Evaluating the effectiveness of Bayesian feature selection

A practical Bayesian approach for inference in neural network models has been available for ten years, and yet it is not used frequently in medical applications. In this chapter we show how both regularisation and feature selection can bring significant benefits in diagnostic tasks through two case studies: heart arrhythmia classification based on ECG data and the prognosis of lupus. In the first of these, the number of variables was reduced by two thirds without significantly affecting performance, while in the second, only the Bayesian models had an acceptable accuracy. In both tasks, neural networks outperformed other pattern recognition approaches.

Semi-Smooth Newton Methods for the Time Optimal Control of Nonautonomous Ordinary Differential Equations

AMS Subj. Classification: 49J15, 49M15

The Frenchness of Marcel Lefebvre and the Society of St Pius X:a new reading

The case of Marcel Lefebvre and the SSPX deserves fresh perspectives. The current historiography is too franco-centric, obsessed with relatively minor matters, rather than with more substantial ones. This article proposes a new analysis of the SSPX’s political discourses in France and internationally over the last fifteen years and undertakes to reframe the relationship between Lefebvre’s life and his congregation by re-examining his African missionary experiences. Such new perspectives will be helpful as the SSPX moves towards regularisation under the pontificate of Pope Francis.

The standardisation of Scottish Gaelic orthography 1750-2007: a corpus approach

This thesis investigates the standardisation of Modern Scottish Gaelic orthography from the mid-eighteenth century to the twenty-first. It presents the results of the first corpus-based analysis of Modern Scottish Gaelic orthographic development combined with an analytic approach that places orthographic choices in their sociolinguistic context. The theoretical framework behind the analysis centres on discussion of how the language ideologies of the phonographic ideal, historicism, autonomy, vernacularism and the ideology of the standard itself have shaped orthographic conventions and debates. It argues that current spelling norms reflect an orthography that is the result of compromise, historical factors and pragmatic function. The research uses a digital corpus to examine how three particular features have been used over time: the dialect variation between <eu> and <ia>; variation in s + stop consonant clusters (sd/st, sg/sc, sb/sp); and the use of the grave and acute accents. Evidence is drawn from the Corpas na Gàidhlig electronic corpus created at the University of Glasgow: the sub-corpus used in this study includes 117 published texts representing a period of over 250 years from 1750 to 2007, and a total size of over four and a quarter million words. The results confirm a key period of reform between 1750 and the early nineteenth century, and thereafter a settled norm being established in the early nineteenth century. Since then, some variation has been acceptable although changes and reform of some features have centred on increasing uniformity and regularisation.

Sparse hierarchical Bayesian models for detecting relevant antigenic sites in virus evolution

Understanding how virus strains offer protection against closely related emerging strains is vital for creating effective vaccines. For many viruses, including Foot-and-Mouth Disease Virus (FMDV) and the Influenza virus where multiple serotypes often co-circulate, in vitro testing of large numbers of vaccines can be infeasible. Therefore the development of an in silico predictor of cross-protection between strains is important to help optimise vaccine choice. Vaccines will offer cross-protection against closely related strains, but not against those that are antigenically distinct. To be able to predict cross-protection we must understand the antigenic variability within a virus serotype, distinct lineages of a virus, and identify the antigenic residues and evolutionary changes that cause the variability. In this thesis we present a family of sparse hierarchical Bayesian models for detecting relevant antigenic sites in virus evolution (SABRE), as well as an extended version of the method, the extended SABRE (eSABRE) method, which better takes into account the data collection process. The SABRE methods are a family of sparse Bayesian hierarchical models that use spike and slab priors to identify sites in the viral protein which are important for the neutralisation of the virus. In this thesis we demonstrate how the SABRE methods can be used to identify antigenic residues within different serotypes and show how the SABRE method outperforms established methods, mixed-effects models based on forward variable selection or l1 regularisation, on both synthetic and viral datasets. In addition we also test a number of different versions of the SABRE method, compare conjugate and semi-conjugate prior specifications and an alternative to the spike and slab prior; the binary mask model. We also propose novel proposal mechanisms for the Markov chain Monte Carlo (MCMC) simulations, which improve mixing and convergence over that of the established component-wise Gibbs sampler. The SABRE method is then applied to datasets from FMDV and the Influenza virus in order to identify a number of known antigenic residue and to provide hypotheses of other potentially antigenic residues. We also demonstrate how the SABRE methods can be used to create accurate predictions of the important evolutionary changes of the FMDV serotypes. In this thesis we provide an extended version of the SABRE method, the eSABRE method, based on a latent variable model. The eSABRE method takes further into account the structure of the datasets for FMDV and the Influenza virus through the latent variable model and gives an improvement in the modelling of the error. We show how the eSABRE method outperforms the SABRE methods in simulation studies and propose a new information criterion for selecting the random effects factors that should be included in the eSABRE method; block integrated Widely Applicable Information Criterion (biWAIC). We demonstrate how biWAIC performs equally to two other methods for selecting the random effects factors and combine it with the eSABRE method to apply it to two large Influenza datasets. Inference in these large datasets is computationally infeasible with the SABRE methods, but as a result of the improved structure of the likelihood, we are able to show how the eSABRE method offers a computational improvement, leading it to be used on these datasets. The results of the eSABRE method show that we can use the method in a fully automatic manner to identify a large number of antigenic residues on a variety of the antigenic sites of two Influenza serotypes, as well as making predictions of a number of nearby sites that may also be antigenic and are worthy of further experiment investigation.