Comparison of sotalol versus amiodarone in maintaining stability of sinus rhythm in patients with atrial fibrillation (Sotalol-Amiodarone Fibrillation Efficacy Trial [Safe-T])

The Sotalol-Amiodarone Fibrillation Efficacy Trial (SAFE-T) is a randomized, double-blind, multicenter, placebo-controlled trial in which the effects of sotalol and amiodarone in maintaining stability of sinus rhythm are being examined in patients with persistent atrial fibrillation at 20 Veterans Affairs medical centers. The time to the occurrence of atrial fibrillation or flutter in patients with atrial fibrillation converted to sinus rhythm is the primary outcome measure, with a number of parameters as secondary end points. SAFE-T had randomized 665 patients when enrollment terminated on October 31, 2001. Follow-up of patients continued until October 31, 2002, for a maximum period of 54 months and a minimum period of 12 months for all patients.     

Genetic diversity of VP3 of goose parvovirus isolated from Southeastern China during 2012–2013

Although vaccine and passive immunotherapy were widely used to prevent goose parvovirus (GPV) infection in goose industry, GPV still poses a big problem in Southeastern China. In this study, 23 GPV isolates were isolated from goslings suspected with GPV infection in Southeastern China during 2012–2013, and the genetic diversity of VP3 of GPV was analyzed. Phylogenetic tree revealed that these isolates could be clustered into two groups, and 11 of 23 could be further clustered into a new subgroup. Moreover, eight novel mutations and seventeen conserved amino acids were found in these 23 isolates in comparison with isolates previously deposited in GenBank. These isolates and findings not only provide insights into the etiology and molecular characteristics of GPV endemic in Southeastern China, but also enrich the GPV genetic information for better controlling the disease.     

A strategy to improve treatment‐related mortality and abandonment of therapy for childhood ALL in a developing country reveals the impact of treatment delays

Background

Treatment‐related mortality and abandonment of therapy are major barriers to successful treatment of childhood acute lymphoblastic leukemia (ALL) in the developing world.

Procedure

A collaboration was undertaken between Instituto Nacional de Cancerologia (Bogota, Colombia), which serves a poor patient population in an upper‐middle income country, and Dana‐Farber/Boston Children's Cancer and Blood Disorders Center (Boston, USA). Several interventions aimed at reducing toxic deaths and abandonment were implemented, including a reduced‐intensity treatment regimen and a psychosocial effort targeting abandonment. We performed a cohort study to assess impact.

Results

The Study Population comprised 99 children with ALL diagnosed between 2007 and 2010, and the Historic Cohort comprised 181 children treated prior to the study interventions (1995–2004). Significant improvements were achieved in the rate of deaths in complete remission (13% to 3%; P = 0.005), abandonment (32% to 9%; P < 0.001), and event‐free survival with abandonment considered an event (47% to 65% at 2 years; P = 0.016). However, relapse rate did not improve. Medically unnecessary treatment delays were common, and landmark analysis revealed that initiating the PIII phase of therapy ≥4 weeks delayed predicted markedly inferior disease‐free survival (P = 0.016). Conversely, patients who received therapy without excessive delays had outcomes approaching those achieved in high‐income countries.

Conclusions

Implementation of a twinning program was followed by reductions in abandonment and toxic deaths, but relapse rate did not improve. Inappropriate treatment delays were common and strongly predicted treatment failure. These findings highlight the importance of adherence to treatment schedule for effective therapy of ALL. Pediatr Blood Cancer 2015;62:1395–1402. © 2015 Wiley Periodicals, Inc.     

The effect of increasing alpha1-acid glycoprotein concentration on the antiviral efficacy of human immunodeficiency virus protease inhibitors

The effect of a 4-fold increase in alpha1-acid glycoprotein (AGP) on the antiviral efficacy of 5 human immunodeficiency virus (HIV) protease inhibitors (PIs) was examined by the effect of HIV PIs on p24 production in peripheral blood mononuclear cells infected with protease wild-type and PI-resistant HIV isolates. For wild-type virus, the efficacy of the PIs at trough concentrations was unaffected by a 4-fold increase in AGP. With the partially HIV PI-resistant isolate, a 4-fold increase in AGP resulted in 2%, 30%, 37%, 37%, and 42% loss of activity for indinavir, saquinavir, nelfinavir, ritonavir, and amprenavir, respectively. The high-level HIV PI-resistant isolate had a greater loss in activity. The change in IC50 secondary to the addition of AGP was the greatest for ritonavir, nelfinavir, and amprenavir and lowest for indinavir. These data suggest that the target plasma concentration for the highly bound HIV PIs may need to be raised in subjects with elevated AGP who harbor partially PI-resistant isolates.     

