Creating a spontaneous conversational speech corpus

Speech recognition and language analysis of spontaneous speech arising in naturally spoken conversations are becoming the subject of much research. However, there is a shortage of spontaneous speech corpora that are freely available for academics. We therefore undertook the building of a natural conversation speech database, recording over 200 hours of conversations in English by over 600 local university students. With few exceptions, the students used their own cell phones from their own rooms or homes to speak to one another, and they were permitted to speak on any topic they chose. Although they knew that they were being recorded and that they would receive a small payment, their conversations in the corpus are probably very close to being natural and spontaneous. This paper describes a detailed case study of the problems we faced and the methods we used to make the recordings and control the collection of these social science data on a limited budget.

EFFECT OF EXERCISE ON THE NASAL TRANSMUCOSAL POTENTIAL DIFFERENCE IN PATIENTS WITH CYSTIC-FIBROSIS AND NORMAL SUBJECTS

Background - Normal subjects have a negative nasal transmucosal potential difference (TPD) at rest which becomes more negative with exercise. Patients with cystic fibrosis have a more negative resting nasal TPD than controls. The present study was designed to determine the effects of exercise on the TPD of patients with cystic fibrosis.

Post-mortem examination of a wild muntjac from Northern Ireland.

1. Deer are of major concern with regards to impacts on biodiversity, forestry and agriculture as well as human health. The invasive Reeves’ muntjac deer Muntiacus reevesi, native to Asia but established in Great Britain, has recently appeared in the wild in Ireland.

2. The first verified record in the wild in Northern Ireland was confirmed during June 2009 as a result of a road traffic accident near Newtownards, Co. Down. The second record was a culled animal shot in the grounds of Mount Stewart, Co. Down during June 2011.

3. The current report aimed to perform a detailed investigation of the most recently obtained animal to establish a) its age, b) its genetic relationship with the first animal and c) the threat it might pose in terms of carrying endo- or ecto-parasites or other micro-pathogens, principally viruses and bacteria.

4. Analysis of its dentition suggested the culled animal was 56 weeks old (range 55-57 weeks).

5. Genetic analysis indicated that there was no possibility of a father-son relationship with the buck killed in 2009. There was a 6.25% probability of both individuals being full-sibs and a 25% probability of a half-sib relationship.

6. The deer was free from all the major pathogens. Most significantly, a novel species of Gammaherpesvirus was detected most closely matching type 2 ruminant rhadinovirus (Gammaherpesvirinae) from mule deer. Further sequencing is required to provide a definitive classification. There is no evidence suggesting this virus is pathogenic but its detection is nonetheless of concern.

7. During late 2009 (after the first animal was recovered) and early 2010, there were a number of sightings of muntjac within the vicinity of Mount Stewart. Our results indicate that the animal shot in June 2011 could not have been this same animal, as the shot animal must have been born in spring 2010. Moreover, genetic analysis indicates that the two recovered individuals were highly unlikely to share the same mother and father, suggesting they were the offspring of a minimum of three breeding adults (two fathers and one mother or two mothers and one father). This brings the total number of known individuals to 5 including the two offspring. The location of the breeding adults is unknown and may be either in captivity with subsequent escapees, or deliberate releases, or be free living in the wild. However, a further sighting during early 2012 may suggest a wild origin.

Flow cytometric analysis of within-strain variation in polysaccharide expression by Bacteroides fragilis by use of murine monoclonal antibodies.

The reactivity of four different monoclonal antibodies (MAbs) with populations of Bacteroides fragilis NCTC 9343, enriched by density gradient centrifugation for a large capsule, small capsule and electron-dense layer (EDL) only visible by electronmicroscopy, was examined. The MAbs reacted strongly with polysaccharides present in both the large capsule- and EDL-enriched populations but not in the small capsule-enriched populations. The pattern of labelling was determined by immunoblotting, immunofluorescence and immuno-electronmicroscopy, and flow cytometry. The MAbs labelled cell membrane-associated epitopes in the large capsule- and EDL-enriched populations and cell-free material in the EDL population. By immunoblotting, ladders of repeating polysaccharide subunits were evident in the EDL population but not in the large capsule population. The proportion of cells labelled within each population was determined by flow cytometry. The reactivity of another MAb with the small capsule population was confirmed by flow cytometry. A qualitative indication of epitope expression was obtained by examination of the flow cytometric profiles. Differential expression of the same saccharide epitope was observed both between and within structurally distinct B. fragilis populations. The MAbs were species-specific and cross-reacted with several recent clinical isolates. These polysaccharides may be relevant to the virulence of B. fragilis.

Immunological detection of Bacteroides fragilis in clinical-samples

A monospecific polyclonal antiserum, prepared against Bacteroides fragilis common polysaccharide antigen purified by polyacrylamide gel immunoblot detected B. fragilis, B. thetaiotaomicron, B. ovatus and Prevotella melaninogenica in pus samples from various anatomical sites by immunofluorescence microscopy of the pus. With standard clinical laboratory culture methods, 36% of 147 samples were positive for one or more of the above bacteria. Of these, B. fragilis accounted for 33%. By immunofluorescent labelling of pus with the common antigen antiserum the detection of these bacteria in the samples increased to 50%. All nine of the blood cultures in which B. fragilis was detected by culture contained bacteria positive for the common antigen. Immunofluorescent labelling of pus samples with a selection of monoclonal antibodies specific for surface polysaccharides which are known to be antigenically variable in culture in vitro and in an animal model of infection showed that these polysaccharides are also variable in natural infection. The results indicate that the common polysaccharide antigen, in contrast to the variable surface polysaccharides, is a suitable target for the immunodetection of B, fragilis in clinical samples from a range of anatomical sites.