Selecting high-dimensional mixed graphical models using minimal AIC or BIC forests

Background  

Chow and Liu showed that the maximum likelihood tree for multivariate discrete distributions may be found using a maximum weight spanning tree algorithm, for example Kruskal's algorithm. The efficiency of the algorithm makes it tractable for high-dimensional problems.     

The nuclear pores of early meiotic prophase nuclei of plants

Summary Pollen mother cells at early meiotic prophase fromFritillaria lanceolata, F. mutica, Tulbaghia violacea, the lily “Formobel”,Triticum aegilopoides, T. dicoccoides, T. aestivum and synaptic and asynaptic forms ofT. durum were studied in thin sections with the electron microscope (a) in relation to distribution of nuclear pores (b) in respect of fine structure of the pore complex in those of the first four. The pores were distributed in random clusters during leptotene to pachytene in all plants, except in the two forms ofT. durum where there were either no pores or so few that they were not detectable. Probably correlated with this, the two membranes of the nuclear envelope were often widely separated and frequently sacculated. No pores were seen at leptotene in the part of the envelope to which, in theFritillarias and lily, the nucleolus was adpressed at this time. Evidence supporting a recent model which proposes that annuli are composed of three rings of eight granular subunits was obtained. These subunits as well as a dense central element, observed in most pores, were composed of filaments about 3 nm in diameter and evidently protein in character. There was evidence of a continuity between filaments in the central element and those in the rings of subunits which encircle the pore aperture at both the nuclear and cytoplasmic sides of the pore. In profiles of pores knobbed filaments were sometimes seen extending laterally from the pore wall into the perinuclear space at two sides. Questions concerning the role of the annulus are discussed. The author wish to thank Mr. R. F. Scott for construction to the model.     

Habitat characteristics contributing to local occupancy and habitat use in rock pool <Emphasis Type="Italic">Daphnia</Emphasis> metapopulations

The effects of invasive macrophytes, water level fluctuations and predation on freshwater unionids Pyganodon grandis and Utterbackia imbecillis were studied in three small impoundments in Northeastern Texas in 2003–2005. Mussel density was sampled with quadrats. Mortality, associated with the water level fluctuations and predation, was estimated by collecting dead shells on the shore at about two month intervals. In two ponds, horizontal distribution of unionids was limited by dense beds of invasive and noxious macrophytes (mainly Eurasian watermilfoil Myriophyllum spicatum and American lotus Nelumbo lutea): mussel densities were significantly lower in these macrophyte beds (P < 0.001). In the third pond with the lowest density of macrophytes (stonewort Chara sp.), unionids were distributed more evenly, and the average unionid biomass was the highest among all ponds studied. Vertical distribution of unionids in all ponds was likely limited by low oxygen at depth >2 m. The total amount of shells found on the shore per year varied from 0.1% to 28% of the total population in the pond and was negatively correlated with water level (r = −0.72 to −0.81, P < 0.005). Mammalian predators consumed up to 19% of the total unionid population and predation was facilitated by water level fluctuations. Handling editor: K. Martens     

The effects of dispersal limitation and topographic heterogeneity on beta diversity and phylobetadiversity in a subtropical forest

We assessed the effects of topographic heterogeneity and stem density on species composition between grains of different sizes (20 × 20, 50 × 50, and 100 × 100 m), based on partial Mantel tests. Similarity in species composition was measured by the abundance-based Jaccard index (C_J) and by an index that incorporates phylogenetic information into C_J (pC_J). Plants were divided into five groups, arbor, subarbor, and shrub according to life form and two other groups: species that produce dry fruits (PDF) and that produce fleshy fruits (PFF). C_J and pC_J between any two grains at each grain size were calculated separately for these groups and for all species combined. In order to examine what influences C_J and pC_J, we analyzed their correlations with topographic heterogeneity variables and two dispersal limitation-related variables (stem and topographic resistance). Our data indicate that at all three grain sizes, C_J and pC_J decrease with increasing distance for all plant groups. Dispersal limitation and topographic heterogeneity were both important at 20 × 20 and 50 × 50 m grain sizes for C_J and pC_J of all plant groups; and at 100 × 100 m grain size, topographic heterogeneity dominates over dispersal limitation for some plant groups. C_J and pC_J of PDFs are less negatively correlated with stem resistance than those of PFFs. We conclude that both beta diversity and phylobetadiversity are dependent on plant groups and grain sizes.     

Extensive variation in synonymous substitution rates in mitochondrial genes of seed plants

Background  

It has long been known that rates of synonymous substitutions are unusually low in mitochondrial genes of flowering and other land plants. Although two dramatic exceptions to this pattern have recently been reported, it is unclear how often major increases in substitution rates occur during plant mitochondrial evolution and what the overall magnitude of substitution rate variation is across plants.