Structured Learning for Non-Smooth Ranking losses

Learning to rank from relevance judgment is an active research area. Itemwise score regression, pairwise preference satisfaction, and listwise structured learning are the major techniques in use. Listwise structured learning has been applied recently to optimize important non-decomposable ranking criteria like AUC (area under ROC curve) and MAP(mean average precision). We propose new, almost-lineartime algorithms to optimize for two other criteria widely used to evaluate search systems: MRR (mean reciprocal rank) and NDCG (normalized discounted cumulative gain)in the max-margin structured learning framework. We also demonstrate that, for different ranking criteria, one may need to use different feature maps. Search applications should not be optimized in favor of a single criterion, because they need to cater to a variety of queries. E.g., MRR is best for navigational queries, while NDCG is best for informational queries. A key contribution of this paper is to fold multiple ranking loss functions into a multi-criteria max-margin optimization.The result is a single, robust ranking model that is close to the best accuracy of learners trained on individual criteria. In fact, experiments over the popular LETOR and TREC data sets show that, contrary to conventional wisdom, a test criterion is often not best served by training with the same individual criterion.

The Nature of Bond Critical Points in Dinuclear Copper(I) Complexes

Closed-shell contacts between two copper(I) ions are expected to be repulsive. However, such contacts are quite frequent and are well documented. Crystallographic characterization of such contacts in unsupported and bridged multinuclear copper(I) complexes has repeatedly invited debates on the existence of cuprophilicity. Recent developments in the application of Baders theory of atoms-in-molecules (AIM) to systems in which weak hydrogen bonds are involved suggests that the copper(I)copper(I) contacts would benefit from a similar analysis. Thus the nature of electron-density distributions in copper(I) dimers that are unsupported, and those that are bridged, have been examined. A comparison of complexes that are dimers of symmetrical monomers and those that are dimers of two copper(I) monomers with different coordination spheres has also been made. AIM analysis shows that a bond critical point (BCP) between two Cu atoms is present in most cases. The nature of the BCP in terms of the electron density, ?, and its Laplacian is quite similar to the nature of critical points observed in hydrogen bonds in the same systems. The ? is inversely correlated to Cu?Cu distance. It is higher in asymmetrical systems than what is observed in corresponding symmetrical systems. By examining the ratio of the local electron potential-energy density (Vc) to the kinetic energy density (Gc), |Vc|/Gc at the critical point suggests that these interactions are not perfectly ionic but have some shared nature. Thus an analysis of critical points by using AIM theory points to the presence of an attractive metallophilic interaction similar to other well-documented weak interactions like hydrogen bonding.

Enhanced non-homologous end joining contributes toward synthetic lethality of pathological RAD51C mutants with poly (ADP-ribose) polymerase

Here, we show that PARP inhibitor-mediated cell death of RAD51C-deficient cells occur by NHEJ-driven illegitimate repair of one-ended double-strand breaks, and the hypomorphic RAD51C pathological mutant cells can be targeted by `synergistic toxicity' induced by low-dose PARP inhibitor and IR.Poly (ADP-ribose) polymerase 1 (PARP1) inhibitors are actively under clinical trials for the treatment of breast and ovarian cancers that arise due to mutations in BRCA1 and BRCA2. The RAD51 paralog RAD51C has been identified as a breast and ovarian cancer susceptibility gene. The pathological RAD51C mutants that were identified in cancer patients are hypomorphic with partial repair function. However, targeting cancer cells that express hypomorphic mutants of RAD51C is highly challenging. Here, we report that RAD51C-deficient cells can be targeted by a `synthetic lethal' approach using PARP inhibitor and this sensitivity was attributed to accumulation of cells in the G(2)/M and chromosomal aberrations. In addition, spontaneous hyperactivation of PARP1 was evident in RAD51C-deficient cells. Interestingly, RAD51C-negative cells exhibited enhanced recruitment of non-homologous end joining (NHEJ) proteins onto chromatin and this accumulation correlated with increased activity of error-prone NHEJ as well as genome instability leading to cell death. Notably, inhibition of DNA-PKcs or depletion of KU70 or Ligase IV rescued this phenotype. Strikingly, stimulation of NHEJ by low dose of ionizing radiation (IR) in the PARP inhibitor-treated RAD51C-deficient cells and cells expressing pathological RAD51C mutants induced enhanced toxicity `synergistically'. These results demonstrate that cancer cells arising due to hypomorphic mutations in RAD51C can be specifically targeted by a `synergistic approach' and imply that this strategy can be potentially applied to cancers with hypomorphic mutations in other homologous recombination pathway genes.

