Environmental and Occupational Health Sciences Institute

Many problems of recent interest in statistics and machine learning can be posed in the framework of convex optimization. Due to the explosion in size and complexity of modern datasets, it is increasingly important to be able to solve problems with a very large number of features or training examples. As a result, both the decentralized collection or storage of these datasets as well as accompanying distributed solution methods are either necessary or at least highly desirable. In this review, we argue that the alternating direction method of multipliers is well suited to distributed convex optimization, and in particular to large-scale problems arising in statistics, machine learning, and related areas. The method was developed in the 1970s, with roots in the 1950s, and is equivalent or closely related to many other algorithms, such as dual decomposition, the method of multipliers, Douglas–Rachford splitting, Spingarn's method of partial inverses, Dykstra's alternating projections, Bregman iterative algorithms for ℓ1 problems, proximal methods, and others. After briefly surveying the theory and history of the algorithm, we discuss applications to a wide variety of statistical and machine learning problems of recent interest, including the lasso, sparse logistic regression, basis pursuit, covariance selection, support vector machines, and many others. We also discuss general distributed optimization, extensions to the nonconvex setting, and efficient implementation, including some details on distributed MPI and Hadoop MapReduce implementations.

0

Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden.The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.

Magnetism and ferroelectricity are essential to many forms of current technology, and the quest for multiferroic materials, where these two phenomena are intimately coupled, is of great technological and fundamental importance. Ferroelectricity and magnetism tend to be mutually exclusive and interact weakly with each other when they coexist. The exciting new development is the discovery that even a weak magnetoelectric interaction can lead to spectacular cross-coupling effects when it induces electric polarization in a magnetically ordered state. Such magnetic ferroelectricity, showing an unprecedented sensitivity to ap plied magnetic fields, occurs in 'frustrated magnets' with competing interactions between spins and complex magnetic orders. We summarize key experimental findings and the current theoretical understanding of these phenomena, which have great potential for tuneable multifunctional devices.

A two-dimensional crystal of molybdenum disulfide (MoS2) monolayer is a photoluminescent direct gap semiconductor in striking contrast to its bulk counterpart. Exfoliation of bulk MoS2 via Li intercalation is an attractive route to large-scale synthesis of monolayer crystals. However, this method results in loss of pristine semiconducting properties of MoS2 due to structural changes that occur during Li intercalation. Here, we report structural and electronic properties of chemically exfoliated MoS2. The metastable metallic phase that emerges from Li intercalation was found to dominate the properties of as-exfoliated material, but mild annealing leads to gradual restoration of the semiconducting phase. Above an annealing temperature of 300 °C, chemically exfoliated MoS2 exhibit prominent band gap photoluminescence, similar to mechanically exfoliated monolayers, indicating that their semiconducting properties are largely restored.

Background Since at least 400 C.E., when the Mayans first used layered clays to make dyes, people have been harnessing the properties of layered materials. This gradually developed into scientific research, leading to the elucidation of the laminar structure of layered materials, detailed understanding of their properties, and eventually experiments to exfoliate or delaminate them into individual, atomically thin nanosheets. This culminated in the discovery of graphene, resulting in a new explosion of interest in two-dimensional materials. Layered materials consist of two-dimensional platelets weakly stacked to form three-dimensional structures. The archetypal example is graphite, which consists of stacked graphene monolayers. However, there are many others: from MoS 2 and layered clays to more exotic examples such as MoO 3 , GaTe, and Bi 2 Se 3 . These materials display a wide range of electronic, optical, mechanical, and electrochemical properties. Over the past decade, a number of methods have been developed to exfoliate layered materials in order to produce monolayer nanosheets. Such exfoliation creates extremely high-aspect-ratio nanosheets with enormous surface area, which are ideal for applications that require surface activity. More importantly, however, the two-dimensional confinement of electrons upon exfoliation leads to unprecedented optical and electrical properties. Advances An important advance has been the discovery that layered crystals can be exfoliated in liquids. There are a number of methods to do this that involve oxidation, ion intercalation/exchange, or surface passivation by solvents. However, all result in liquid dispersions containing large quantities of nanosheets. This brings considerable advantages: Liquid exfoliation allows the formation of thin films and composites, is potentially scaleable, and may facilitate processing by using standard technologies such as reel-to-reel manufacturing. Although much work has focused on liquid exfoliation of graphene, such processes have also been demonstrated for a host of other materials, including MoS 2 and related structures, layered oxides, and clays. The resultant liquid dispersions have been formed into films, hybrids, and composites for a range of applications. Outlook There is little doubt that the main advances are in the future. Multifunctional composites based on metal and polymer matrices will be developed that will result in enhanced mechanical, electrical, and barrier properties. Applications in energy generation and storage will abound, with layered materials appearing as electrodes or active elements in devices such as displays, solar cells, and batteries. Particularly important will be the use of MoS 2 for water splitting and metal oxides as hydrogen evolution catalysts. In addition, two-dimensional materials will find important roles in printed electronics as dielectrics, optoelectronic devices, and transistors. To achieve this, much needs to be done. Production rates need to be increased dramatically, the degree of exfoliation improved, and methods to control nanosheet properties developed. The range of layered materials that can be exfoliated must be expanded, even as methods for chemical modification must be developed. Success in these areas will lead to a family of materials that will dominate nanomaterials science in the 21st century.

