of new bioisosteric drug candidates, and a program, COSMOsim, which enables the efficient calculation of this similarity for large libraries, making use of our COSMOfrag technology. In several examples COSMOsim has already demonstrated 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.

Despite these initial successes, which clearly proof the significance of the σ-profile similarity for drug similarity and bioisoster screening, we are aware that high SMS similarity cannot be a sufficient condition for drug activity, because no 3D-information is included in this measure. More refined methods combining 3D aspects and SMS similarity are under development – partly in academic collaborations – and will be added to our COSMOsim program during the next one or two years. Nevertheless, even in its present early development stage, COSMOsim should open a novel view to drug similarity, and provide many new and complementary ideas compared to the conventional more structure based similarity search methods.