Quantitative pharmacophore models with inductive logic programming

Three-dimensional models, or pharmacophores, describing Euclidean constraints on the location on small molecules of functional groups (like hydrophobic groups, hydrogen acceptors and donors, etc.), are often used in drug design to describe the medicinal activity of potential drugs (or ‘ligands’). This medicinal activity is produced by interaction of the functional groups on the ligand with a binding site on a target protein. In identifying structure-activity relations of this kind there are three principal issues: (1) It is often difficult to “align” the ligands in order to identify common structural properties that may be responsible for activity; (2) Ligands in solution can adopt different shapes (or `conformations’) arising from torsional rotations about bonds. The 3-D molecular substructure is typically sought on one or more low-energy conformers; and (3) Pharmacophore models must, ideally, predict medicinal activity on some quantitative scale. It has been shown that the logical representation adopted by Inductive Logic Programming (ILP) naturally resolves many of the difficulties associated with the alignment and multi-conformation issues. However, the predictions of models constructed by ILP have hitherto only been nominal, predicting medicinal activity to be present or absent. In this paper, we investigate the construction of two kinds of quantitative pharmacophoric models with ILP: (a) Models that predict the probability that a ligand is “active”; and (b) Models that predict the actual medicinal activity of a ligand. Quantitative predictions are obtained by the utilising the following statistical procedures as background knowledge: logistic regression and naive Bayes, for probability prediction; linear and kernel regression, for activity prediction. The multi-conformation issue and, more generally, the relational representation used by ILP results in some special difficulties in the use of any statistical procedure. We present the principal issues and some solutions. Specifically, using data on the inhibition of the protease Thermolysin, we demonstrate that it is possible for an ILP program to construct good quantitative structure-activity models. We also comment on the relationship of this work to other recent developments in statistical relational learning. Editors: Tamás Horváth and Akihiro Yamamoto     

Egg handling and storage

The relative importance of fat and lean tissue mass in determining bone mineral mass among postmenopausal women was examined in this 1-year longitudinal study. Fifty postmenopausal Caucasian women entered the study; 45 of them completed a 1-year follow-up. Dual-energy X-ray absorptiometry was employed for measuring total and regional bone mineral density (BMD) and bone mineral content (BMC), fat tissue mass (FTM), lean tissue mass (LTM), and body weight. Results from linear regression analysis using the cross-sectional data (n = 50) of the study indicated that LTM explained a larger percentage of variation in bone mineral mass than did FTM. FTM and LTM were found to be moderately correlated (r = 0.55); when FTM was entered in the same predicting regression models, LTM was a significant predictor (p < 0.05) of the total and regional BMC, but not BMD. The percent FTM (and inversely %LTM) was correlated with BMD and BMC, but significant correlation was primarily found only for total body BMD (or BMC). Weight was the best predictor of total body BMD and BMC. Longitudinally (n = 45), annual changes in both FTM and weight were significantly associated with annual changes in regional BMD after adjustment for initial bone mineral values (p < 0.05). We conclude that bone mineral mass is more closely related to LTM than to FTM, while annual changes in regional BMD are more closely correlated with changes in FTM in healthy postmenopausal women. Meanwhile, increased body weight is significantly associated with increased bone mineral mass.     

MANAGING DISTRIBUTED COMPUTING Five Practices for Success

Management of distributed environments requires corporate IS managers to shift from being operators of a central utility to facilitators who can recognize the synergy among loosely related groups of users. Success depends on the development of new ways of visualizing and measuring the service delivery process.     

Simulation of magnetic component models in electric circuitsincluding dynamic thermal effects

It is essential in the simulation of power electronics applications to model magnetic components accurately. In addition to modeling the nonlinear hysteresis behavior, eddy currents and winding losses must be included to provide a realistic model. In practice the losses in magnetic components give rise to significant temperature increases which can lead to major changes in the component behavior. In this paper a model of magnetic components is presented which integrates a nonlinear model of hysteresis, electro-magnetic windings and thermal behavior in a single model for use in circuit simulation of power electronics systems. Measurements and simulations are presented which demonstrate the accuracy of the approach for the electrical, magnetic and thermal domains across a variety of operating conditions, including static thermal conditions and dynamic self heating     

What We Mean by Future-Proofing

Perhaps the most powerful claim that advocates of fiber can make,on top of the massive bandwidthfiber makes possible today, is that fiber is future-proof. That is, the fiber that is laid today can be upgraded so that the same strands of glass can carry massively more bandwidth in the future.     

Functional topography of cat primary auditory cortex: response latencies

Minimum onset latency (Lmin) of single- and multiple-unit responses were mapped in the primary auditory cortex (AI) of barbiturate-anesthetized cats. Contralateral Lmin for multiple units was non-homogeneously distributed along the dorso-ventral/isofrequency axis of the AI. Responses with shorter latencies were more often located in the central, more sharply tuned region while longer latencies were more frequently encountered in the dorsal and ventral portions of the AI. For single units, a large scatter of Lmin values was found throughout the extent of the AI including cortical depth. The relationship between Lmin and previously reported spectral, intensity and temporal parameters was analyzed and revealed statistically significant correlations between minimum onset latency and the following response properties in some but not all studied animals: sharpness of tuning of a frequency response area 10 dB above threshold, broadband transient response, strongest response level, monotonicity of rate/level functions, dynamic range, and preferred frequency modulation sweep direction. This analysis suggests that Lmin is determined by several independent factors and that the prediction of Lmin based on relationships with other spectral and temporal response properties is inherently weak. The spatial distribution and the functional relationship between these response parameters may provide an important aspect of the time-based cortical representation of specific features in the animal's natural environment.