Surrogate Regret Bounds for Bipartite Ranking via Strongly Proper Losses

The problem of bipartite ranking, where instances are labeled positive or negative and the goal is to learn a scoring function that minimizes the probability of mis-ranking a pair of positive and negative instances (or equivalently, that maximizes the area under the ROC curve), has been widely studied in recent years. A dominant theoretical and algorithmic framework for the problem has been to reduce bipartite ranking to pairwise classification; in particular, it is well known that the bipartite ranking regret can be formulated as a pairwise classification regret, which in turn can be upper bounded using usual regret bounds for classification problems. Recently, Kotlowski et al. (2011) showed regret bounds for bipartite ranking in terms of the regret associated with balanced versions of the standard (non-pairwise) logistic and exponential losses. In this paper, we show that such (non-pairwise) surrogate regret bounds for bipartite ranking can be obtained in terms of a broad class of proper (composite) losses that we term as strongly proper. Our proof technique is much simpler than that of Kotlowski et al. (2011), and relies on properties of proper (composite) losses as elucidated recently by Reid and Williamson (2010, 2011) and others. Our result yields explicit surrogate bounds (with no hidden balancing terms) in terms of a variety of strongly proper losses, including for example logistic, exponential, squared and squared hinge losses as special cases. An important consequence is that standard algorithms minimizing a (non-pairwise) strongly proper loss, such as logistic regression and boosting algorithms (assuming a universal function class and appropriate regularization), are in fact consistent for bipartite ranking; moreover, our results allow us to quantify the bipartite ranking regret in terms of the corresponding surrogate regret. We also obtain tighter surrogate bounds under certain low-noise conditions via a recent result of Clemencon and Robbiano (2011).

Milking Diatoms for Sustainable Energy: Biochemical Engineering versus Gasoline-Secreting Diatom Solar Panels

In the face of increasing CO2 emissions from conventional energy (gasoline), and the anticipated scarcity of Crude oil, a worldwide effort is underway for cost-effective renewable alternative energy sources. Here, we review a simple line of reasoning: (a) geologists claim that Much crude oil comes from diatoms; (b) diatoms do indeed make oil; (c) agriculturists Claim that diatoms could make 10-200 times as much oil per hectare as oil seeds; and (d) therefore, sustainable energy could be made from diatoms. In this communication, we propose ways of harvesting oil from diatoms, using biochemical engineering and also a new solar panel approach that utilizes genomically modifiable aspects of diatom biology, offering the prospect of ``milking'' diatoms for Sustainable energy by altering them to actively secrete oil products. Secretion by and milking of diatoms may provide a way around the puzzle of how to make algae that both grow quickly and have a very high oil content.

Modulating the preferred O-H center dot center dot center dot O hydrogen-bonding motif in a conformationally constrained environment through hydroxy-group derivatization

The crystal structures of three conformationally locked esters, namely the centrosymmetric tetrabenzoate of all-axial per-hydronaphthalene- 2,3,4a, 6,7,8a-hexaol, viz. trans-4a, 8a-dihydroxyperhydronaphthalene-2,3,6,7-tetrayl tetrabenzoate, C38H34O10, and the diacetate and dibenzoate of all-axial perhydronaphthalene-2,3,4a, 8a-tetraol, viz. (2R*,3R*,4aS*,8aS*)-4a, 8a-dihydroxyperhydronaphthalene-2,3-diyl diacetate, C-14-H22O6, and (2R*, 3R*, 4aS*, 8aS*)-4a, 8a-dihydroxyperhydronaphthalene- 2,3-diyl dibenzoate, C24H26O6, have been analyzed in order to examine the preference of their supramolecular assemblies towards competing inter-and intramolecular O-H center dot center dot center dot O hydrogen bonds. It was anticipated that the supramolecular assembly of the esters under study would adopt two principal hydrogen-bonding modes, namely one that employs intermolecular O-H center dot center dot center dot O hydrogen bonds (mode 1) and another that sacrifices those for intramolecular O-H center dot center dot center dot O hydrogen bonds and settles for a crystal packing dictated by weak intermolecular interactions alone (mode 2). Thus, while the molecular assembly of the two crystalline diacyl derivatives conformed to a combination of hydrogen-bonding modes 1 and 2, the crystal packing in the tetrabenzoate preferred to follow mode 2 exclusively.