In a seminal paper, Garrett Hardin argued in 1968 that users of a commons are caught in an inevitable process that leads to the destruction of the resources on which they depend. This article discusses new insights about such problems and the conditions most likely to favor sustainable uses of common-pool resources. Some of the most difficult challenges concern the management of large-scale resources that depend on international cooperation, such as fresh water in international basins or large marine ecosystems. Institutional diversity may be as important as biological diversity for our long-term survival.

The Female Sexual Function Index (FSFI) is a brief, multidimensional scale for assessing sexual function in women. The scale has received initial psychometric evaluation, including studies of reliability, convergent validity, and discriminant validity (Meston, 2003 Meston, C. M. 2003. Validation of the Female Sexual Function Index (FSFI) in women with female orgasmic disorder and in women with hypoactive sexual desire disorder. Journal of Sex & Marital Therapy, 29: 39–46. [Taylor & Francis Online], [Web of Science ®] , [Google Scholar]; Rosen et al., 2000 Rosen, R. C., Brown, C., Heiman, J., Leiblum, S., Meston, C. M., Shabsigh, R., Ferguson, D. and D'Agostino, R. 2000. The Female Sexual Function Index (FSFI): A multidimensional self-report instrument for the assessment of female sexual function. Journal of Sex & Marital Therapy, 26: 191–208. [Taylor & Francis Online], [Web of Science ®] , [Google Scholar]). The present study was designed to cross-validate the FSFI in several samples of women with mixed sexual dysfunctions (N = 568) and to develop diagnostic cut-off scores for potential classification of women's sexual dysfunction. Some of these samples were drawn from our previous validation studies (N = 414), and some were added for purposes of the present study (N = 154). The combined data set consisted of multiple samples of women with sexual dysfunction diagnoses (N = 307), including female sexual arousal disorder (FSAD), hypoactive sexual desire disorder (HSDD), female sexual orgasm disorder (FSOD), dyspareunia/vaginismus (pain), and multiple sexual dysfunctions, in addition to a large sample of nondysfunctional controls (n = 261). We conducted analyses on the individual and combined samples, including replicating the original factor structure using principal components analysis with varimax rotation. We assessed Cronbach's alpha (internal reliability) and interdomain correlations and tested discriminant validity by means of a MANOVA (multivariate analysis of variance; dysfunction diagnosis x FSFI domain), with Bonferroni-corrected post hoc comparisons. We developed diagnostic cut off scores by means of standard receiver operating characteristics–curves and the CART (Classification and Regression Trees) procedure. Principal components analysis replicated the original five-factor structure, including desire/arousal, lubrication, orgasm, pain, and satisfaction. We found the internal reliability for the total FSFI and six domain scores to be good to excellent, with Cronbach alpha's > 0.9 for the combined sample and above 0.8 for the sexually dysfunctional and nondysfunctional samples, independently. Discriminant validity testing confirmed the ability of both total and domain scores to differentiate between functional and nondysfunctional women. On the basis of sensitivity and specificity analyses and the CART procedure, we found an FSFI total score of 26.55 to be the optimal cut score for differentiating women with and without sexual dysfunction. On the basis of this cut-off, we found 70.7% of women with sexual dysfunction and 88.1% of the sexually functional women in the cross-validation sample to be correctly classified. Addition of the lubrication score in the model resulted in slightly improved specificity (from .707 to .772) at a slight cost of sensitivity (from .881 to .854) for identifying women without sexual dysfunction. We discuss the results in terms of potential strengths and weaknesses of the FSFI, as well in terms of further clinical and research implications.

