Covariance Partition Priors: A Bayesian Approach to Simultaneous Covariance Estimation for Longitudinal Data

The estimation of the covariance matrix is a key concern in the analysis of longitudinal data. When data consist of multiple groups, it is often assumed the covariance matrices are either equal across groups or are completely distinct. We seek methodology to allow borrowing of strength across potentially similar groups to improve estimation. To that end, we introduce a covariance partition prior that proposes a partition of the groups at each measurement time. Groups in the same set of the partition share dependence parameters for the distribution of the current measurement given the preceding ones, and the sequence of partitions is modeled as a Markov chain to encourage similar structure at nearby measurement times. This approach additionally encourages a lower-dimensional structure of the covariance matrices by shrinking the parameters of the Cholesky decomposition toward zero. We demonstrate the performance of our model through two simulation studies and the analysis of data from a depression study. This article includes Supplementary Materials available online.     

DOES LOW-INTENSITY He-Ne LASER RADIATION PRODUCE A PHOTOBIOLOGICAL GROWTH RESPONSE IN Escherichia coli?

Abstract A photobiological study was camed out on the bacterium Escherichia coli in order to determine whether stimulation of growth occurred after irradiation of an inoculum with coherent red light. No enhancement or inhibition of growth was observed for cultures of the bacterium following irradiation of inocula with a Helium-neon laser (continuous wave, λ= 632.8 nm) at irradiances of 7.7 × 10¹⁵ and 1.8 × 10¹⁶ photons cm⁻² s⁻¹ using fluences of 4.5 × 10^−-1 and 4.5 J cm⁻² at each irradiance. Bacterial growth in irradiated and control cultures was monitored during a growth period of ca 2 h using a viable count technique after inocula in the early exponential phase had been diluted with fresh growth medium. These results do not provide support for the work of Kam et al . (1983, Nuov. Cim . 2D, 1138–1144), and Tiphlova and Karu (1988, Photochem. Photobiol . 48 , 467–471), which appear to show substantial enhancement of E. coli growth under these conditions.     

Radiationless transition rates of thymine and uracil in neutral aqueous solution at 300°K

Recent experiments in low-intensity fluorescence spectrometry, flash photolysis and continuous photolysis at 300°K in an aqueous medium have provided data for uracil and thymine from which all the pertinent phenomenological characteristics (Φ_fl^0′, Φ_isc^0′, Φ′_chem) of the 0′ level can be evaluated. It is necessary to postulate considerable internal conversion ¹S → ⁰S with τ_ic ≈ 1–2 × 10⁻¹² sec, this process correlating with the dielectric relaxation time of water.     

EXCITED STATES OF DNA and ITS COMPONENTS AT ROOM TEMPERATURE–V. SPECTRAL, POLARIZATION and QUANTUM YIELD STUDIES OF CYTIDYLYL-(3',5')-ADENOSINE

Abstract— The corrected normalized emission spectrum, quantum yield, and emission anisotropy are reported for partially-stacked CpA excited at 266 nm. Utilizing parameters determined separately for cytidine and AMP, the emission spectrum of CpA is quantitatively resolved into components characteristic of the two monomers plus a red shifted emission characteristic of complex formation. The results fit accurately to a simple stacking model in which monomer-like fluorescence originates only from the unstacked fraction and "complex" fluorescence from the stacked fraction. It is shown that the stacked fraction absorbs in a manner qualitatively distinct from the unstacked fraction and the "complex" emission is really the excited state of a ground state complex. This emission, which peaks at 440 nm and contains at least two species can be resolved into three Gaussian components. It is suggested that the emitting species may originate in distinct stacked conformations.     

Kinetics and mechanism of the general-acid-catalyzed ring-closure of the malondialdehyde-DNA adduct, N2-(3-oxo-1-propenyl)deoxyguanosine (N2OPdG-), to 3-(2'-Deoxy-beta-D-erythro-pentofuranosyl)pyrimido[1,2-alpha]purin- 10(3H)-one (M1dG)

