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     

Structural properties of inorganic materials: tungsten bronze SBN ceramics

Rare-earth doped strontium barium niobates were synthesized using usual ceramic technique. The dopants are La, Ce, Gd, Sm and Nd. The materials were characterized by XRD and density measurements. The grain sizes were determined from SEM analysis. Lattice parameters changed uniformly with rare-earth dopants in unfilled structures. Density measurements and SEM analysis confirmed only minute changes in the densities of the ceramics.     

Comparison of mass transfer efficiency in horizontal rotating packed beds and rotating biological contactors

In this study, the mass transfer efficiencies of a novel horizontal rotating packed (h‐RPB) bed and the conventional disc‐type rotating biological contactor (RBC) were studied at four speeds and seven submergences. Pall rings of two different sizes (25, 38 mm), superintalox saddles and a wiremesh spiral bundle were used as packings in the h‐RPB. Volumetric gas–liquid mass transfer coefficients were determined by unsteady state absorption of atmospheric oxygen in de‐aerated water. Power consumption per unit liquid volume has been found for all geometries tested. The oxygen transfer efficiency values for the h‐RPB were found to be 2–5 kg kWh^?1 and for the disc RBC were found to be 1–2 kg kWh^?1. The performance of the h‐RPB was also compared with other gas–liquid contactors such as surface aerators. The study proves that the h‐RPB is a energy efficient alternative to conventional contactors. Copyright © 2005 Society of Chemical Industry     

Design considerations for high performance avalanche photodiodemultiplication layers

We have studied the effect of the thickness of the multiplication region on the noise performance characteristics of avalanche photodiodes (APD's). Our simulation results are based on a full band Monte Carlo model with anisotropic threshold energies for impact ionization. Simulation results suggest that the well known McIntyre expression for the excess noise factor is not directly applicable for devices with a very thin multiplication region. Since the number of ionization events is drastically reduced when the multiplication layer is very thin, the “ionization coefficient” is not a good physical parameter to characterize the process. Instead “effective quantum yield,” which is a measure of the total electron-hole pair generation in the device, is a more appropriate parameter to consider. We also show that for the device structure considered here, modeling the excess noise factor using a “discrete Bernoulli trial” model as opposed to the conventional “continuum theory” produces closer agreement to experimental measurements. Our results reinforce the understanding that impact ionization is a strong function of carrier energy and the use of simplified field-dependent models to characterize this high energy process fails to accurately model this phenomenon     

A microfluidic device for thermal particle detection

We demonstrate the use of heat to count microscopic particles. A thermal particle detector (TPD) was fabricated by combining a 500-nm-thick silicon nitride membrane containing a thin-film resistive temperature detector with a silicone elastomer microchannel. Particles with diameters of 90 and 200 μm created relative temperature changes of 0.11 and ?0.44 K, respectively, as they flowed by the sensor. A first-order lumped thermal model was developed to predict the temperature changes. Multiple particles were counted in series to demonstrate the utility of the TPD as a particle counter.     

A New Method for Getting Rational Approximation for Fractional Order Differintegrals

In this paper a new algorithm is presented to calculate the poles and zeros to approximate a fractional order (FO) differintegral (s^±α,α∈(0,1)) by a rational function on a finite frequency band ω∈(ω_l,ω_h). The constant phase property of the FO differintegral is the basis for development of the algorithm. Interlacing of real poles and zeros is used to achieve the constant phase. The calculations are done using the asymptotic Bode phase plot. A brief investigation is made to get a good approximation for the Bode phase plot. Two design parameters are introduced to keep the average phase close to the desired phase angle and to keep the error within the allowed bounds. A study is done to empirically understand the relationship between the error and the design parameters. The results thus obtained help in the further calculations. The algorithm is computationally simple and inexpensive, and gives a fairly good approximation of fractance frequency response on the specified frequency band.     

Mixed-Mode Stress Intensity Factor Determination of Riveted Lap Joints Used in Aircraft Fuselage Structures

Journal of Failure Analysis and Prevention - In the present work, mixed-mode stress intensity factor (SIF) of multiple cracks in a riveted lap joint has been determined, with and without the...     

A simulation study on the US rehabilitation service process for patients with modest-to-severe traumatic brain injury

Traumatic brain injury (TBI) is a devastating injury with severe consequences. In this paper, we conduct a simulation study on the commonly implemented care delivery process for TBI rehabilitation in the US, which covers three care categories: inpatient acute, outpatient sub-acute and general residential care. Our investigation is focused on assessing how coverage duration of publicly funded rehabilitation impacts two key system outcomes: sub-acute rehabilitation readmission and total rehabilitation spending. We develop prediction models on the above two outcomes for patients of different conditions. We introduce the notions of forceful transition and medical necessity adjustment, and embed the notions in a state-transition simulation model. Our simulation results suggest that to minimise the care spending, the duration of publicly insured outpatient sub-acute rehab be set smaller than what is currently implemented but not to the point where coverage should be completely removed. Our sensitivity analysis justifies the robustness of our results under variations on model parameters.     

Uncovering Global Icebergs in Distributed Streams: Results and Implications

Discovering icebergs in distributed streams of data is an important problem for a number of applications in networking and databases. While previous work has concentrated on measuring these icebergs in the non-distributed streaming case or in the non-streaming distributed case, we present a general framework that allows for distributed processing across multiple streams of data. We compare several of the state-of-the-art streaming algorithms for estimating local elephants in the individual streams. However, since an iceberg may be hidden by being distributed across many different streams, we add a sampling component to handle such cases. We provide a novel taxonomy of current sketches and perform a thorough analysis of the strengths and weaknesses of each scheme under various QoS metrics, using both real and synthetic Internet trace data. We summarize their performance and discuss the implications for the future design of sketches.