Blue photoluminescene (PL) from chemically derived graphene oxide has been observed. The PL is attributed to radiative recombination of electron–hole pairs in isolated “molecular” sp2 domains, which are present within the carbon–oxygen sp3 matrix in chemically derived graphene oxide. The PL intensity is correlated to the evolution of sp2 domains during reduction.

MicroRNAs (miRNAs) are small noncoding RNA gene products about 22 nt long that are processed by Dicer from precursors with a characteristic hairpin secondary structure. Guidelines are presented for the identification and annotation of new miRNAs from diverse organisms, particularly so that miRNAs can be reliably distinguished from other RNAs such as small interfering RNAs. We describe specific criteria for the experimental verification of miRNAs, and conventions for naming miRNAs and miRNA genes. Finally, an online clearinghouse for miRNA gene name assignments is provided by the Rfam database of RNA families.

In higher eukaryotes, translation is modulated at the level of initiation by five aspects of mRNA structure: (i) the m7G cap; (ii) the primary sequence or context surrounding the AUG codon; (iii) the position of the AUG codon, i.e. whether or not it s “first”; (iv) secondary structure both upstream and downstream from the AUG codon; and (v) leader length. Here I briefly discuss how experimental manipulation of these features affects the fidelity and/or efficiency of initiation. Elsewhere (1) I discuss the extent to which natural mRNA leader sequences conform to these experimentally determined requirements for initiation. Although my primary concern is to document the occurrence and consequences of the five structural elements in mRNAs, I will allude now and then to why each feature has the effect it does. The explanations will invariably hark back to the scanning process by which ribosomes are thought to initiate translation. In its simplest form, the scanning model (2) postulates that a 40 S ribosomal subunit, carrying MettRNA? and an imperfectly defined set of initiation factors (3), enters at the 5‘-end of the mRNA and migrates linearly until it reaches the first AUG codon, whereupon a 60 S ribosomal subunit joins and the first peptide bond is formed (Fig. 1). Evidence in support of the scanning model has been adduced previously (2, 4, 5). The most recent evidence includes the apparent queuing of 40 S ribosomal subunits on a long, unstructured leader sequence (6) and the demonstration that 40 S subunits stall on the 5’-side of a stable hairpin structure introduced between the cap and the AUG codon (7).

BACKGROUND: Epidemiologic evidence indicates that diets high in fruits and vegetables are associated with a reduced risk of several cancers, including cancers of the esophagus and stomach. Vitamins and minerals in these foods may contribute to the reduced cancer risk. The people of Linxian County, China, have one of the world's highest rates of esophageal/gastric cardia cancer and a persistently low intake of several micronutrients. PURPOSE: We sought to determine if dietary supplementation with specific vitamins and minerals can lower mortality from or incidence of cancer as well as mortality from other diseases in Linxian. METHODS: Individuals of ages 40-69 were recruited in 1985 from four Linxian communes. Mortality and cancer incidence during March 1986-May 1991 were ascertained for 29,584 adults who received daily vitamin and mineral supplementation throughout this period. The subjects were randomly assigned to intervention groups according to a one-half replicate of a 2(4) factorial experimental design. This design enabled testing for the effects of four combinations of nutrients: (A) retinol and zinc; (B) riboflavin and niacin; (C) vitamin C and molybdenum; and (D) beta carotene, vitamin E, and selenium. Doses ranged from one to two times U.S. Recommended Daily Allowances. RESULTS: A total of 2127 deaths occurred among trial participants during the intervention period. Cancer was the leading cause of death, with 32% of all deaths due to esophageal or stomach cancer, followed by cerebrovascular disease (25%). Significantly (P = .03) lower total mortality (relative risk [RR] = 0.91; 95% confidence interval [CI] = 0.84-0.99) occurred among those receiving supplementation with beta carotene, vitamin E, and selenium. The reduction was mainly due to lower cancer rates (RR = 0.87; 95% CI = 0.75-1.00), especially stomach cancer (RR = 0.79; 95% CI = 0.64-0.99), with the reduced risk beginning to arise about 1-2 years after the start of supplementation with these vitamins and minerals. No significant effects on mortality rates from all causes were found for supplementation with retinol and zinc, riboflavin and niacin, or vitamin C and molybdenum. Patterns of cancer incidence, on the basis of 1298 cases, generally resembled those for cancer mortality. CONCLUSIONS: The findings indicate that vitamin and mineral supplementation of the diet of Linxian adults, particularly with the combination of beta carotene, vitamin E, and selenium, may effect a reduction in cancer risk in this population. IMPLICATIONS: The results on their own are not definitive, but the promising findings should stimulate further research to clarify the potential benefits of micronutrient supplements.