3-(2'-Deoxy-beta-D-erythro-pentofuranosyl)pyrimido[1,2-alpha]purin-10(3H)-one (M1dG) is the major product of the reaction of deoxyguanosine with malondialdehyde (MDA). M1dG blocks replication by DNA polymerases in vitro and is mutagenic in vivo. M1dG reacts with hydroxide to form the N2-(3-oxo-1-propenyl)deoxyguanosine anion (N2OPdG-). This reaction is pH-dependent and reverses under neutral and acidic conditions to form M1dG. Here we describe the kinetics and mechanism of the ring-closure reaction in both the nucleoside and oligonucleotides. Kinetic analysis of absorbance and fluorescence changes demonstrates that ring-closure is biphasic, leading to the rapid formation of an intermediate that slowly converts to M1dG in a general-acid-catalyzed reaction. The dependence of the rate of the rapid phase on pH reveals the pKa for protonated N2OPdG is 6.9. One-dimensional 1H NMR and DQF-COSY experiments identified two distinct intermediates, N2OPdG-H and 8-hydroxy-6,7-propenodeoxyguanosine (HO-Prene-dG), that are formed upon acidification of N2OPdG-. Characterization of ring-closure in single-stranded and in melted duplex oligonucleotides shows M1dG formation is also acid-catalyzed in single-stranded oligonucleotides and that the denaturation of an oligonucleotide duplex enhances ring-closure. This work details the complexity of ring-closure in the nucleoside and oligonucleotides and provides new insight into the role of duplex DNA in catalyzing ring-opening and ring-closing of M1dG and N2OPdG.     

Oxidation of Ni3(dpa)4Cl2 and Cu3(dpa)4Cl2: nickel-nickel bonding interaction,but no copper-copper bonds

Of the known trinuclear dipyridylamido complexes of the first-row transition metals, M(3)(dpa)(4)Cl(2) (dpa is the anion of di(2-pyridyl)amine, M = Cr, Co, Ni, Cu), the one-electron-oxidation products of only Cr(3)(dpa)(4)Cl(2) and Co(3)(dpa)(4)Cl(2) have been isolated previously. Here we report one-electron-oxidation products of Ni(3)(dpa)(4)Cl(2) (1) and Cu(3)(dpa)(4)Cl(2) (3): Ni(3)(dpa)(4)(PF(6))(3) (2) and [Cu(3)(dpa)(4)Cl(2)]SbCl(6) (4). While there are no Ni-Ni bonds in 1, the Ni-Ni distances in 2 are 0.15 A shorter than those in 1, very suggestive of metal-metal bonding interactions. In contrast, the oxidation of 3 to 4 is accompanied by a lengthening of the Cu-Cu distances, as expected for an increase in electrostatic charge between positively charged nonbonded metal ions, which is further evidence against Cu-Cu bonding in either 3 or 4. A qualitative model of the electronic structures of all [M(3)(dpa)(4)Cl(2)](n+) (n = 0, 1) compounds is presented and discussed.     

Magnetically induced electric field gradients

Several investigations of the Mössbauer spectrum of Fe₂As have yielded results which do not agree with those obtained by neutron diffraction studies. We report here a new investigation, and we show that the earlier disagreement is caused by a lack of appreciation of several factors. These include fluctuating hyperfine magnetic fields (relaxation) and an asymmetry of the electric field gradient produced by the spontaneous magnetization.On leave of absence at Laboratoire de Spectroscopie Hertzienne de l'Ecole Normale Superieure, 24 Rue Lhomond, 75005 Paris, France     

Synthetic magnetic opals

We present studies of novel nanocomposites of BiNi impregnated into the structure of opals as well as inverse opals. Atomic force microscopy and high resolution elemental analyses show a highly ordered structure and uniform distribution of the BiNi filler in the matrix. These BiNi-based nanocomposites are found to exhibit distinct ferromagnetic-like ordering with transition temperature of about 675 K. As far as we know there exists no report in literature on any BiNi compound which is magnetic.     

Statistics of defect trajectories in spatio-temporal chaos in inclined layer convection and the complex Ginzburg-Landau equation

For spatio-temporal chaos observed in numerical simulations of the complex Ginzburg-Landau equation (CGL) and in experiments on inclined-layer convection (ILC) we report numerical and experimental data on the statistics of defects and of defect loops. These loops consist of defect trajectories in space-time that are connected to each other through the pairwise annihilation or creation of the associated defects. While most such loops are small and contain only a few defects, the loop distribution functions decay only slowly with the quantities associated with the loop size, consistent with power-law behavior. For the CGL, two of the three power-law exponents are found to agree, within our computational precision, with those from previous investigations of a simple lattice model. In certain parameter regimes of the CGL and ILC, our results for the single-defect statistics show significant deviations from the previously reported findings that the defect dynamics are consistent with those of random walkers that are created with fixed probability and annihilated through random collisions.