Multi-task learning using shared and task specific information

Multi-task learning solves multiple related learning problems simultaneously by sharing some common structure for improved generalization performance of each task. We propose a novel approach to multi-task learning which captures task similarity through a shared basis vector set. The variability across tasks is captured through task specific basis vector set. We use sparse support vector machine (SVM) algorithm to select the basis vector sets for the tasks. The approach results in a sparse model where the prediction is done using very few examples. The effectiveness of our approach is demonstrated through experiments on synthetic and real multi-task datasets.

Accurate prediction of interfacial residues in two-domain proteins using evolutionary information: Implications for three-dimensional modeling

With the preponderance of multidomain proteins in eukaryotic genomes, it is essential to recognize the constituent domains and their functions. Often function involves communications across the domain interfaces, and the knowledge of the interacting sites is essential to our understanding of the structure-function relationship. Using evolutionary information extracted from homologous domains in at least two diverse domain architectures (single and multidomain), we predict the interface residues corresponding to domains from the two-domain proteins. We also use information from the three-dimensional structures of individual domains of two-domain proteins to train naive Bayes classifier model to predict the interfacial residues. Our predictions are highly accurate (approximate to 85%) and specific (approximate to 95%) to the domain-domain interfaces. This method is specific to multidomain proteins which contain domains in at least more than one protein architectural context. Using predicted residues to constrain domain-domain interaction, rigid-body docking was able to provide us with accurate full-length protein structures with correct orientation of domains. We believe that these results can be of considerable interest toward rational protein and interaction design, apart from providing us with valuable information on the nature of interactions. Proteins 2014; 82:1219-1234. (c) 2013 Wiley Periodicals, Inc.

Mapping Post-translational Modifications of Mammalian Testicular Specific Histone Variant TH2B in Tetraploid and Haploid Germ Cells and Their Implications on the Dynamics of Nucleosome Structure

Histones regulate a variety of chromatin templated events by their post-translational modifications (PTMs). Although there are extensive reports on the PTMs of canonical histones, the information on the histone variants remains very scanty. Here, we report the identification of different PTMs, such as acetylation, methylation, and phosphorylation of a major mammalian histone variant TH2B. Our mass spectrometric analysis has led to the identification of both conserved and unique modifications across tetraploid spermatocytes and haploid spermatids. We have also computationally derived the 3-dimensional model of a TH2B containing nucleosome in order to study the spatial orientation of the PTMs identified and their effect on nucleosome stability and DNA binding potential. From our nucleosome model, it is evident that substititution of specific amino acid residues in TH2B results in both differential histone-DNA and histone-histone contacts. Furthermore, we have also observed that acetylation on the N-terminal tail of TH2B weakens the interactions with the DNA. These results provide direct evidence that, similar to somatic H2B, the testis specific histone TH2B also undergoes multiple PTMs, suggesting the possibility of chromatin regulation by such covalent modifications in mammalian male germ cells.

Enriching the annotation of Mycobacterium tuberculosis H37Rv proteome using remote homology detection approaches: Insights into structure and function

The availability of the genome sequence of Mycobacterium tuberculosis H37Rv has encouraged determination of large numbers of protein structures and detailed definition of the biological information encoded therein; yet, the functions of many proteins in M. tuberculosis remain unknown. The emergence of multidrug resistant strains makes it a priority to exploit recent advances in homology recognition and structure prediction to re-analyse its gene products. Here we report the structural and functional characterization of gene products encoded in the M. tuberculosis genome, with the help of sensitive profile-based remote homology search and fold recognition algorithms resulting in an enhanced annotation of the proteome where 95% of the M. tuberculosis proteins were identified wholly or partly with information on structure or function. New information includes association of 244 proteins with 205 domain families and a separate set of new association of folds to 64 proteins. Extending structural information across uncharacterized protein families represented in the M. tuberculosis proteome, by determining superfamily relationships between families of known and unknown structures, has contributed to an enhancement in the knowledge of structural content. In retrospect, such superfamily relationships have facilitated recognition of probable structure and/or function for several uncharacterized protein families, eventually aiding recognition of probable functions for homologous proteins corresponding to such families. Gene products unique to mycobacteria for which no functions could be identified are 183. Of these 18 were determined to be M. tuberculosis specific. Such pathogen-specific proteins are speculated to harbour virulence factors required for pathogenesis. A re-annotated proteome of M. tuberculosis, with greater completeness of annotated proteins and domain assigned regions, provides a valuable basis for experimental endeavours designed to obtain a better understanding of pathogenesis and to accelerate the process of drug target discovery. (C) 2014 Elsevier Ltd. All rights reserved.

