核磁共振代谢组学数据的尺度归一化新方法

在核磁共振代谢组学数据预处理中,尺度归一化主要目的是提高特征代谢物信息的权重,减小噪声及无关代谢物信息的影响,从而降低后续模式识别分析的难度. 本文提出一种新的尺度归一化方法,该方法不强调各变量在尺度上的归一,而是在原始数据的基础上,通过提高那些稳定性高、且在不同类别样本中具有显著差异性的变量的权重,以增强与特征代谢物相关的信息. 文中分别采用模拟数据和真实代谢组学数据对新归一化方法的性能进行评估,并与单位方差法(Unit Variance)、变量稳定性(Variable Stability)和尺度缩放法(Level Scaling)等常用的尺度归一化方法做比较. 研究结果表明：新归一化方法能够提高多变量统计模型的预测能力,较好地保留核磁共振谱的分子信息,有助于特征代谢物的识别,并使后续的数据分析结果具有更好的可解释性.     

Low-load rotor-synchronised Hahn-echo pulse train (RS-HEPT) 1H decoupling in solid-state NMR: factors affecting MAS spin-echo dephasing times

Transverse dephasing times T(2)' in spin-echo MAS NMR using rotor-synchronised Hahn-echo pulse-train (RS-HEPT) low-load (1)H decoupling are evaluated. Experiments were performed at 300 and 600 MHz for (13)CH-labelled L-alanine and (15)NH(delta)-labelled L-histidine.HCl.H(2)O, together with SPINEVOLUTION simulations for a ten-spin system representing the crystal structure environment of the (13)CH carbon in L-alanine. For 30 kHz MAS and nu(1)((1)H) = 100 kHz at 300 MHz, a RS-HEPT T(2)' value of 17 +/- 1 ms was obtained for (13)CH-labelled L-alanine which is approximately 50% of the XiX T(2)' value of 33 +/- 2 ms. Optimum RS-HEPT decoupling performance is observed for a relative phase of alternate RS-HEPT pi-pulses, Deltaphi = phi'- phi, between 40 and 60 degrees . For experiments at 600 MHz and 30 kHz MAS with (13)CH-labelled L-alanine, the best RS-HEPT (nu(1)((1)H) = 100 kHz) T(2)' value was 3 times longer than that observed for low-power continuously applied sequences with nu(1)((1)H) < or =40 kHz, i.e. corresponding to the same average power dissipated in the probe. A marked improvement in RS-HEPT (1)H decoupling is observed for increasing MAS frequency: at 55.6 kHz MAS, a best RS-HEPT T(2)' value of 34 +/- 5 ms was recorded for (13)CH-labelled L-alanine. Much improved RS-HEPT broadband performance was also observed at 55.6 kHz MAS as compared to 30 kHz MAS.     

Quantifying hydrogen-bonding strength: the measurement of 2hJNN couplings in self-assembled guanosines by solid-state 15N spin-echo MAS NMR

(2h)J(NN) hydrogen-bond mediated J couplings are measured in the solid state for two synthetic deoxyguanosine derivatives by (15)N MAS NMR spin-echo experiments. The use of rotor-synchronised Hahn-echo pulse train (RS-HEPT) (1)H decoupling, with a duty cycle of 6%, allows spin-echo durations out to 200 ms, hence enabling the accurate determination of J couplings as small as 3.8 Hz. A single-crystal X-ray diffraction structure exists for the shorter alkyl chain derivative dG(C(3))(2): the observation of significantly different (2h)J(NN) couplings, 6.2 +/- 0.4 and 7.4 +/- 0.4 Hz, for the two resolved N7 resonances is to be expected given the NH...N hydrogen-bonding distances of 2.91 and 2.83 A for the two distinct molecules in the asymmetric unit cell. For the longer alkyl chain derivative, dG(C(10))(2), for which there is no single-crystal diffraction structure, a (15)N refocused INADEQUATE spectrum (Pham et al., J. Am. Chem. Soc., 2005, 127, 16018-16019) has demonstrated the presence of N2-H...N7 intermolecular hydrogen-bonds indicative of a quartet-like structure. The (2h)J(NN) hydrogen-bond mediated J coupling of 5.9 +/- 0.2 Hz is at the lower end of the range (5.9-8.2 Hz) of (2h)J(NN) couplings determined from solution-state NMR of guanosine quartets in quadruplex DNA. A full discussion of the determination of error bars on the fitted parameters is given; specifically, error bars determined by a non-linear fitting (using the covariance matrix) or in a Monte-Carlo fashion are found to give effectively identical results.     