Ligand 5,10,15,20-Tetra(N-methyl-4-pyridyl)porphine (TMPyP4) Prefers the Parallel Propeller-Type Human Telomeric G-Quadruplex DNA over Its Other Polymorphs

The binding of ligand 5,10,15,20-tetra(N-methyl-4-pyridyl)porphine (TMPyP4) with telomeric and genomic G-quadruplex DNA has been extensively studied. However, a comparative study of interactions of TMPyP4 with different conformations of human telomeric G-quadruplex DNA, namely, parallel propeller-type (PP), antiparallel basket-type (AB), and mixed hybrid-type (MH) G-quadruplex DNA, has not been done. We considered all the possible binding sites in each of the G-quadruplex DNA structures and docked TMPyP4 to each one of them. The resultant most potent sites for binding were analyzed from the mean binding free energy of the complexes. Molecular dynamics simulations were then carried out, and analysis of the binding free energy of the TMPyP4-G-quadruplex complex showed that the binding of TMPyP4 with parallel propeller-type G-quadruplex DNA is preferred over the other two G-quadruplex DNA conformations. The results obtained from the change in solvent excluded surface area (SESA) and solvent accessible surface area (SASA) also support the more pronounced binding of the ligand with the parallel propeller-type G-quadruplex DNA.

Study of three elastic-plastic constitutive models by non-proportional finite deformations of OFHC copper

Experimental stress-strain data of OFHC copper first under torsion to 13% and then under torsion-tension to about 10% are used to study the characteristics of three elastic-plastic constitutive models: Chaboche's super-positional nonlinear model, Dafalias and Popov's two surface model and Watanabe and Atluri's version of the endochronic model. The three models, originally oriented for infinitesimal deformation, have been extended for finite deformation. The results show (a) the Mises-type yield surface used in the three models brings about significant departure of the predictions from the experimental data; (b) Chaboche's and Dafalias' models are easier than Watanabe and Atluri's model in determining the material parameters in them, and (c) Chaboche's and Watanabe & Atluri's models produce almost the same prediction to the data, while Dafalias' model cannot accurately predict the plastic deformations when a loading path changes in its direction. Copyright (C) 1996 Elsevier Science Ltd

Electron transfer in cytochrome c oxidase : the rate of electron equillibration between cytochrome a and copper A

Intramolecular electron transfer in partially reduced cytochrome c oxidase has been studied by means of perturbed equilibrium techniques. We have prepared a three electron reduced, CO inhibited form of the enzyme in which cytochrome a and copper A are partially reduced an in intramolecular redox equilibrium. When these samples were photolyzed using a nitrogen laser (0.6 µs, 1.0 mJ pulses) changes in absorbance at 598 nm and 830 nm were observed which are consistent with a fast electron from cytochrome a to copper A. The absorbance changes at 598 nm have an apparent rate of 17,200 ± 1,700 s^(-1) (1σ), at pH 7.0 and 25.5 °C. These changes were not observed in either the CO mixed valence or CO inhibited fully reduced forms of the enzyme. The rate is fastest at about pH 8.0, and falls off in either direction, and there is a small, but clear temperature dependence. The process was also observed in the cytochrome c -- cytochrome c oxidase high affinity complex.

This rate is far faster than any rate measured or inferred previously for the cytochrome a -- copper A electron equilibration, but the interpretation of these results is hampered by the fact that the relaxation could only be followed during the time before CO became rebound to the oxygen binding site. The meaning of our our measured rate is discussed, along with other reported rates for this process. In addition, a temperature-jump experiment on the same system is discussed.

We have also prepared a partially reduced, cyanide inhibited form of the enzyme in which cytochrome a, copper A and copper B are partially reduced and in redox equilibrium. Warming these samples produced absorbance changes at 605 nm which indicate that cytochrome a was becoming more oxidized, but there were no parallel changes in absorbance at 830 nm as would be expected if copper A was becoming reduced. We concluded that electrons were being redistributed from cytochrome a to copper B. The kinetics of the absorbance changes at 605 nm were investigated by temperature-jump methods. Although a rate could not be resolved, we concluded that the process must occur with an (apparent) rate larger than 10,000 s^(-1).

