166 research outputs found
A4NT : Author Attribute Anonymity by Adversarial Training of Neural Machine Translation
No full textText-based analysis methods enable an adversary to reveal privacy relevant author attributes such as gender, age and can identify the text's author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called the Adversarial Author Attribute Anonymity Neural Translation (), to combat such text-based adversaries. Unlike prior works on obfuscation, we propose a system that is fully automatic and learns to perform obfuscation entirely from the data. This allows us to easily apply the system to obfuscate different author attributes. We propose a sequence-to-sequence language model, inspired by machine translation, and an adversarial training framework to design a system which learns to transform the input text to obfuscate the author attributes without paired data. We also propose and evaluate techniques to impose constraints on our model to preserve the semantics of the input text. learns to make minimal changes to the input to successfully fool author attribute classifiers, while preserving the meaning of the input text. Our experiments on two datasets and three settings show that the proposed method is effective in fooling the attribute classifiers and thus improves the anonymity of authors
Fresh-pack potatoes: handling, packaging and transportation in refrigerated railcars
No full textBulletin no. 804 Moscow, Idaho :University of Idaho, College of Agriculture, Agriculture Experiment Station, 1998-07-01. Author(s): Shetty, Kiran; Casada, Mark; Zhu, Hua; Thornton, Mike; Nolte, Phili
Differential involvement of Ca2+/calmodulin-dependent protein kinases and mitogen-activated protein kinases in the dopamine D1/D5 receptor-mediated potentiation in hippocampal CA1 pyramidal neurons
No full textAbstractDopaminergic neurotransmission modulates and influences hippocampal CA1 synaptic plasticity, learning and long-term memory mechanisms. Investigating the mechanisms involved in the slow-onset potentiation induced by the dopamine D1/D5 receptor agonists in hippocampal CA1 region, we have reported recently that it could play a role in regulating synaptic cooperation and competition. We have also shown that a sustained activation of MEK/MAP kinase pathway was involved in the maintenance of this long-lasting potentiation (Shivarama Shetty, Gopinadhan, & Sajikumar, 2016). However, the molecular aspects of the induction of dopaminergic slow-onset potentiation are not known. Here, we investigated the involvement of MEK/MAPK pathway and Ca2+-calmodulin-dependent protein kinases (CaMKII and CaMKIV) in the induction and maintenance phases of the D1/D5 receptor-mediated slow-onset potentiation. We report differential involvement of these kinases in a dose-dependent manner wherein at weaker levels of dopaminergic activation, both CaMKII and MEK1/2 activation is necessary for the establishment of potentiation and with sufficiently stronger dopaminergic activation, the role of CaMKII becomes dispensable whereas MEK activation remains crucial for the long-lasting potentiation. The results are interesting in view of the involvement of the hippocampal dopaminergic system in a variety of cognitive abilities including memory formation and also in neurological diseases such as Alzheimer’s disease and Parkinson’s disease
Structure Analysis Of Plant Lectin Domains
No full textLectins are multivalent carbohydrate binding proteins that specifically recognise diverse sugar structures and mediate a variety of biological processes, such as cell-cell and host-pathogen interactions, serum glycoprotein turnover and innate immune responses. Lectins have received considerable attention in recent years on account of their properties leading to wide use in research and biomedical applications. Seeds of leguminous plants are mainly rich sources of lectins, but lectins are also found in all classes and families of organisms. Legume lectins have similar tertiary structures, but exhibit a large variety of quaternary structures. The carbohydrate binding site in them is made up of four loops, the first three of which are highly conserved in all legume lectins. The fourth loop, which is variable, is implicated in conferring specificity. Legume lectins which share the same monosaccharide specificity often exhibit markedly different oligosaccharide specificities. This thesis primarily concerns with structure solution and analysis of lectins from the legume and β-prism II fold families using X-ray crystallography. Apart from having the property of specifically and reversibly binding to carbohydrates, lectins are also interesting models to study sequence-structure relationships, especially of how minor change in the sequence may bring about major changes in oligomerization and binding.
Chapter 1 gives an overview of different structural types of plant lectins and describes in detail, their carbohydrate binding features. The details of the various experimental procedures employed during the course of this research, are explained in Chapter 2.
Chapter 3 describes the crystal structure of a β-prism II fold lectin (RVL), from Remusatia vivipara, an epiphytic plant of traditional medicinal value, and analysis of its binding properties. This lectin was established to have distinct binding properties and has nematicidal activity against a root-knot nematode with the localization site identified as the high-mannose displaying gut-lining in the nematode. The crystal structure of RVL revealed a new quaternary association of this homodimeric lectin, different from those of reported β-prism II lectins. Functional studies on RVL showed that it fails to bind to simple mannose moieties yet showed agglutination with rabbit blood cells (which have mannose moieties on the surface) and some high mannose containing glycoproteins like mucin and asialofetuin. Further, ELISA and glycan array experiments indicated that RVL has high affinity to N-glycans like trimannose pentasaccharide such as in gp120, a capsid glycoprotein of HIV virus, necessary in virus-association with the host cell. The structural basis for this N-glycan binding was revealed through structure analysis and molecular modelling, and it was demonstrated that there are two distinct binding sites per monomer, making RVL a truly multivalent lectin. Evolutionary phylogeny revealed the divergence in the β-prism II fold proteins with regards to the number of sugar-binding regions per domain, oligomerization and specificity.
