No abstract

The paper aims to give a unified account of the central concepts in recent work on bandit processes and dynamic allocation indices; to show how these reduce some previously intractable problems to the problem of calculating such indices; and to describe how these calculations may be carried out. Applications to stochastic scheduling, sequential clinical trials and a class of search problems are discussed.

The prevention and control of Campylobacter colonization of poultry flocks are important public health strategies for the control of human campylobacteriosis. A critical review of the literature on interventions to control Campylobacter in poultry on farms was undertaken using a systematic approach. Although the focus of the review was on aspects appropriate to the United Kingdom poultry industry, the research reviewed was gathered from worldwide literature. Multiple electronic databases were employed to search the literature, in any language, from 1980 to September 2008. A primary set of 4,316 references was identified and scanned, using specific agreed-upon criteria, to select relevant references related to biosecurity-based interventions. The final library comprised 173 references. Identification of the sources of Campylobacter in poultry flocks was required to inform the development of targeted interventions to disrupt transmission routes. The approach used generally involved risk factor-based surveys related to culture-positive or -negative flocks, usually combined with a structured questionnaire. In addition, some studies, either in combination or independently, undertook intervention trials. Many of these studies were compromised by poor design, sampling, and statistical analysis. The evidence for each potential source and route of transmission on the poultry farm was reviewed critically, and the options for intervention were considered. The review concluded that, in most instances, biosecurity on conventional broiler farms can be enhanced and this should contribute to the reduction of flock colonization. However, complementary, non-biosecurity-based approaches will also be required in the future to maximize the reduction of Campylobacter-positive flocks at the farm level.

High-throughput screening has made a significant impact on drug discovery, but there is an acknowledged need for quantitative methods to analyze screening results and predict the activity of further compounds. In this paper we introduce one such method, binary kernel discrimination, and investigate its performance on two datasets; the first is a set of 1650 monoamine oxidase inhibitors, and the second a set of 101 437 compounds from an in-house enzyme assay. We compare the performance of binary kernel discrimination with a simple procedure which we call "merged similarity search", and also with a feedforward neural network. Binary kernel discrimination is shown to perform robustly with varying quantities of training data and also in the presence of noisy data. We conclude by highlighting the importance of the judicious use of general pattern recognition techniques for compound selection.