Quinolinone and pyridopyrimidinone inhibitors of DNA-dependent protein kinase

8-Substituted 2-morpholin-4-yl-quinolin-4-ones and 9-substituted 2-morpholin-4-yl-pyrido[1,2-a]pyrimidin-4-ones with selected aryl and heteroaryl groups as the substituent have been synthesised as potential inhibitors of DNA-dependent protein kinase. A multiple-parallel approach, employing Suzuki cross-coupling methodology, was utilised in the preparation of 8-substituted 2-morpholin-4-yl-quinolin-4-ones. For this purpose 8-bromo-2-morpholin-4-yl-quinolin-4-one was required as an intermediate. This compound was obtained by adapting a literature route in which thermal cyclocondensation of (2-bromoanilino)-morpholin-4-yl-5-methylene-2,2-dimethyl[1,3]dioxane-4,6-dione afforded 8-bromo-2-morpholin-4-yl-quinolin-4-one. A multiple-parallel approach, employing Suzuki cross-coupling methodology, was also utilised to prepare 9-substituted 2-morpholin-4-yl-pyrido[1,2-a]pyrimidin-4-ones using 9-hydroxy-2-morpholin-4-yl-pyrido[1,2-a]pyrimidin-4-one O-trifluoromethanesulfonate as an intermediate. 8-Substituted 2-morpholin-4-yl-quinolin-4-ones and 9-substituted 2-morpholin-4-yl-pyrido[1,2-a]pyrimidin-4-ones were both inhibitors of DNA-dependent protein kinase. When the substituent was dibenzothiophen-4-yl, dibenzofuran-4-yl or biphen-3-yl, IC50 values in the low nanomolar range were observed. Interestingly, the pyridopyrimidinones and quinolinones were essentially equipotent with the corresponding 8-substituted 2-morpholin-4-yl-chromen-4-ones previously reported (I. R. Hardcastle, X. Cockcroft, N. J. Curtin, M. Desage El-Murr, J. J. J. Leahy, M. Stockley, B. T. Golding, L. Rigoreau, C. Richardson, G. C. M. Smith and R. J. Griffin, J. Med. Chem., 2005, 48, 7829-7846).     

Electrostatically stabilized metal oxide particle dispersions in carbon dioxide

We report electrostatic stabilization of micrometer-sized TiO(2) particles at long range (several micrometers) in liquid and supercritical CO(2) despite the ultralow dielectric constant, as low as 1.5. The counterions were solubilized in dry reverse micelles, formed with a low-molecular weight cationic perfluoropolyether trimethylammonium acetate surfactant, to prevent ion pairing with the particle surface. Dynamic light scattering and settling velocities indicate a particle diameter of 620-740 nm. The electrophoretic mobility of -2.3 x 10(-8) m(2)/V s indicated a particle charge on the order of -1.7 x 10(-17) C, or 105 elementary negative charges per particle. The balance of particle compression by an electric field versus electrostatic repulsion generated an amorphous arrangement of particles with 5-9 mum spacing, indicating Debye lengths greater than 1 mum. Scattering patterns also indicate that chains of particles may be achieved in CO(2) by dielectrophoresis with alternating fields. The electrostatic stabilization has been achieved by solubilizing a small concentration of counterions in only a small fraction of the reverse micelles in the double layer. Whereas many low-molecular weight surfactants have been shown to form reverse micelles in CO(2), very few polymers are able to stabilize micrometer-sized colloids sterically. Thus, electrostatic stabilization has the potential to expand markedly the domain of colloid science in apolar supercritical fluids.     

Recoupling in solid state NMR using γ prepared states and phase matching

The paper describes two-dimensional solid state NMR experiments that use powdered dephased antiphase coherence (γ preparation) to encode chemical shifts in the indirect dimension. Both components of this chemical shift encoded gamma-prepared states can be refocused into inphase coherence by a recoupling element. This helps to achieve sensitivity enhancement in 2D NMR experiments by quadrature detection. The powder dependence of the gamma-prepared states allows for manipulating them by suitable insertion of delays in the recoupling periods. This helps to design experiments that suppress diagonal peaks in 2D spectra, leading to improved resolution. We describe some new phase modulated heteronuclear and homonuclear recoupling pulse sequences that simplify the implementation of the described experiments based on γ prepared states. Recoupling in the heteronuclear spin system is achieved by matching the difference in the amplitude of the sine/cosine modulated phase on the two rf-channels to the spinning frequency while maintaining the same power on the two rf-channels.     