During the course of the temperature-jump experiments, we also found that non-redox related, temperature dependent absorbance changes in fully reduced CO inhibited cytochrome c oxidase, and in the cyanide mixed valence enzyme, took place with an (apparent) rate faster that 30,000 s^(-1).

Electrochemical and electron transfer investigations of copper proteins

A study of the pH and temperature dependence of the redox potentials of azurins from five species of bacteria has been performed. The variations in the potentials with pH have been interpreted in terms of electrostatic interactions between the copper site and titrating histidine residues, including the effects of substitutions in the amino acid sequences of the proteins on the electrostatic interactions. A comparison of the observed pH dependences with predictions based on histidine pK_a values known for Pseudomonas aeruginosa (Pae), Alcaligenes denitrificans (Ade), and Alcaligenes faecalis (Afa) azurins indicates that the Pae and Ade redox potentials exhibit pH dependences in line with electrostatic arguments, while Afa azurin exhibits more complex behavior. Redox enthalpies and entropies for four of the azurins at low and high pH values have also been obtained. Based on these results in conjuction with the variable pH experiments, it appears that Bordetella bronchiseptica azurin may undergo a more substantial conformational change with pH than has been observed for other species of azurin.

The temperature dependence of the redox potential of bovine erythrocyte superoxide dismutase (SOD) has been determined at pH 7.0, with potassium ferricyanide as the mediator. The following thermodynamic parameters have been obtained (T = 25°C): E°' = 403±5 mV vs. NHE, ΔG°' = -9.31 kcal/mol, ΔH°' = -21.4 kcal/mol, ΔS°' = -40.7 eu, ΔS°'_(rc) = -25.1 eu. It is apparent from these results that ΔH°', rather than ΔS°', is the dominant factor in establishing the high redox potential of SOD. The large negative enthalpy of reduction may also reflect the factors which give SOD its high specificity toward reduction and oxidation by superoxide.

Sequence specific complexation of b DNA at sites containing g,c base pairs

A series of eight related analogs of distamycin A has been synthesized. Footprinting and affinity cleaving reveal that only two of the analogs, pyridine-2- car box amide-netropsin (2-Py N) and 1-methylimidazole-2-carboxamide-netrops in (2-ImN), bind to DNA with a specificity different from that of the parent compound. A new class of sites, represented by a TGACT sequence, is a strong site for 2-PyN binding, and the major recognition site for 2-ImN on DNA. Both compounds recognize the G•C bp specifically, although A's and T's in the site may be interchanged without penalty. Additional A•T bp outside the binding site increase the binding affinity. The compounds bind in the minor groove of the DNA sequence, but protect both grooves from dimethylsulfate. The binding evidence suggests that 2-PyN or 2-ImN binding induces a DNA conformational change.

In order to understand this sequence specific complexation better, the Ackers quantitative footprinting method for measuring individual site affinity constants has been extended to small molecules. MPE•Fe(II) cleavage reactions over a 10^5 range of free ligand concentrations are analyzed by gel electrophoresis. The decrease in cleavage is calculated by densitometry of a gel autoradiogram. The apparent fraction of DNA bound is then calculated from the amount of cleavage protection. The data is fitted to a theoretical curve using non-linear least squares techniques. Affinity constants at four individual sites are determined simultaneously. The distamycin A analog binds solely at A•T rich sites. Affinities range from 10^(6)- 10^(7)M^(-1) The data for parent compound D fit closely to a monomeric binding curve. 2-PyN binds both A•T sites and the TGTCA site with an apparent affinity constant of 10^(5) M^(-1). 2-ImN binds A•T sites with affinities less than 5 x 10^(4) M^(-1). The affinity of 2-ImN for the TGTCA site does not change significantly from the 2-PyN value. At the TGTCA site, the experimental data fit a dimeric binding curve better than a monomeric curve. Both 2-PyN and 2-ImN have substantially lower DNA affinities than closely related compounds.

In order to probe the requirements of this new binding site, fourteen other derivatives have been synthesized and tested. All compounds that recognize the TGTCA site have a heterocyclic aromatic nitrogen ortho to the N or C-terminal amide of the netropsin subunit. Specificity is strongly affected by the overall length of the small molecule. Only compounds that consist of at least three aromatic rings linked by amides exhibit TGTCA site binding. Specificity is only weakly altered by substitution on the pyridine ring, which correlates best with steric factors. A model is proposed for TGTCA site binding that has as its key feature hydrogen bonding to both G's by the small molecule. The specificity is determined by the sequence dependence of the distance between G's.

