Hidden Conditional Random Fields for Visual Speech Recognition

In this paper we present the application of Hidden Conditional Random Fields (HCRFs) to modelling speech for visual speech recognition. HCRFs may be easily adapted to model long range dependencies across an observation sequence. As a result visual word recognition performance can be improved as the model is able to take more of a contextual approach to generating state sequences. Results are presented from a speaker-dependent, isolated digit, visual speech recognition task using comparisons with a baseline HMM system. We firstly illustrate that word recognition rates on clean video using HCRFs can be improved by increasing the number of past and future observations being taken into account by each state. Secondly we compare model performances using various levels of video compression on the test set. As far as we are aware this is the first attempted use of HCRFs for visual speech recognition.

Spatial genetic structuring in a vagile species, the European wood mouse

We examined the genetic structure of natural populations of the European wood mouse Apodemus sylvaticus at the microgeographic ( 30 km) scales. Ecological and behavioural studies indicate that this species exhibits considerable dispersal relative to its home-range size. Thus, there is potential for high gene flow over larger geographic areas. As levels of population genetic structure are related to gene flow, we hypothesized that population genetic structuring at the microgeographic level should be negligible, increasing only with geographic distance. To test this, four sites were sampled within a microgeographic scale with two additional samples at the macrogeographic level. Individuals (n=415) were screened and analysed for seven polymorphic microsatellite loci. Contrary to our hypothesis, significant levels of population structuring were detected at both scales. Comparing genetic differentiation with geographic distance suggests increasing genetic isolation with distance. However, this distance effect was non-significant being confounded by surprisingly high levels of differentiation among microgeographic samples. We attribute this pattern of genetic differentiation to the effect of habitat fragmentation, splitting large populations into components with small effective population sizes resulting in enhanced genetic drift. Our results indicate that it is incorrect to assume genetic homogeneity among populations even where there is no evidence of physical barriers and dispersal can occur freely. In the case of A. sylvaticus, it is not clear whether dispersal does not occur across habitat barriers or behavioural dispersal occurs without consequent gene flow.

Prediction of microRNAs affecting mRNA expression during retinal development.

Background: MicroRNAs (miRNAs) are small RNA molecules (similar to 22 nucleotides) which have been shown to play an important role both in development and in maintenance of adult tissue. Conditional inactivation of miRNAs in the eye causes loss of visual function and progressive retinal degeneration. In addition to inhibiting translation, miRNAs can mediate degradation of targeted mRNAs. We have previously shown that candidate miRNAs affecting transcript levels in a tissue can be deduced from mRNA microarray expression profiles. The purpose of this study was to predict miRNAs which affect mRNA levels in developing and adult retinal tissue and to confirm their expression.

Results: Microarray expression data from ciliary epithelial retinal stem cells (CE-RSCs), developing and adult mouse retina were generated or downloaded from public repositories. Analysis of gene expression profiles detected the effects of multiple miRNAs in CE-RSCs and retina. The expression of 20 selected miRNAs was confirmed by RT-PCR and the cellular distribution of representative candidates analyzed by in situ hybridization. The expression levels of miRNAs correlated with the significance of their predicted effects upon mRNA expression. Highly expressed miRNAs included miR-124, miR-125a, miR-125b, miR-204 and miR-9. Over-expression of three miRNAs with significant predicted effects upon global mRNA levels resulted in a decrease in mRNA expression of five out of six individual predicted target genes assayed.

Conclusions: This study has detected the effect of miRNAs upon mRNA expression in immature and adult retinal tissue and cells. The validity of these observations is supported by the experimental confirmation of candidate miRNA expression and the regulation of predicted target genes following miRNA over-expression. Identified miRNAs are likely to be important in retinal development and function. Misregulation of these miRNAs might contribute to retinal degeneration and disease. Conversely, manipulation of their expression could potentially be used as a therapeutic tool in the future.