Longitudinal data were obtained from a nonclinical sample of 1,308 male and female adolescents covering the age range from 12 to 21. Factor analyses of 52 symptoms and/or consequences of alcohol use yielded three problem dimensions. In addition, a unidimensional, 23-item scale (the Rutgers Alcohol Problem Index, RAPI) was constructed with an internal consistency of .92. Correlations between RAPI and alcohol-use intensity were moderately strong for all age groups at each test occasion (ranging from .20 to .57), yet low enough to suggest that identification of problem drinkers requires both types of measures. The results suggest that the RAPI may be a useful tool for the standardized and efficient assessment of problem drinking during adolescence.

Five structural features in mRNAs have been found to contribute to the fidelity and efficiency of initiation by eukaryotic ribosomes. Scrutiny of vertebrate cDNA sequences in light of these criteria reveals a set of transcripts--encoding oncoproteins, growth factors, transcription factors, and other regulatory proteins--that seem designed to be translated poorly. Thus, throttling at the level of translation may be a critical component of gene regulation in vertebrates. An alternative interpretation is that some (perhaps many) cDNAs with encumbered 5' noncoding sequences represent mRNA precursors, which would imply extensive regulation at a posttranscriptional step that precedes translation.

ATP-dependent nucleosome remodeling and core histone acetylation and deacetylation represent mechanisms to alter nucleosome structure. NuRD is a multisubunit complex containing nucleosome remodeling and histone deacetylase activities. The histone deacetylases HDAC1 and HDAC2 and the histone binding proteins RbAp48 and RbAp46 form a core complex shared between NuRD and Sin3-histone deacetylase complexes. The histone deacetylase activity of the core complex is severely compromised. A novel polypeptide highly related to the metastasis-associated protein 1, MTA2, and the methyl-CpG-binding domain-containing protein, MBD3, were found to be subunits of the NuRD complex. MTA2 modulates the enzymatic activity of the histone deacetylase core complex. MBD3 mediates the association of MTA2 with the core histone deacetylase complex. MBD3 does not directly bind methylated DNA but is highly related to MBD2, a polypeptide that binds to methylated DNA and has been reported to possess demethylase activity. MBD2 interacts with the NuRD complex and directs the complex to methylated DNA. NuRD may provide a means of gene silencing by DNA methylation.

In this review most of the various known, suspected, or postulated functions of osteopontin, a secreted highly acidic phosphoprotein, are discussed in terms of what we currently know about the protein. These include 1) binding of OPN both to cells via a GRGDS cell adhesion sequence that recognizes the alpha v beta 3 integrin and to extracellular matrix components via poorly characterized motifs, 2) regulation of the formation and remodeling of mineralized tissue, 3) recruiting and stimulating macrophages and lymphocytes as part of a nonspecific response to microbial infections, 4) multiple interactions with Ca2+ that likely influence OPN protein conformation and may be important in Ca(2+)-mediated or Ca(2+)-dependent processes, 5) inhibiting the growth of calcium oxalate crystals by disruption of the growing crystal lattice, 6) effects on gene expression, Ca2+ regulation, and nitric oxide production, and 7) involvement in cell migration. OPN production is frequently augmented when cell signaling pathways are activated by any of a variety of stimuli, for example in cancer cells.

We show that spiral waves in galaxy discs churn the stars and gas in a manner that largely preserves the overall angular momentum distribution and leads to little increase in random motion. Changes in the angular momenta of individual stars are typically as large as ∼50 per cent over the lifetime of the disc. The changes are concentrated around the corotation radius for an individual spiral wave, but since transient waves with a wide range of pattern speeds develop in rapid succession, the entire disc is affected. This behaviour has profound consequences for the metallicity gradients with radius in both stars and gas, since the interstellar medium is also stirred by the same mechanism. We find observational support for stirring, propose a simple model for the distribution of stars over metallicity and age, and discuss other possible consequences.