Exploring the Crystal Structure Landscape with a Heterosynthon Module: Fluorobenzoic Acid:1,2-Bis(4-pyridyl)ethylene 2:1 Cocrystals

The crystal structure landscape of the 2:1 benzoic acid:dipyridylethylene cocrystal (BA:DPE-I) is explored experimentally with fluoro-substituted benzoic acids and extended with studies employing the Cambridge Structural Database (CSD). The interpretation of the cocrystal landscape is facilitated by considering the kinetically favored and robust acidpyridine heterosynthon as a modular unit. Information based on high-throughput crystallography shows that polymorphs and pseudopolymorphs may belong to the same landscape but arise from different crystallization pathways because of complex and different kinetic features, and secondary synthon preferences. Using the CSD as a guide, the coformer was changed from 1,2-bis(4-pyridyl)ethylene (DPE-I) to 1,2-bis(4-pyridyl)ethane (DPE-II) and this provides an extended interpretation of the BA:DPE-I cocrystal landscape, also highlighting the complexity of the kineticthermodynamic dichotomy during the molecule-to-crystal progression.

Airblast spray in crossflow - Structure, trajectory and droplet sizing

This study reports results of an experimental investigation of airblast spray of water and ethanol in crossflow. Laser shadowgraphy and Particle/Droplet Imaging Analysis (PDIA) are used to derive spray trajectory and drop size information while Particle Tracking Velocimetry (PTV) is used to measure droplet velocities. A new phenomenon of spray bifurcation is observed for low Gas to Liquid Ratio (GLR) cases. The reasons for the spatial bifurcation can be attributed to a combination of reasons. These are (a) presence of large ligaments and droplets in the near-nozzle region for low GLRs (b) secondary breakup experienced by ligaments/droplets leading to formation of a large number of small droplets, and (c) the crossflow causing differential dispersion of the small and large droplets. A novel correlation for spray trajectory is proposed incorporating the momentum ratio and liquid surface tension. This correlation is shown to be effective in predicting the non-linear spray trajectory over a large range of conditions for not only water but ethanol and Jet-A also. It is observed that the larger droplets penetrate further into the crossflow, in the direction of injection. Thus, with increase in height of the measurement location from the injection plane, the droplet Sauter Mean Diameter (SMD) is found to increase. Moreover, as the droplets travel downstream in the crossflow direction, the droplet SMD is observed to decrease. The effect of drag is assessed by comparing velocity of different sizes of droplets at various locations. Smaller droplets are entrained into the crossflow at much lower elevations, whereas larger droplets tend to penetrate further into the crossflow. (C) 2015 Elsevier Ltd. All rights reserved.

Differences in host species relationships and biogeographic influences produce contrasting patterns of prevalence, community composition and genetic structure in two genera of avian malaria parasites in southern Melanesia

1. Host-parasite interactions have the potential to influence broadscale ecological and evolutionary processes, levels of endemism, divergence patterns and distributions in host populations. Understanding the mechanisms involved requires identification of the factors that shape parasite distribution and prevalence. 2. A lack of comparative information on community-level host-parasite associations limits our understanding of the role of parasites in host population divergence processes. Avian malaria (haemosporidian) parasites in bird communities offer a tractable model system to examine the potential for pathogens to influence evolutionary processes in natural host populations. 3. Using cytochrome b variation, we characterized phylogenetic diversity and prevalence of two genera of avian haemosporidian parasites, Plasmodium and Haemoproteus, and analysed biogeographic patterns of lineages across islands and avian hosts, in southern Melanesian bird communities to identify factors that explain patterns of infection. 4. Plasmodium spp. displayed isolation-by-distance effects, a significant amount of genetic variation distributed among islands but insignificant amounts among host species and families, and strong local island effects with respect to prevalence. Haemoproteus spp. did not display isolation-by-distance patterns, showed marked structuring of genetic variation among avian host species and families, and significant host species prevalence patterns. 5. These differences suggest that Plasmodium spp. infection patterns were shaped by geography and the abiotic environment, whereas Haemoproteus spp. infection patterns were shaped predominantly by host associations. Heterogeneity in the complement and prevalence of parasite lineages infecting local bird communities likely exposes host species to a mosaic of spatially divergent disease selection pressures across their naturally fragmented distributions in southern Melanesia. Host associations for Haemoproteus spp. indicate a capacity for the formation of locally co-adapted host-parasite relationships, a feature that may limit intraspecific gene flow or range expansions of closely related host species.