Crystal Structures of a New Polymorphic Form of Gabapentin Monohydrate and the E and Z Isomers of 4-Tertiarybutylgabapentin

Gabapentin, a widely used antiepileptic drug, crystallizes in multiple polymorphic forms. A new crystal form of gabapentin monohydrate in the space group Pbca is reported and the packing arrangement compared with that of a previously reported polymorph in the space group P2(1)/c [Ibers, J.A. (2001) Acta Crystallogr; C57:641]. Gabapentin polymorphs can also occur from a selection of one of the two distinct chair forms of the 1,1-disubstituted cyclohexane. Crystal structures of the E and Z isomers of 4-tert-butylgabapentin provide models for analyzing anticipated packing modes in the conformational isomers of gabapentin. The E isomer crystallized in the space group Pca2(1), while the Z isomer crystallized in the space group P2(1)/c. The crystal structure of E-4-tert-butylgabapentin provides the only example of a structure in a non-centrosymmetric space group. Crystal structures of the hydrochloride and hydrobromide salts of 4-tert-butyl derivatives are reported. The results suggest that for gabapentin, a large 'polymorph-space' may be anticipated, in view of the multiple conformational states that are accessible to the molecule.

Characterization of the Entamoeba histolytica Ornithine Decarboxylase-Like Enzyme

Background: The polyamines putrescine, spermidine, and spermine are organic cations that are required for cell growth and differentiation. Ornithine decarboxylase (ODC), the first and rate-limiting enzyme in the polyamine biosynthetic pathway, is a highly regulated enzyme. Methodology and Results: To use this enzyme as a potential drug target, the gene encoding putative ornithine decarboxylase (ODC)-like sequence was cloned from Entamoeba histolytica, a protozoan parasite causing amoebiasis. DNA sequence analysis revealed an open reading frame (ORF) of similar to 1,242 bp encoding a putative protein of 413 amino acids with a calculated molecular mass of 46 kDa and a predicted isoelectric point of 5.61. The E. histolytica putative ODC-like sequence has 33% sequence identity with human ODC and 36% identity with the Datura stramonium ODC. The ORF is a single-copy gene located on a 1.9-Mb chromosome. The recombinant putative ODC protein (48 kDa) from E. histolytica was heterologously expressed in Escherichia coli. Antiserum against recombinant putative ODC protein detected a band of anticipated size similar to 46 kDa in E. histolytica whole-cell lysate. Difluoromethylornithine (DFMO), an enzyme-activated irreversible inhibitor of ODC, had no effect on the recombinant putative ODC from E. histolytica. Comparative modeling of the three-dimensional structure of E. histolytica putative ODC shows that the putative binding site for DFMO is disrupted by the substitution of three amino acids-aspartate-332, aspartate-361, and tyrosine-323-by histidine-296, phenylalanine-305, and asparagine-334, through which this inhibitor interacts with the protein. Amino acid changes in the pocket of the E. histolytica enzyme resulted in low substrate specificity for ornithine. It is possible that the enzyme has evolved a novel substrate specificity. Conclusion: To our knowledge this is the first report on the molecular characterization of putative ODC-like sequence from E. histolytica. Computer modeling revealed that three of the critical residues required for binding of DFMO to the ODC enzyme are substituted in E. histolytica, resulting in the likely loss of interactions between the enzyme and DFMO.

