Confidence in Classification: A Bayesian Approach

Bayesian classification is currently of considerable interest. It provides a strategy for eliminating the uncertainty associated with a particular choice of classifiermodel parameters, and is the optimal decision-theoretic choice under certain circumstances when there is no single “true” classifier for a given data set. Modern computing capabilities can easily support the Markov chain Monte Carlo sampling that is necessary to carry out the calculations involved, but the information available in these samples is not at present being fully utilised. We show how it can be allied to known results concerning the “reject option” in order to produce an assessment of the confidence that can be ascribed to particular classifications, and how these confidence measures can be used to compare the performances of classifiers. Incorporating these confidence measures can alter the apparent ranking of classifiers as given by straightforward success or error rates. Several possible methods for obtaining confidence assessments are described, and compared on a range of data sets using the Bayesian probabilistic nearest-neighbour classifier.     

Serotonergic and cholinergic interaction in the regulation of pituitary-adrenal function in rats

It has been proposed that the central serotonergic inputs which modulate pituitary-adrenal secretion are mediated by cholinergic neurons. We have tested this hypothesis in intact rats. Male Sprague-Dawley rats were injected with cholinergic and serotonergic agents which enhanced transmitter function and with receptor blocking agents. Agents were injected, singly and in combination, into both unstressed and stressed animals. Since the response to cholinergic agents might be due to changes to vasopressin release, Brattleboro (vasopressin deficient) rats were also injected with cholinergic agents. The level of plasma corticosterone at 1-h post-injection was determined. Results indicate that the serotonin receptor blockade decreased the stimulatory, cholinergic effect of physostigmine. Cholinergic receptor blockers did not significantly reduce the corticosterone rise induced by 5-hydroxytryptophan. These results do not support the hypothesis of cholinergic mediation of serotonergic input. Nicotinic and muscarinic receptors appeared to exert opposing influences on the system. The nicotinic receptor antagonist was able to block the stimulatory effect of physostigmine. The muscarinic receptor antagonist significantly elevated plasma corticosterone levels. No differences were found in the effect of physostigmine on Brattleboro rats as compared to controls. These data are interpreted as suggesting that 1) the acetylcholine-induced stimulation of pituitary-adrenal function is mediated, in part, by serotonergic neurons; and 2) stimulation of nicotinic receptors is facilitatory whereas stimulation of muscarinic receptors is inhibitory to pituitary-adrenal function.     

Travelling waves in the occurrence of dengue haemorrhagic fever in Thailand

Dengue fever is a mosquito-borne virus that infects 50-100 million people each year. Of these infections, 200,000-500,000 occur as the severe, life-threatening form of the disease, dengue haemorrhagic fever (DHF). Large, unanticipated epidemics of DHF often overwhelm health systems. An understanding of the spatial-temporal pattern of DHF incidence would aid the allocation of resources to combat these epidemics. Here we examine the spatial-temporal dynamics of DHF incidence in a data set describing 850,000 infections occurring in 72 provinces of Thailand during the period 1983 to 1997. We use the method of empirical mode decomposition to show the existence of a spatial-temporal travelling wave in the incidence of DHF. We observe this wave in a three-year periodic component of variance, which is thought to reflect host-pathogen population dynamics. The wave emanates from Bangkok, the largest city in Thailand, moving radially at a speed of 148 km per month. This finding provides an important starting point for detecting and characterizing the key processes that contribute to the spatial-temporal dynamics of DHF in Thailand.     

A putative RUNX1 binding site variant between SLC9A3R1 and NAT9 is associated with susceptibility to psoriasis

Psoriasis (OMIM 177900) is a chronic inflammatory skin disorder of unknown pathogenesis affecting approximately 2% of the Western population. It occurs more frequently in individuals with human immunodeficiency virus, and 20-30% of individuals with psoriasis have psoriatic arthritis. Psoriasis is associated with HLA class I alleles, and previous linkage analysis by our group identified a second psoriasis locus at 17q24-q25 (PSORS2; ref. 7). Linkage to this locus was confirmed with independent family sets. Additional loci have also been proposed to be associated with psoriasis. Here we describe two peaks of strong association with psoriasis on chromosome 17q25 separated by 6 Mb. Associated single-nucleotide polymorphisms (SNPs) in the proximal peak lie in or near SLC9A3R1 (also called EBP50 and NHERF1) and NAT9, a new member of the N-acetyltransferase family. SLC9A3R1 is a PDZ domain-containing phosphoprotein that associates with members of the ezrin-radixin-moesin family and is implicated in diverse aspects of epithelial membrane biology and immune synapse formation in T cells. The distal peak of association is in RAPTOR (p150 target of rapamycin (TOR)-scaffold protein containing WD-repeats). Expression of SLC9A3R1 is highest in the uppermost stratum Malpighi of psoriatic and normal skin and in inactive versus active T cells. A disease-associated SNP lying between SLC9A3R1 and NAT9 leads to loss of RUNX1 binding. This is the second example of loss of a RUNX1 binding site associated with susceptibility to an autoimmune disease. It also suggests defective regulation of SLC9A3R1 or NAT9 by RUNX1 as a susceptibility factor for psoriasis.     

