Dynamic Latent Class Model Averaging for Online Prediction

We consider the problem of online prediction when it is uncertain what the best prediction model to use is. We develop a method called dynamic latent class model averaging, which combines a state‐space model for the parameters of each of the candidate models of the system with a Markov chain model for the best model. We propose a polychotomous regression model for the transition weights to assume that the probability of a change in time depends on the past through the values of the most recent time periods and spatial correlation among the regions. The evolution of the parameters in each submodel is defined by exponential forgetting. This structure allows the ‘correct’ model to vary over both time and regions. In contrast to existing methods, the proposed model naturally incorporates clustering and prediction analysis in a single unified framework. We develop an efficient Gibbs algorithm for computation, and we demonstrate the value of our framework on simulated experiments and on a real‐world problem: forecasting IBM's corporate revenue. Copyright © 2014 John Wiley & Sons, Ltd.     

Identification of Tim4 as a phosphatidylserine receptor

In programmed cell death, a large number of cells undergo apoptosis, and are engulfed by macrophages to avoid the release of noxious materials from the dying cells. In definitive erythropoiesis, nuclei are expelled from erythroid precursor cells and are engulfed by macrophages. Phosphatidylserine is exposed on the surface of apoptotic cells and on the nuclei expelled from erythroid precursor cells; it works as an 'eat me' signal for phagocytes. Phosphatidylserine is also expressed on the surface of exosomes involved in intercellular signalling. Here we established a library of hamster monoclonal antibodies against mouse peritoneal macrophages, and found an antibody that strongly inhibited the phosphatidylserine-dependent engulfment of apoptotic cells. The antigen recognized by the antibody was identified by expression cloning as a type I transmembrane protein called Tim4 (T-cell immunoglobulin- and mucin-domain-containing molecule; also known as Timd4). Tim4 was expressed in Mac1+ cells in various mouse tissues, including spleen, lymph nodes and fetal liver. Tim4 bound apoptotic cells by recognizing phosphatidylserine via its immunoglobulin domain. The expression of Tim4 in fibroblasts enhanced their ability to engulf apoptotic cells. When the anti-Tim4 monoclonal antibody was administered into mice, the engulfment of apoptotic cells by thymic macrophages was significantly blocked, and the mice developed autoantibodies. Among the other Tim family members, Tim1, but neither Tim2 nor Tim3, specifically bound phosphatidylserine. Tim1- or Tim4-expressing Ba/F3 B cells were bound by exosomes via phosphatidylserine, and exosomes stimulated the interaction between Tim1 and Tim4. These results indicate that Tim4 and Tim1 are phosphatidylserine receptors for the engulfment of apoptotic cells, and may also be involved in intercellular signalling in which exosomes are involved.     

零下温度时二维通道内界面活性剂减阻流动的实验研究

为了进行零下温度时区域供冷系统的界面活性剂减阻设计,测量了-5℃时以质量分数为20%的甘醇不冻液为溶剂的界面活性剂溶液油醇基二羟乙基甲基氯化铵在二维通道内的减阻特性,同时用粒子成像测速仪对减阻流动的湍流特性进行了测量.实验发现:质量分数为5×10-5~5×10-4的低浓度油醇基二羟乙基甲基氯化铵界面活性剂溶液在一定的雷诺数范围内均呈现减阻特性,且其减阻特性与质量分数和温度呈复杂关系,最大减阻值可高达70%以上;在湍流液体中只需加入少量该界面活性剂即可大幅度减少液体输送的动力消耗.与单独的质量分数为20%的甘醇不冻液湍流流动相比,添加界面活性剂抑制了湍流速度脉动,使雷诺应力完全消失.     

Vegetation Spatial Heterogeneity of Different Soil Regions in Inner Mongolia, China

The vegetation spatial heterogeneity and ecological characteristics in different soil regions were analyzed by surveying the vegetation in 12 different soil regions of Inner Mongolia, China, including conifer- ous-broadleaf deciduous forests, shrub, grassland, and desert regions with 1122 large 2 cm × 2 cm quad- rats (actual size 30 km × 30 km, referred to as L-quadrat hereafter) in about 1.18 million km2. Each L- quadrat was divided into four small 1 cm × 1 cm quadrats (actual size 15 km × 15 km, S-quadrat). The vegetation was analyzed based on the beta-binomial distribution to describe the frequency of occurrence and spatial heterogeneity for each kind of vegetation. The weighted average of the heterogeneity of all vegeta- tion in the same soil region provides a measure of the soil regional landscape level heterogeneity which de- scribes the spatial complexity of the regional landscape composition of the existing vegetation. Comparison of the vegetation characteristics in the 12 soil regions shows that, the calcic gray soil has the highest average vegetation type per quadrat. The largest soil region is calcic chestnut soil and has the most vegetation types. Every soil region has its own dominant vegetation sequence which dominates in occurrence and dominant vegetation types which dominates in spatial heterogeneity. For the Inner Mongolian vegetation, the weighted average of the heterogeneity is 0.60 and the vegetation diversity index is 4.47.     