Hypermethylation of CpG islands in the promoter regions is an important mechanism to silence the expression of many important genes in cancer. The hypermethylation status is passed to the daughter cells through the methylation of the newly synthesized DNA strand by 5-cytosine DNA methyltransferase (DNMT). We report herein that (-)-epigallocatechin-3-gallate (EGCG), the major polyphenol from green tea, can inhibit DNMT activity and reactivate methylation-silenced genes in cancer cells. With nuclear extracts as the enzyme source and polydeoxyinosine-deoxycytosine as the substrate, EGCG dose-dependently inhibited DNMT activity, showing competitive inhibition with a K(i) of 6.89 microM. Studies with structural analogues of EGCG suggest the importance of D and B ring structures in the inhibitory activity. Molecular modeling studies also support this conclusion, and suggest that EGCG can form hydrogen bonds with Pro(1223), Glu(1265), Cys(1225), Ser(1229), and Arg(1309) in the catalytic pocket of DNMT. Treatment of human esophageal cancer KYSE 510 cells with 5-50 microM of EGCG for 12-144 h caused a concentration- and time-dependent reversal of hypermethylation of p16(INK4a), retinoic acid receptor beta (RARbeta), O(6)-methylguanine methyltransferase (MGMT), and human mutL homologue 1 (hMLH1) genes as determined by the appearance of the unmethylation-specific bands in PCR. This was accompanied by the expression of mRNA of these genes as determined by reverse transcription-PCR. The re-expression of RARbeta and hMLH1 proteins by EGCG was demonstrated by Western blot. Reactivation of some methylation-silenced genes by EGCG was also demonstrated in human colon cancer HT-29 cells, esophageal cancer KYSE 150 cells, and prostate cancer PC3 cells. The results demonstrate for the first time the inhibition of DNA methylation by a commonly consumed dietary constituent and suggest the potential use of EGCG for the prevention or reversal of related gene-silencing in the prevention of carcinogenesis.

Finite element exterior calculus is an approach to the design and understanding of finite element discretizations for a wide variety of systems of partial differential equations. This approach brings to bear tools from differential geometry, algebraic topology, and homological algebra to develop discretizations which are compatible with the geometric, topological, and algebraic structures which underlie well-posedness of the PDE problem being solved. In the finite element exterior calculus, many finite element spaces are revealed as spaces of piecewise polynomial differential forms. These connect to each other in discrete subcomplexes of elliptic differential complexes, and are also related to the continuous elliptic complex through projections which commute with the complex differential. Applications are made to the finite element discretization of a variety of problems, including the Hodge Laplacian, Maxwell’s equations, the equations of elasticity, and elliptic eigenvalue problems, and also to preconditioners.

Tea is one of the most popular beverages consumed worldwide. The relationship between tea consumption and human cancer incidence is an important concern. This topic has been studied in different populations by many investigators, but no clear-cut conclusion can be drawn. Whereas some studies have shown a protective effect of tea consumption against certain types of cancers, other studies have indicated an opposite effect. Our purpose is to provide a critical review of this topic, covering basic chemistry and biochemical activity of tea, epidemiologic investigations, and laboratory studies, as well as possible directions for future research. Studies have demonstrated either a lack of association between tea consumption and cancer incidence at specific organ sites or inconsistent results. On the other hand, many laboratory studies have demonstrated inhibitory effects of tea preparations and tea polyphenols against tumor formation and growth. This inhibitory activity is believed to be mainly due to the antioxidative and possible antiproliferative effects of polyphenolic compounds in green and black tea. These polyphenolics may also inhibit carcinogenesis by blocking the endogenous formation of N-nitroso compounds, suppressing the activation of carcinogens, and trapping of genotoxic agents. The effect of tea consumption on cancer is likely to depend on the causative factors of the specific cancer. Therefore, a protective effect observed on a certain cancer with a specific population may not be observable with a cancer of a different etiology. On the basis of this concept, we suggest future laboratory and epidemiologic studies to elucidate the relationship between tea consumption and human cancer risk.