Efficient inference of population size histories and locus-specific mutation rates from large-sample genomic variation data

With the recent increase in study sample sizes in human genetics, there has been growing interest in inferring historical population demography from genomic variation data. Here, we present an efficient inference method that can scale up to very large samples, with tens or hundreds of thousands of individuals. Specifically, by utilizing analytic results on the expected frequency spectrum under the coalescent and by leveraging the technique of automatic differentiation, which allows us to compute gradients exactly, we develop a very efficient algorithm to infer piecewise-exponential models of the historical effective population size from the distribution of sample allele frequencies. Our method is orders of magnitude faster than previous demographic inference methods based on the frequency spectrum. In addition to inferring demography, our method can also accurately estimate locus-specific mutation rates. We perform extensive validation of our method on simulated data and show that it can accurately infer multiple recent epochs of rapid exponential growth, a signal that is difficult to pick up with small sample sizes. Lastly, we use our method to analyze data from recent sequencing studies, including a large-sample exome-sequencing data set of tens of thousands of individuals assayed at a few hundred genic regions.The demography of an evolving population strongly influences the genetic variation found within it, and understanding the intricate interplay between natural selection, genetic drift, and demography is a key aim of population genomics. For example, the human census population has expanded more than 1000-fold in the last 400 generations (Keinan and Clark 2012), resulting in a state that is profoundly out of equilibrium with respect to genetic variation. Recently, there has been much interest in studying the consequences of such rapid expansion on mutation load and the genetic architecture of complex traits (Gazave et al. 2013; Lohmueller 2014; Simons et al. 2014). Estimating the population demography is necessary for developing more accurate null models of neutral evolution in order to identify genomic regions subject to natural selection (Williamson et al. 2005; Boyko et al. 2008; Lohmueller et al. 2008). The problem of inferring demography from genomic data also has several other important applications. In particular, the population demography is needed to correct for spurious genotype-phenotype associations in genome-wide association studies due to hidden population substructure (Marchini et al. 2004; Campbell et al. 2005; Clayton et al. 2005), to date historical population splits, migrations, admixture, and introgression events (Gravel et al. 2011; Li and Durbin 2011; Lukić and Hey 2012; Sankararaman et al. 2012), to compute random match probabilities accurately in forensic applications (Balding and Nichols 1997; Graham et al. 2000), for examples.A commonly used null model in population genetics assumes that individuals are randomly sampled from a well-mixed population of constant size that evolves neutrally according to some model of random mating (Ewens 2004). However, several recent large-sample sequencing studies in humans (Coventry et al. 2010; Fu et al. 2012; Nelson et al. 2012; Tennessen et al. 2012) have found an excess of single nucleotide variants (SNVs) that have very low minor allele frequency (MAF) in the sample compared to that predicted by coalescent models with a constant effective population size. For example, in a sample of ∼12,500 individuals of European descent analyzed by Nelson et al. (2012), >74% of the SNVs have only one or two copies of the minor allele, and >95% of the SNVs have an MAF <0.5%. On the other hand, assuming a constant population size over time, Kingman’s coalescent predicts that the number of neutral SNVs is inversely proportional to the sample frequency of the variant (Fu 1995). Keinan and Clark (2012) have suggested that such an excess of sites segregating with low MAF can be explained by recent exponential population growth. In particular, a rapid population expansion produces genealogical trees that have long branch lengths at the tips of the trees, leading to a large fraction of mutations being limited to a single individual in the sample. Motivated by these findings and rapidly increasing sample sizes in population genomics, we here tackle the problem of developing an efficient algorithm for inferring historical effective population sizes and locus-specific mutation rates using a very large sample, with tens or hundreds of thousands of individuals.At the coarsest level, previous approaches to inferring demography from genomic variation data can be divided according to the representation of the data that they operate on. Full sequence-based approaches for inferring the historical population size such as the works of Li and Durbin (2011) and Sheehan et al. (2013) use between two and a dozen genomes to infer piecewise constant models of historical population sizes. Since these approaches operate genome wide, they can take into account linkage information between neighboring SNVs. On the other hand, they are computationally very expensive and cannot be easily applied to infer recent demographic events from large numbers of whole genomes. A slightly more tractable approach to inferring potentially complex demographies involves comparing the length distribution of identical-by-descent and identical-by-state tracts between pairs of sequences (Palamara et al. 2012; Harris and Nielsen 2013).The third class of methods, and the one that our approach also belongs to, summarizes the variation in the genome sequences by the sample frequency spectrum (SFS). The SFS of a sample of size n counts the number of SNVs as a function of their mutant allele frequency in the sample. Since the SFS is a very efficient dimensional reduction of large-scale population genomic data that summarizes the variation in n sequences by n − 1 numbers, it is naturally attractive for computational and statistical purposes. Furthermore, the expected SFS of a random sample drawn from the population strongly depends on the underlying demography, and there have been several previous approaches that exploit this relationship for demographic inference. Nielsen (2000) developed a method based on coalescent tree simulations to infer exponential population growth from single nucleotide polymorphisms that are far enough apart to be in linkage equilibrium. Coventry et al. (2010) developed a similar coalescent simulation-based method that additionally infers per-locus mutation rates and applied this method to exome-sequencing data from ∼10,000 individuals at two genes. Nelson et al. (2012) have also applied this method to a larger data set of 11,000 individuals of European ancestry (CEU) sequenced at 185 genes to infer a recent epoch of exponential population growth. The common feature of all these methods is that they use Monte Carlo simulations to empirically estimate the expected SFS under a given demographic model, and then they compute a pseudo-likelihood function for the demographic model by comparing the expected and observed SFS. The optimization over the demographic models is then performed via grid search procedures. More recently, Excoffier et al. (2013) have developed a software package that employs coalescent tree simulations to estimate the expected joint SFS of multiple subpopulations for inferring potentially very complex demographic scenarios from multipopulation genomic data. The problem of demographic inference has also been approached from the perspective of diffusion processes. Given a demographic model, one can derive a partial differential equation (PDE) for the density of segregating sites at a given derived allele frequency as a function of time. Gutenkunst et al. (2009) used numerical methods to approximate the solution to this PDE, while Lukić et al. (2011) approximated this solution using an orthogonal polynomial expansion. The coalescent-based method of Excoffier et al. (2013), fastsimcoal, and the diffusion-based method of Gutenkunst et al. (2009), ∂a∂i, can infer the joint demography of multiple subpopulations with changing population sizes and complex patterns of migration between subpopulations.In this paper, we focus on the problem of inferring the effective population size as a function of time for a single randomly mating population. As mentioned above, our method is based on the SFS. By restricting our inference to a single population, we are able to compute the expected SFS exactly, rather than using Monte Carlo simulations or solving PDEs numerically. Briefly, we utilize the theoretical work of Polanski et al. (2003) and Polanski and Kimmel (2003), which relate the expected SFS for a sample of size n from a single population to the expected waiting times to the first coalescence event for all sample sizes ≤n. We show that the latter quantities can be computed efficiently and numerically stably for very large sample sizes and for an arbitrary piecewise-exponential model of the historical effective population size. Further, our method utilizes the technique of automatic differentiation to compute exact gradients of the likelihood with respect to the parameters of the effective population size function, thereby facilitating optimization over the space of demographic parameters. These techniques result in our method being both more accurate and more computationally efficient than ∂a∂i and fastsimcoal. In what follows, we carry out an extensive simulation study to demonstrate that our method can infer multiple recent epochs of rapid exponential growth and estimate locus-specific mutation rates with a high accuracy. We then apply our method to analyze data from recent sequencing studies.     