纳米级超显微硬度法对镉变形行为的研究

利用纳米级超显微硬度压痕法研究了具有六方密排结构纯Ｃｄ(９９９９％)的应力应变及加工硬化行为·根据载荷压痕贯穿深度的测量,证明了在Ｃｄ的压痕变形中存在弹性、弹塑性和完全塑性三个阶段,与每个阶段相对应在其载荷压痕深度曲线上都出现一个斜率的改变,从而可以近似确定金属开始塑性变形的应力范围·其加工硬化指数是随着施加的最大载荷由５ｍＮ时的０１增加到４０ｍＮ时的０３,这主要是和加载的速度有关并满足线性方程·     

Elucidating the principles of the molecular organization of heteropolymeric tight junction strands

Paracellular barrier properties of tissues are mainly determined by the composition of claudin heteropolymers. To analyze the molecular organization of tight junctions (TJ), we investigated the ability of claudins (Cld) to form homo- and heteromers. Cld1, -2, -3, -5, and -12 expressed in cerebral barriers were investigated. TJ-strands were reconstituted by claudin-transfection of HEK293-cells. cis-Interactions and/or spatial proximity were analyzed by fluorescence resonance energy transfer inside and outside of strands and ranked: Cld5/Cld5?>?Cld5/Cld1?>?Cld3/Cld1?>?Cld3/Cld3?>?Cld3/Cld5, no Cld3/Cld2. Classic Cld1, -3, and -5 but not non-classic Cld12 showed homophilic trans-interaction. Freeze-fracture electron microscopy revealed that, in contrast to classic claudins, YFP-tagged Cld12 does not form homopolymers. Heterophilic trans-interactions were analyzed in cocultures of differently monotransfected cells. trans-Interaction of Cld3/Cld5 was less pronounced than that of Cld3/Cld1, Cld5/Cld1, Cld5/Cld5 or Cld3/Cld3. The barrier function of reconstituted TJ-strands was demonstrated by a novel imaging assay. A model of the molecular organization of TJ was generated.     

The Forecasting Performance of a Finite Mixture Regime‐Switching Model for Daily Electricity Prices

Forecasting prices in electricity markets is a crucial activity for both risk management and asset optimization. Intra‐day power prices have a fine structure and are driven by an interaction of fundamental, behavioural and stochastic factors. Furthermore, there are reasons to expect the functional forms of price formation to be nonlinear in these factors and therefore specifying forecasting models that perform well out‐of‐sample is methodologically challenging. Markov regime switching has been widely advocated to capture some aspects of the nonlinearity, but it may suffer from overfitting and unobservability in the underlying states. In this paper we compare several extensions and alternative regime‐switching formulations, including logistic specifications of the underlying states, logistic smooth transition and finite mixture regression. The finite mixture approach to regime switching performs well in an extensive, out‐of‐sample forecasting comparison. Copyright © 2014 John Wiley & Sons, Ltd.     

Disruption of P2RY5, an orphan G protein-coupled receptor, underlies autosomal recessive woolly hair

The genetic determinants of hair texture in humans are largely unknown. Several human syndromes exist in which woolly hair comprises a part of the phenotype; however, simple autosomal recessive inheritance of isolated woolly hair has only rarely been reported. To identify a gene involved in controlling hair texture, we performed genetic linkage analysis in six families of Pakistani origin with autosomal recessive woolly hair (ARWH; OMIM 278150). All six families showed linkage to chromosome 13q14.2-14.3 (Z = 17.97). In all cases, we discovered pathogenic mutations in P2RY5, which encodes a G protein-coupled receptor and is a nested gene residing within intron 17 of the retinoblastoma 1 (RB1) gene. P2RY5 is expressed in both Henle's and Huxley's layers of the inner root sheath of the hair follicle. Our findings indicate that disruption of P2RY5 underlies ARWH and, more broadly, uncover a new gene involved in determining hair texture in humans.