Abstract: Models predict activity of chemical compounds by employing “transformed” descriptor values. The descriptors are transformed via transformation functions that convert the raw descriptor values to new values better representing the contribution of the descriptors to the activity in question. Typically, these transformation functions are non-linear parametric functions such as Gaussian functions or sigmoid functions. Typically, the model will employ at least two different descriptors, each transformed by its own non-linear parametric function.

Abstract: Models predict activity of chemical compounds by employing “transformed” descriptor values. The descriptors are transformed via transformation functions that convert the raw descriptor values to new values better representing the contribution of the descriptors to the activity in question. Typically, these transformation functions are non-linear parametric functions such as Gaussian functions or sigmoid functions. Typically, the model will employ at least two different descriptors, each transformed by its own non-linear parametric function.

Abstract: Descriptor based models, employing at least two descriptors, predict activities of compounds under consideration. Those activities may be biology-based activities such as the ability of the compound to cross the blood brain barrier. In one example, the model predicts a compound's solubility, its ability to be absorbed in the intestine, and its ability to cross the blood brain barrier. The descriptors of interest are typically physicochemical properties of the whole molecule. Examples include a log P or log D, molecular weight or related size-based descriptors, the number of hydrogen bond donors and/or hydrogen bond acceptors, formal charge, lipophilicity, and the like.

Abstract: Systems and method are provided for modeling substrate molecules so that the various pathway reaction rates, and thus their overall reaction rates and metabolic properties, can be predicted. The current invention provides various systems and methods for stoichiometrically measuring the pathway reaction rates, both directly and indirectly. By repeating this for a class or several classes of substrate molecules, a general model of pathway reaction rates can be developed by correlating observed pathway reaction rates to the actual structural descriptors of the molecules, in particular, features around the reactive sites. The model can then be used to predict and design substrates according to desired metabolic characteristics. The systems and methods are particularly applicable to metabolism of substrate molecules by the cytochrome P450 enzymes.

Abstract: Accessibility correction factors may be used to modify values predicted by models of electronic component substrate reactivity. Most of the correction factors described herein pertain to either steric or orientation effects on substrate accessibility. The correction factors may be derived from one or more “descriptors” of the substrate structure. Each group of descriptors and associated correction factor pertain to a particular site on the substrate. Examples of such descriptors include site polarity, protrusion, partial surface area, partial charge, etc. Often the correction factor is a function of multiple descriptors. The function may be an expression comprising multiple terms, each representing the weighted contribution of a particular descriptor. In other embodiments, the correction factor is simply a descriptor or a descriptor multiplied by a coefficient or other function.