Priming of CD8+ and CD4+ T Cells in Experimental Leishmaniasis Is Initiated by Different Dendritic Cell Subtypes

The biological role of Langerin(+) dendritic cells (DCs) such as Langerhans cells and a subset of dermal DCs (dDCs) in adaptive immunity against cutaneous pathogens remains enigmatic. Thus, we analyzed the impact of Langerin(+) DCs in adaptive T cell-mediated immunity toward Leishmania major parasites in a Lang-DTR mouse model that allows conditional diphtheria toxin (DT)-induced ablation of The biological role of Langerin+ dendritic cells (DCs) such as Langerhans cells and a subset of dermal DCs (dDCs) in adaptive immunity against cutaneous pathogens remains enigmatic. Thus, we analyzed the impact of Langerin+ DCs in adaptive T cell-mediated immunity toward Leishmania major parasites in a Lang-DTR mouse model that allows conditional diphtheria toxin (DT)-induced ablation of Langerin+ DCs in vivo. For the first time, infection experiments with DT-treated Lang-DTR mice revealed that proliferation of L. major-specific CD8+ T cells is significantly reduced during the early phase of the immune response following depletion of Langerin+ DCs. Consequently, the total number of activated CD8+ T cells within the draining lymph node and at the site of infection is diminished. Furthermore, we show that the impaired CD8+ T cell response is due to the absence of Langerin+ dDCs and not Langerhans cells. Nevertheless, the CD4+ T cell response is not altered and the infection is cleared as effectively in DT-treated Lang-DTR mice as in control mice. This clearly demonstrates that Langerin+ DCs are, in general, dispensable for an efficient adaptive immune response against L. major parasites. Thus, we propose a novel concept that, in the experimental model of leishmaniasis, priming of CD4+ T cells is mediated by Langerin− dDCs, whereas Langerin+ dDCs are involved in early priming of CD8+ T cells.

Persistent measles virus infection of mouse neural cells lacking known human entry receptors

Aims: Infection of the mouse central nervous system with wild type (WT) and vaccine strains of measles virus (MV) results in lack of clinical signs and limited antigen detection. It is considered that cell entry receptors for these viruses are not present on murine neural cells and infection is restricted at cell entry.

Methods: To examine this hypothesis, virus antigen and caspase 3 expression (for apoptosis) was compared in primary mixed, neural cell cultures infected in vitro or prepared from mice infected intracerebrally with WT, vaccine or rodent neuroadapted viruses. Viral RNA levels were examined in mouse brain by nested and real-time reverse transcriptase polymerase chain reaction.

Results: WT and vaccine strains were demonstrated for the first time to infect murine oligodendrocytes in addition to neurones despite a lack of the known MV cell receptors. Unexpectedly, the percentage of cells positive for viral antigen was higher for WT MV than neuroadapted virus in both in vitro and ex vivo cultures. In the latter the percentage of positive cells increased with time after mouse infection. Viral RNA (total and mRNA) was detected in brain for up to 20 days, while cultures were negative for caspase 3 in WT and vaccine virus infections.

Conclusions: WT and vaccine MV strains can use an endogenous cell entry receptor(s) or alternative virus uptake mechanism in murine neural cells. However, viral replication occurs at a low level and is associated with limited apoptosis. WT MV mouse infection may provide a model for the initial stages of persistent MV human central nervous system infections.

Isolation and culture of mouse intestinal cells.

Complex cell signal transduction mechanisms regulate intestinal epithelial shape, polarity, motility, organelles, cell membrane components as well as physical and mechanical properties to influence alimentary digestion, absorption, secretion, detoxification and fluid balance. Interactions between the epithelial cells and adjacent mesenchyme are central to intestinal homeostasis although the key regulatory molecules of specific differentiation steps remain unclear. Isolation and primary culture of heterotypic murine intestinal cells provides a model system for elucidation of essential molecular cross-talk between epithelium and mesenchyme that may provide several biological and practical advantages over transformed cell lines. An in vitro primary culture system for neonatal rat or mouse intestinal cells has been established that forms monolayers, expresses intestine-specific epithelial features including intestinal brush borders and appropriate hydrolase enzymes. Our studies confirm the promise of this method which may advance our understanding of heterotypic cellular interactions implicated in intestinal function and may provide important insights into the pathobiology of disease.