Multiple conformational states in crystals and in solution in alpha gamma hybrid peptides. Fragility of the C-12 helix in short sequences

The conformational properties of foldamers generated from alpha gamma hybrid peptide sequences have been probed in the model sequence Boc-Aib-Gpn-Aib-Gpn-NHMe. The choice of alpha-aminoisobutyryl (Aib) and gabapentin (Gpn) residues greatly restricts sterically accessible coil formational space. This model sequence was anticipated to be a short segment of the alpha gamma C-12 helix, stabilized by three successive 4 -> 1 hydrogen bonds, corresponding to a backbone-expanded analogue of the alpha polypeptide 3(10)-helix. Unexpectedly, three distinct crystalline polymorphs were characterized in the solid state by X-ray diffraction. In one form, two successive C-12 hydrogen bonds were obtained at the N-terminus, while a novel C-17 hydrogen-bonded gamma alpha gamma turn was observed at the C-terminus. In the other two polymorphs, isolated C-9 and C-7 hydrogen-bonded turns were observed at Gpn (2) and Gpn (4). Isolated C-12 and C-9 turns were also crystallographically established in the peptides Boc-Aib-Gpn-Aib-OMe and Boc-Gpn-Aib-NHMe, respectively. Selective line broadening of NH resonances and the observation of medium range NH(i)<-> NH(i+2) NOEs established the presence of conformational heterogeneity for the tetrapeptide in CDCl3 solution. The NMR results are consistent with the limited population of the continuous C-12 helix conformation. Lengthening of the (alpha gamma)(n) sequences in the nonapeptides Boc-Aib-Gpn-Aib-Gpn-Aib-Gpn-Aib-Gpn-Xxx (Xxx = Aib, Leu) resulted in the observation of all of the sequential NOEs characteristic of an alpha gamma C-12 helix. These results establish that conformational fragility is manifested in short hybrid alpha gamma sequences despite the choice of conformationally constrained residues, while stable helices are formed on chain extension.

Helical Conformations of Hexapeptides Containing N-Terminus Diproline Segments

The role of N-terminus diproline segments in facilitating helical folding in short peptides has been investigated in a set of model hexapeptides of the type Piv-Xxx-Yyy-Aib-Leu-Aib-Phe-OMe (Piv, pivaloyl). Nine sequences have been investigated with the following N-terminus dipeptide segments: (D)Pro-Ala (4) and Pro-Psi Pro (5, Psi, pseudoproline), Ala-Ala (6), Ala-Pro (7), Pro-Ala (8), Aib-Ala (9), Ala-Aib (10). The analog sequences Piv-Pro-Pro-Ala-Leu-Aib-Phe-OMe (2) and Piv-Pro-Pro-Ala-Aib-Ala-Aib-OMe (3) have also been studied. Solid state conformations have been determined by X-ray crystallography for peptides 4, 6, and 8 and compared with the previously determined crystal structure of peptide 1 (Boc-Pro-Pro-Aib-Leu-Aib-Val-OMe); (Rai et al., JACS 2006, 128, 7916-7928). Peptides 1 and 6 adopt almost identical helical conformations with unfolding of the helix at the N-terminus Pro (1) residue. Peptide 4 reveals the anticipated (D)Pro-Ala type II' beta-turn, followed by a stretch of 3(10)-helix. Peptide 8 adopts a folded conformation stabilized by four successive 4 -> 1 intramolecular hydrogen bonds. Ala (2) adopts an alpha(L) conformation, resulting in a type II beta-turn conformation followed by a stretch of 3(10)-helix. Conformational properties in solution were probed using solvent perturbation of NH chemical shifts which permit delineation of hydrogen bonded NH groups and nuclear Overhauser effects (NOEs) between backbone protons, which are diagnostic of local residue conformations. The results suggest that continuous helical conformations are indeed significantly populated for peptides 2 and 3. Comparison of the results for peptides 1 and 2, suggest that there is a significant influence of the residue that follows diproline segments in influencing backbone folding. (C) 2010 Wiley Periodicals, Inc. Biopolymers (Pept Sci) 94: 360-370, 2010.

Liquid Limit Of Kaolinitic Soils

This paper elucidates the relative importance of soil structure and various compositional factors in influencing the liquid limit of natural kaolinitic soils. Earlier studies dealt with purified systems and anticipated that the liquid limit of the soils would increase with percentage clay size fraction and surface area, and that soils with a greater degree of paricle flocculation would possess a higher liquid limit than soils with a more parallel particle arrangement. The results revealed that the inter-particle attraction and repulsion forces have a prominent role in determining the liquid limit of kaolinitic soils. These forces determine the particle arrangement (clay fabric) which in turn regulates the liquid limit values. The influence of clay fabric was ascertained from the relationships of liquid limit with shrinkage limit and sediment volume (in water) values. It was anticipated that kaolinitic soils with a greater degree of particle flocculatin and a higher liquid limit would shrink less and occupy a higher sedimentation volume. As expected an increase in liquid limit was accompanied by an increase in shrinkage limit and sediment volume in water.