The Structure of the Holo-Acyl Carrier Protein of Leishmania major Displays a Remarkably Different Phosphopantetheinyl Transferase Binding Interface

The genome of Leishmania major encodes a type II fatty acid biosynthesis pathway for which no structural or biochemical information exists. Here, for the first time, we have characterized the central player of the pathway, the acyl carrier protein (LmACP), using nuclear magnetic resonance (NMR). Structurally, the LmACP molecule is similar to other type II ACPs, comprising a four-helix bundle, enclosing a hydrophobic core. Dissimilarities in sequence, however, exist in helix II (recognition helix) of the protein. The enzymatic conversion of apo-LmACP into the holo form using type I (Escherichia coli AcpS) and type II (Sfp type) phosphopantetheinyl transferases (PPTs) is relatively slow. Mutagenesis studies underscore the importance of the residues present at the protein protein interaction interface of LmACP in modulating the activity of PPTs. Interestingly, the cognate PPT for this ACP, the L. major 4'-phosphopantetheinyl transferase (LmPPT), does not show any enzymatic activity toward it, though it readily converts other type I and type II ACPs into their holo forms. NMR chemical shift perturbation studies suggest a moderately tight complex between LmACP and its cognate PPT, suggesting inhibition. We surmise that the unique surface of LmACP might have evolved to complement its cognate enzyme (LmPPT), possibly for the purpose of regulation.

Timing Calculations for a General N-Level Dodecagonal Space Vector Structure Using Only Reference Phase Voltages

Multilevel inverters with dodecagonal (12-sided polygon) voltage space vector (SV) structures have advantages like extension of linear modulation range, elimination of fifth and seventh harmonics in phase voltages and currents for the full modulation range including extreme 12-step operation, reduced device voltage ratings, lesser dv/dt stresses on devices and motor phase windings resulting in lower EMI/EMC problems, and lower switching frequency-making it more suitable for high-power drive applications. This paper proposes a simple method to obtain pulsewidth modulation (PWM) timings for a dodecagonal voltage SV structure using only sampled reference voltages. In addition to this, a carrier-based method for obtaining the PWM timings for a general N-level dodecagonal structure is proposed in this paper for the first time. The algorithm outputs the triangle information and the PWM timing values which can be set as the compare values for any carrier-based hardware PWM module to obtain SV PWM like switching sequences. The proposed method eliminates the need for angle estimation, computation of modulation indices, and iterative search algorithms that are typical in multilevel dodecagonal SV systems. The proposed PWM scheme was implemented on a five-level dodecagonal SV structure. Exhaustive simulation and experimental results for steady-state and transient conditions are presented to validate the proposed method.

Residue proximity information and protein model discrimination using saturation-suppressor mutagenesis

Identification of residue-residue contacts from primary sequence can be used to guide protein structure prediction. Using Escherichia coli CcdB as the test case, we describe an experimental method termed saturation-suppressor mutagenesis to acquire residue contact information. In this methodology, for each of five inactive CcdB mutants, exhaustive screens for suppressors were performed. Proximal suppressors were accurately discriminated from distal suppressors based on their phenotypes when present as single mutants. Experimentally identified putative proximal pairs formed spatial constraints to recover >98% of native-like models of CcdB from a decoy dataset. Suppressor methodology was also applied to the integral membrane protein, diacylglycerol kinase A where the structures determined by X-ray crystallography and NMR were significantly different. Suppressor as well as sequence co-variation data clearly point to the Xray structure being the functional one adopted in vivo. The methodology is applicable to any macromolecular system for which a convenient phenotypic assay exists.

A Non-Parametric Dimension Test of the Term Structure

Published as an article in: Studies in Nonlinear Dynamics & Econometrics, 2004, vol. 8, issue 3, article 6.

Endogenous Formation of Competing Partnership with Moral Hazard

Published as an article in: Games and Economic Behavior, 2003, vol. 44, issue 1, pages 183-194.

The Metabolic Core and Catalytic Switches Are Fundamental Elements in the Self-Regulation of the Systemic Metabolic Structure of Cells

19 p.