Dynamic nuclear polarization at 9T using a novel 250 GHz gyrotron microwave source. 2003

In this communication, we report enhancements of nuclear spin polarization by dynamic nuclear polarization (DNP) in static and spinning solids at a magnetic field strength of 9 T (250 GHz for g = 2 electrons, 380 MHz for ¹H). In these experiments, ¹H enhancements of up to 170 ± 50 have been observed in 1-¹³C-glycine dispersed in a 60:40 glycerol/water matrix at temperatures of 20 K; in addition, we have observed significant enhancements in ¹⁵N spectra of unoriented pf1-bacteriophage. Finally, enhancements of ∼17 have been obtained in two-dimensional ¹³C–¹³C chemical shift correlation spectra of the amino acid U–¹³C, ¹⁵N-proline during magic angle spinning (MAS), demonstrating the stability of the DNP experiment for sustained acquisition and for quantitative experiments incorporating dipolar recoupling. In all cases, we have exploited the thermal mixing DNP mechanism with the nitroxide radical 4-amino-TEMPO as the paramagnetic dopant. These are the highest frequency DNP experiments performed to date and indicate that significant signal enhancements can be realized using the thermal mixing mechanism even at elevated magnetic fields. In large measure, this is due to the high microwave power output of the 250 GHz gyrotron oscillator used in these experiments.     

Finite element modelling of surface waves in a channel

A variational formulation of the vertically-integrated differential equations for free surface wave motion is presented. A finite element model is derived for solving this nonlinear system of hydrodynamic equations. The time integration scheme employed is discussed and the results obtained demonstrate its good stability and accuracy.Several applications of the model are considered: the first problem is an open channel of uniform depth and the second an open channel of linearly varying depth. The ‘inflow’ boundary condition is prescribed in terms of the velocity which represents a wavemaker and/or a flow source, while the ‘outflow’ boundary condition is specified in terms of the water elevation. The outflow condition is adjusted for two cases, a reflecting boundary (finite channel) and a non-reflecting boundary (open-ended channel). The latter boundary condition is examined in some detail and the results obtained show that the numerical model can produce the non-reflecting boundary that is similar to the analytical radiation condition for waves. Computational results for a third problem, involving wave reflection from a submerged cylinder, are also presented and compared with both experimental data and analytical predictions.The simplicity and the performance of the computational model suggest that free surface waves can be simulated without excessively complicated numerical schemes. The ability of the model to simulate outflow boundary conditions properly is of special importance since these conditions present serious problems for many numerical algorithms.     

Multi-criteria logistics modeling for military humanitarian assistance and disaster relief aerial delivery operations

Given that it is not always feasible to reach an affected area via land or sea within the first week following a natural disaster, aerial delivery provides the primary means to rapidly supply the affected population. Further, it is often the case that high density delivery of humanitarian aid supplies are taken over by non-friendly groups within the affected population. By using direct airdrop systems to deliver large quantities of individually wrapped food and water items, dispersion among the affected disaster relief population will occur more quickly. In this paper, we proffer a multiple criteria decision analysis (MCDA) framework to optimize the military humanitarian assistance/disaster relief (HA/DR) aerial delivery supply chain network. The model uses stochastic, mixed-integer, weighted goal programming to optimize network design, logistics costs, staging locations, procurement amounts, and inventory levels. The MCDA framework enables decision-makers to explore the trade-offs between military HA/DR aerial delivery supply chain efficiency and responsiveness, while optimizing across a wide range of real-world, probabilistic scenarios to account for the inherent uncertainty in the location of global humanitarian disasters as well as the amount of demand to be met.     

Using an iterative linear solver in an interior-point method for generating support vector machines

This paper concerns the generation of support vector machine classifiers for solving the pattern recognition problem in machine learning. A method is proposed based on interior-point methods for convex quadratic programming. This interior-point method uses a linear preconditioned conjugate gradient method with a novel preconditioner to compute each iteration from the previous. An implementation is developed by adapting the object-oriented package OOQP to the problem structure. Numerical results are provided, and computational experience is discussed.