Crystallization and preliminary X-ray studies of a galactose-specific lectin from the seeds of bitter gourd (Momordica charantia)

A galactose-specific lectin from the seeds of bitter gourd (Momordica charantia) is a four-chain type II ribosome-inactivating protein (RIP) resulting from covalent association through a disulfide bridge between two identical copies of a two-chain unit. The available structural information on such four-chain RIPs is meagre. The bitter gourd lectin was therefore crystallized for structural investigation and the crystals have been characterized. It is anticipated that the structure of the orthorhombic crystals will be analysed using molecular replacement by taking advantage of its sequence, and presumably structural, homology to normal two-chain type II RIPs.

Study of the growth of capped ZnO nanocrystals: A route to rational synthesis

We report the study of complex and unexpected dependencies of nanocrystal size as well as nanocrystalsize distribution on various reaction parameters in the synthesis of ZnO nanocrystals using poly(vinyl pyrollidone) (PVP) as a capping agent. This method establishes a qualitatively different growth mechanism to the anticipated Ostwald ripening behavior. The study of size-distribution kinetics and an understanding of the observed non-monotonic behaviors provides a route to rational synthesis. We used a simple, but accurate, approach to estimate the size-distribution function of nanocrystals from the UV-absorption spectrum. Our results demonstrate the accuracy and generality of this approach, and we also illustrate its application to various semiconducting nanocrystals, such as ZnO, ZnS, and CdSe, over a wide size range (1.8-5.3 nm).

Radical Annulation Strategy to Chiral Pupukeanones: Total Synthesis of (+)-10-ex&l(- Naphthyl)pupukean-9-onea nd (+)-lO-exo-l(- Naphthyl)-5-epipupukean-9-ones1

Intramolecular alkylation reaction of the bromoenone 12, obtained from S-carvone in three steps, furnished the bicyclo[2.2.2]octenone 13. Contrary to the anticipated radical annulation reaction, the bicyclic bromides 14 and 15, obtained from the enone 13, generated exclusively the cyclopropane product 18 via a 3-exo-trig radical cyclization on reaction with nBu3SnH and AIBN, even in the presence of a large excess of a radicophile. On the other hand, bromoenone 24, synthesized from R-carvone via S-naphthylcarvone 21, underwent radical annulation reaction in the presence of radicophiles to furnish the isotwistanes 25-28 in a regio- and stereospecific manner. Hydrogenation of the olefin 34, obtained from the diketone 27 via a regiospecific Wittig reaction, furnished the naphthyl-5-epipupukean-9-one 33, whereas stereoselective hydrogenation of the enone 36, prepared from the keto ester 25 via a Grignard reaction and dehydration sequence, generated the naphthylpupukeanone 32.

Hybrid Polypeptides: Gabapentin as a Stereochemically Constrained gamma-Amino Acid Residue

The design of folded structures in peptides containing the higher homologues of alpha-amino acid residues requires the restriction of the range of local conformational choices In alpha-amino acids stereochemically constrained residues like alpha,alpha-dialkylated residue, aminoisobutyric acid (Aib), and D-Proline ((D)Pro) have proved extremely useful in the design of helices and hairpins in short peptides Extending this approach, backbone substitution and cyclization are anticipated to bc useful in generating conformationally constrained beta- and gamma-residues This brief review provides a survey of work on hybrid peptide sequences concerning the conformationally constrained gamma-amino acid residue 1-aminomethyl cyclohexane acetic acid, gabapentin (Gpn) This achiral, beta,beta-disubstituted, gamma-residue strongly favors gauche-gauche conformations about the C-alpha-C-beta (0(2)) and C-alpha-C-gamma (0(1)) bonds, facilitating local folding The Gpn residue can adopt both C-7 (NH1 -> CO1) and C-9 (CO1 (I)<- NH1+I) hydrogen bonds which are analogous to the C-5 and C7 (gamma-turn) conformations at alpha-residues In conjunction with adjacent residues, Gpn may be used in ay and gamma alpha segments to generate C-12 hydrogen bonded conformations which may be considered as expanded analogs of conventional beta-turns The structural characterization of C-12 helices, C-12/C-10 helices with mixed hydrogen bond directionalities and beta-hairpins incorporating Gpn residues at the turn segment is illustrated (C) 2010 Wiley Periodicals, Inc Biopolymers (Pept Sci) 94 733-741 2010