One derivative of 2-PyN exhibits pH dependent sequence specificity. At low pH, 4-dimethylaminopyridine-2-carboxamide-netropsin binds tightly to A•T sites. At high pH, 4-Me_(2)NPyN binds most tightly to the TGTCA site. In aqueous solution, this compound protonates at the pyridine nitrogen at pH 6. Thus presence of the protonated form correlates with A•T specificity.

The binding site of a class of eukaryotic transcriptional activators typified by yeast protein GCN4 and the mammalian oncogene Jun contains a strong 2-ImN binding site. Specificity requirements for the protein and small molecule are similar. GCN4 and 2-lmN bind simultaneously to the same binding site. GCN4 alters the cleavage pattern of 2-ImN-EDTA derivative at only one of its binding sites. The details of the interaction suggest that GCN4 alters the conformation of an AAAAAAA sequence adjacent to its binding site. The presence of a yeast counterpart to Jun partially blocks 2-lmN binding. The differences do not appear to be caused by direct interactions between 2-lmN and the proteins, but by induced conformational changes in the DNA protein complex. It is likely that the observed differences in complexation are involved in the varying sequence specificity of these proteins.

Associação entre variantes do gene da adiponectina (ADIPOQ): -11391 G/A, -11377 C/G, 45T>G e I164T e os níveis circulantes de adiponectina de alto peso molecular e fatores de risco cardiometabólico

A adiponectina, um hormônio produzido pelo tecido adiposo, atua na regulação do metabolismo energético e interfere favoravelmente na sensibilidade à insulina através de suas ações no fígado e musculatura esquelética. Ao contrário da maioria das outras adipocitocinas, associa-se inversamente com a obesidade visceral, resistência à insulina, diabetes tipo 2 e doença cardiovascular. Inúmeros estudos demonstraram nos últimos anos os efeitos de variantes genéticas no gene ADIPOQ sobre os níveis circulantes de adiponectina, resistência à insulina, diabetes e obesidade. Entretanto, além de resultados contraditórios, a maior parte desses estudos foi realizada em populações Caucasianas e Asiáticas. Avaliar, em uma população multiétnica adulta do município do Rio de Janeiro, as possíveis associações das variantes genéticas (-11391 G>A, -11377C>G, +45T>G e T517G) no gene ADIPOQ com o fenótipo obeso, níveis circulantes de adiponectina de alto peso molecular e fatores de risco cardiometabólico. Trata-se de um estudo transversal. Foram estudados 100 indivíduos eutróficos (IMC 18,5 24,9 kg/m2, idade: 32,5 + 9,8 anos) e 100 obesos (IMC 30 58,2 kg/m2, idade 37,5 + 14,1 anos), igualmente divididos entre homens e mulheres. Os indivíduos obesos apresentaram valores significativamente maiores de circunferência abdominal, pressão arterial sistólica, diastólica e média, glicemia de jejum, triglicerídeos, LDL-colesterol, leptina, insulina, HOMA-IR e proteína C reativa, quando comparados aos eutróficos. Contrariamente, exibiram menores valores de adiponectina e HDL-colesterol. Análises de correlação mostraram relação inversa e significativa entre a adiponectina, circunferência abdominal, insulina, HOMA-IR e pressão arterial. Com os níveis de HDL-colesterol, a correlação foi positiva. Por meio de análise de regressão múltipla foi possível identificar os determinantes dos níveis séricos de adiponecinta. Sexo masculino, circunferência abdominal, HOMA-IR e a variante genética -11391G>A, foram os principais responsáveis por essa variação, com um R2 de 30%. Quanto à análise genética, não encontramos nenhuma associação entre essas variantes e o fenótipo obeso. Entretanto, os indivíduos carreadores do alelo mutante -11391A apresentaram menores valores de glicemia, pressão arterial e relação cintura-quadril e maiores concentrações sanguíneas de adiponectina, quando comparados aos indivíduos ditos selvagens. Ademais, os carreadores do alelo mutante -11377G apresentaram menores valores de pressão arterial sistólica, diastólica e média. Os resultados do presente estudo demonstram que níveis de adiponectina diferem entre eutróficos e obesos e que concentrações mais baixas dessa adipocitocina estão associadas a um pior perfil cardiometabólico. Variantes no gene ADIPOQ podem interferir nessa relação e alguns polimorfismos parecem ter um perfil protetor no risco cardiovascular.