Three-dimensional honeycomb-like porous carbon derived from corncob for the removal of heavy metals from water by capacitive deionization

In this study, porous carbon (3DHPC) with a 3D honeycomb-like structure was synthesized from waste biomass corncob via hydrothermal carbonization coupled with KOH activation and investigated as a capacitive deionization (CDI) electrode material. The obtained 3DHPC possesses a hierarchal macroporous and mesoporous structure, and a large accessible specific surface area (952 m² g⁻¹). Electrochemical tests showed that the 3DHPC electrode exhibited a specific capacitance of 452 F g⁻¹ and good electric conductivity. Moreover, the feasibility of electrosorptive removal of chromium(vi) from an aqueous solution using the 3DHPC electrode was demonstrated. When 1.0 V was applied to a solution containing 30 mg L⁻¹ chromium(vi), the 3DHPC electrode exhibited a higher removal efficiency of 91.58% compared with that in the open circuit condition. This enhanced adsorption results from the improved affinity between chromium(vi) and the electrode under electrochemical assistance involving a non-faradic process. Consequently, the 3DHPC electrode with typical double-layer capacitor behavior is demonstrated to be a favorable electrode material for capacitive deionization.

A porous carbon electrode with a 3D honeycomb-like structure demonstrates a high removal efficiency for the removal of chromium(vi) from water.     

4-Phenylpyrimidine monolayer protection of a copper surface from salt corrosion

4-Phenylpyrimidine (4-PPM) containing N heteroatoms can easily form compact and uniform layers on metallic surfaces. In this work, the protection of a copper surface from corrosion in 3 wt% NaCl by a 4-PPM layer was investigated by electrochemical impedance spectroscopy (EIS) and polarization methods. Under optimum conditions, the inhibition efficiency of a 4-PPM layer for copper corrosion reached 83.2%. Raman analysis in conjunction with calculations using density functional theory (DFT) based on the B3LYP/LANL2DZ basis set suggested that the 4-PPM molecule anchored on the copper surface via the N₁ atom to construct a uniform layer.

The efficiency of a 4-phenylpyrimidine monolayer optimally self-assembled on a copper surface against corrosion by a 3 wt% NaCl solution could reach 83.2%.