A non-cooperative game-theoretic approach to channel assignment in multi-channel multi-radio wireless networks

Channel assignment in multi-channel multi-radio wireless networks poses a significant challenge due to scarcity of number of channels available in the wireless spectrum. Further, additional care has to be taken to consider the interference characteristics of the nodes in the network especially when nodes are in different collision domains. This work views the problem of channel assignment in multi-channel multi-radio networks with multiple collision domains as a non-cooperative game where the objective of the players is to maximize their individual utility by minimizing its interference. Necessary and sufficient conditions are derived for the channel assignment to be a Nash Equilibrium (NE) and efficiency of the NE is analyzed by deriving the lower bound of the price of anarchy of this game. A new fairness measure in multiple collision domain context is proposed and necessary and sufficient conditions for NE outcomes to be fair are derived. The equilibrium conditions are then applied to solve the channel assignment problem by proposing three algorithms, based on perfect/imperfect information, which rely on explicit communication between the players for arriving at an NE. A no-regret learning algorithm known as Freund and Schapire Informed algorithm, which has an additional advantage of low overhead in terms of information exchange, is proposed and its convergence to the stabilizing outcomes is studied. New performance metrics are proposed and extensive simulations are done using Matlab to obtain a thorough understanding of the performance of these algorithms on various topologies with respect to these metrics. It was observed that the algorithms proposed were able to achieve good convergence to NE resulting in efficient channel assignment strategies.

Centers of Excellence in Nanoelectronics in India

Nanoelectronics is considered an emerging area all over the world and is widely anticipated to hold the key to the future electronic innovations. Realizing its importance, the Government of India has set up two Centers of Excellence in Nanoelectronics (CEN) one at Indian Institute of Science (IISc), Bangalore and the other at Indian Institute of Technology Bombay (IITB) in 2006. The first phase of this program is going to be over in December 2010 and the second phase is going to continue. This paper discusses the objectives of the project, activities, research areas, students training, various research activities, Incubation and Entrepreneurship activity, and interaction with Academic institutions / industries both at National and International level.

The intercalation reaction of pyridine with manganese thiophosphate, MnPS3

The intercalation of pyridine in the layered manganese thiophosphate, MnPS3, has been examined in detail by a variety of techniques. The reaction is interesting since none of the anticipated changes in optical and electrical properties associated with intercalation of electron donating molecules is observed. The only notable change in the properties of the host lattice is in the nature of the low-temperature magnetic ordering; while MnPS3 orders antiferromagnetically below 78 K, the intercalated compound shows weak ferromagnetism probably due to a canted spin structure. Vibrational spectra clearly show that the intercalated species are pyridinium ions solvated by neutral pyridine molecules. The corresponding reduced sites of the host lattice, however, were never observed. The molecules in the solvation shell are exchangeable. Although the reaction appears to be topotactic and reversible, from XRD, a more detailed analysis of the products of deintercalation reveal that it is not so. The intercalation proceeds by an ion exchange/intercalation mechanism wherein the intercalated species are pyridinium ions solvated by neutral molecules with charge neutrality being preserved not by electron transfer but by a loss of Mn2+ ions from the lattice. The experimental evidence leading to this conclusion is discussed and it is shown that this model can account satisfactorily for the observed changes (or lack of it) in the optical, electrical, vibrational, and magnetic properties.