Concentrations of Polychlorinated Biphenyls (PCB’s), Chlorinated Pesticides, and Heavy Metals and Other Elements in Tissues of Belugas, Delphinapterus leucas, from Cook Inlet

Tissues from Cook Inlet beluga whales, Delphinapterus leucas, that were collected as part of the Alaska Marine Mammal Tissue Archival Project were analyzed for polychlorinated biphenyls (PCB’s), chlorinated pesticides, and heavy metals and other elements. Concentrations of total PCB’s (ΣPCB’s), total DDT (ΣDDT), chlordane compounds, hexachlorobenzene (HCB), dieldrin, mirex, toxaphene, and hexachlorocyclohexane (HCH) measured in Cook Inlet beluga blubber were compared with those reported for belugas from two Arctic Alaska locations (Point Hope and Point Lay), Greenland, Arctic Canada, and the highly contaminated stock from the St. Lawrence estuary in eastern Canada. The Arctic and Cook Inlet belugas had much lower concentrations (ΣPCB’s and ΣDDT were an order of magnitude lower) than those found in animals from the St. Lawrence estuary. The Cook Inlet belugas had the lowest concentrations of all (ΣPCB’s aver-aged 1.49 ± 0.70 and 0.79 ± 0.56 mg/kg wet mass, and ΣDDT averaged 1.35 ± 0.73 and 0.59 ± 0.45 mg/kg in males and females, respectively). Concentrations in the blubber of the Cook Inlet males were significantly lower than those found in the males of the Arctic Alaska belugas (ΣPCB’s and ΣDDT were about half). The lower levels in the Cook Inlet animals might be due to differences in contaminant sources, food web differences, or different age distributions among the animals sampled. Cook Inlet males had higher mean and median concentrations than did females, a result attributable to the transfer of these compounds from mother to calf during pregnancy and during lactation. Liver concentrations of cadmium and mercury were lower in the Cook Inlet belugas (most cadmium values were <1 mg/kg and mercury values were 0.704–11.42 mg/kg wet mass), but copper levels were significantly higher in the Cook Inlet animals (3.97–123.8 mg/kg wet mass) than in Arctic Alaska animals and similar to those reported for belugas from Hudson Bay. Although total mercury levels were the lowest in the Cook Inlet population, methylmercury concentrations were similar among all three groups of the Alaska animals examined (0.34–2.11 mg/kg wet mass). As has been reported for the Point Hope and Point Lay belugas, hepatic concentrations of silver were re

Concentration and distribution of heavy metals (lead, cadmium, chromium, copper, nickel, zinc) in Karachi shore and offshore sediments

A study to measure the heavy metal pollution level in the sediment of coastal and offshore area indicates that high concentration of heavy metals were found around Manora channel and eastern coast of Karachi. In comparison with coastal areas, relatively low concentration of heavy metals was recorded in the offshore area. The result shows that sewage and industrial wastes are the main source of heavy metal pollution in the coastal area. The concentration of heavy metals in the sediments is as follows: Chromium 10.4-33.69, nickel 13.3-47.6, lead 10.0-39.04, cadmium 0.08-0.21, zinc 7.4-73.2 and copper 9.44-18.56 mg/kg. In the offshore areas strong correlation was observed between copper and organic carbon, and calcium carbonate and cadmium. In the shore area such correlation has been recorded among nickel, chromium, zinc, and chromium and copper. The Karachi. coast is viewed as moderately polluted when compared to other continental coastal areas.

Energy scattering in weakly non-linear systems.

The Chinese Tam-Tam exhibits non-linear behavior in its vibro-acoustic response. The frequency content of the response during free, unforced vibration smoothly changes, with energy being progressively smeared out over a greater bandwidth with time. This is used as a motivating case for the general study of the phenomenon of energy cascading through weak nonlinearity. Numerical models based upon the Fermi-Pasta-Ulam system of non-linearly coupled oscillators, modified with the addition of damping, have been developed. These were used to study the response of ensembles of systems with randomized natural frequencies. Results from simulations will be presented here. For un-damped systems, individual ensemble members exhibit cyclical energy exchange between linear modes, but the ensemble average displays a steady state. For the ensemble response of damped systems, lightly damped modes can exhibit an effective damping which is higher than predicated by linear theory. The presence of a non-linearity provides a path for energy flow to other modes, increasing the apparent damping spectrum at some frequencies and reducing it at others. The target of this work is a model revealing the governing parameters of a generic system of this type and leading to predictions of the ensemble response.