Nuclear cataract caused by a lack of DNA degradation in the mouse eye lens

The eye lens is composed of fibre cells, which develop from the epithelial cells on the anterior surface of the lens. Differentiation into a lens fibre cell is accompanied by changes in cell shape, the expression of crystallins and the degradation of cellular organelles. The loss of organelles is believed to ensure the transparency of the lens, but the molecular mechanism behind this process is not known. Here we show that DLAD ('DNase II-like acid DNase', also called DNase IIbeta) is expressed in human and murine lens cells, and that mice deficient in the DLAD gene are incapable of degrading DNA during lens cell differentiation--the undigested DNA accumulates in the fibre cells. The DLAD-/- mice develop cataracts of the nucleus lentis, and their response to light on electroretinograms is severely reduced. These results indicate that DLAD is responsible for the degradation of nuclear DNA during lens cell differentiation, and that if DNA is left undigested in the lens, it causes cataracts of the nucleus lentis, blocking the light path.     

Somatic diversification of variable lymphocyte receptors in the agnathan sea lamprey

Although jawless vertebrates are apparently capable of adaptive immune responses, they have not been found to possess the recombinatorial antigen receptors shared by all jawed vertebrates. Our search for the phylogenetic roots of adaptive immunity in the lamprey has instead identified a new type of variable lymphocyte receptors (VLRs) composed of highly diverse leucine-rich repeats (LRR) sandwiched between amino- and carboxy-terminal LRRs. An invariant stalk region tethers the VLRs to the cell surface by means of a glycosyl-phosphatidyl-inositol anchor. To generate rearranged VLR genes of the diversity necessary for an anticipatory immune system, the single lamprey VLR locus contains a large bank of diverse LRR cassettes, available for insertion into an incomplete germline VLR gene. Individual lymphocytes express a uniquely rearranged VLR gene in monoallelic fashion. Different evolutionary strategies were thus used to generate highly diverse lymphocyte receptors through rearrangement of LRR modules in agnathans (jawless fish) and of immunoglobulin gene segments in gnathostomes (jawed vertebrates).     

The genomic basis of adaptive evolution in threespine sticklebacks

Marine stickleback fish have colonized and adapted to thousands of streams and lakes formed since the last ice age, providing an exceptional opportunity to characterize genomic mechanisms underlying repeated ecological adaptation in nature. Here we develop a high-quality reference genome assembly for threespine sticklebacks. By sequencing the genomes of twenty additional individuals from a global set of marine and freshwater populations, we identify a genome-wide set of loci that are consistently associated with marine-freshwater divergence. Our results indicate that reuse of globally shared standing genetic variation, including chromosomal inversions, has an important role in repeated evolution of distinct marine and freshwater sticklebacks, and in the maintenance of divergent ecotypes during early stages of reproductive isolation. Both coding and regulatory changes occur in the set of loci underlying marine-freshwater evolution, but regulatory changes appear to predominate in this well known example of repeated adaptive evolution in nature.     

Experimental adaptation of an influenza H5 HA confers respiratory droplet transmission to a reassortant H5 HA/H1N1 virus in ferrets

Highly pathogenic avian H5N1 influenza A viruses occasionally infect humans, but currently do not transmit efficiently among humans. The viral haemagglutinin (HA) protein is a known host-range determinant as it mediates virus binding to host-specific cellular receptors. Here we assess the molecular changes in HA that would allow a virus possessing subtype H5 HA to be transmissible among mammals. We identified a reassortant H5 HA/H1N1 virus-comprising H5 HA (from an H5N1 virus) with four mutations and the remaining seven gene segments from a 2009 pandemic H1N1 virus-that was capable of droplet transmission in a ferret model. The transmissible H5 reassortant virus preferentially recognized human-type receptors, replicated efficiently in ferrets, caused lung lesions and weight loss, but was not highly pathogenic and did not cause mortality. These results indicate that H5 HA can convert to an HA that supports efficient viral transmission in mammals; however, we do not know whether the four mutations in the H5 HA identified here would render a wholly avian H5N1 virus transmissible. The genetic origin of the remaining seven viral gene segments may also critically contribute to transmissibility in mammals. Nevertheless, as H5N1 viruses continue to evolve and infect humans, receptor-binding variants of H5N1 viruses with pandemic potential, including avian-human reassortant viruses as tested here, may emerge. Our findings emphasize the need to prepare for potential pandemics caused by influenza viruses possessing H5 HA, and will help individuals conducting surveillance in regions with circulating H5N1 viruses to recognize key residues that predict the pandemic potential of isolates, which will inform the development, production and distribution of effective countermeasures.     

Hierarchical hypergames and Bayesian games: A generalization of the theoretical comparison of hypergames and Bayesian games considering hierarchy of perceptions

This paper discusses the relationship of two independently developed models of games with incomplete information, hierarchical hypergames and Bayesian games. It can be considered as a generalization of the previous study on the theoretical comparison of simple hypergames and Bayesian games (Sasaki and Kijima, 2012) by taking into account hierarchy of perceptions, i.e., an agent’s perception about the other agents’ perceptions, and so on. The authors first introduce the general way of transformation of any hierarchical hypergames into corresponding Bayesian games, which was called as the Bayesian representation of hierarchical hypergames. The authors then show that some equilibrium concepts for hierarchical hypergames can be associated with those for Bayesian games and discuss implications of the results.