Chapter 4 deals with the structural studies on a galactose-specific legume lectin (DLL-II) from Dolichos lablab, a leguminous plant. The lectin was found to be a planar tetramer in the crystal structures of the native and ligand bound forms, as expected from our solution studies and phylogenetic analysis. The protein is a heterotetramer with subunits differing only in the presence or absence of a C-terminal helical region at the core of the tetramer. Due to the static disorder in all the crystals, the central helix could be oriented in either direction. Structure analysis of DLL-II proved to be an interesting endeavour as static disorder compounded with twinning in the crystal made the data processing and structure solution a challenging process. Subsequent structure and sequence alignments led to the identification of an adenine-binding pocket in the hydrophobic core of the tetramer. Based on this, DLL-II lectin was co-crystallized with adenine and the structure revealed the presence of adenine at the predicted binding site.
Chapter 5 describes the identification and analysis of potential plant lectins/lectin-like domains in the genome of Oryza sativa, using bioinformatics approaches. This project was initiated to study the occurrence of legume-lectin like domains (a predominant dicot feature) in O. sativa, which is a monocot. Later, a large scale genome analysis for all types of lectin domains was carried out through exhaustive PSI-BLAST, profile matching by HMMer, CDD and MulPSSM. The final validation was carried out by assessing the carbohydrate binding potential of the domain by examining the sugar binding sites. The primary interest in undertaking this work was to find the occurrence of association of these domains with other domains as in protein receptor kinases, where lectin is the receptor domain. Though primarily initiated as a bioinformatics project, further structural characterization was attempted by cloning, expression and purification of some of the annotated lectin proteins using prokaryotic expression systems. The protein expression was attained in reasonable amounts for a few of the annotated legume lectin homologs, however purification is yet to be achieved as the expressed proteins are insoluble.
A part of the results described in this thesis and the other related projects that the author was involved are reported in the following publications.
1) Purification, characterization and molecular cloning of a monocot mannose-binding lectin from Remusatia vivipara with nematicidal activity Bhat GG, Shetty KN, Nagre NN, Neekhra VV, Lingaraju S, Bhat RS, Inamdar SR, Suguna K, Swamy BM. 2010. Glycoconjugate J. 27(3):309-320
2) Modification of the sugar specificity of a plant lectin: structural studies on a point mutant of Erythrina corallodendron lectin Thamotharan S, Karthikeyan T, Kulkarni KA, Shetty KN, Surolia A, Vijayan M & Suguna K. 2011. Acta Crystallographica D 67(3):218-227
3) Crystal structure of a β-prism II lectin from Remusatia vivipara Shetty KN, Bhat GG, Inamdar SR, Swamy BM, Suguna K. 2012. Glycobiology 22(1): 56-69.
4) Structure of a galactose binding lectin from Dolichos lablab
Shetty KN, Lavanyalatha V, Rao RN, SivaKumar N & Suguna K (Under review)
5) Occurrence of lectin-like domains: Oryza sativa genome analysis. Shetty KN & Suguna K. (Manuscript in preparation
Weighted K-nearest neighbor algorithm as an object localization technique using passive RFID tags
No full textTechnologies using identification by radio frequencies (RFID) are experiencing rapid development and healthcare is a major application area benefiting from it. Highly pervasive RFID enables remote identification, tracking and localization of the medical staff, patients, medications and equipment, thus increasing safety, optimizing in real-time management and providing support for new ambient-intelligent services. This thesis describes and evaluates an algorithm that enables object localization and tracking using passive RFID tags. This thesis also describes scenarios of how this technology can be used as a part of building a smart trauma resuscitation room by tracking the equipments. The main contribution of this thesis is the adaptation of the Weighted K-Nearest Neighbor Algorithm as a localization technique to track objects in a confined and crowded space by using passive RFID tags. The input parameter to the algorithm is the received signal strength indicator (RSSI), which gives a measure of back-scattered radio frequencies from passive tags. While using RFID technology special attention has to be given to the placement of antennas to get the optimum result. Therefore, we analyzed various antenna placement configurations with mean error and error consistency as the two performance parameters. The detection of multiple tags and human occlusion are two major concerns while tracking tags in a confined space with many team members collaborating on solving a problem. The RF signal can be interrupted by people walking around randomly and holding multiple (tagged) instruments at the same time. While the algorithm worked fine when tracking multiple tags, we had to modify the experimental set-up and attach an antenna onto the ceiling (which we call a vertical antenna), so that even if all the wall antennas are blocked we get at least one input parameter to base our localization decision on. We evaluated the algorithm for different combinations of configurations and number of neighbors, and achieved the following results. The best results were obtained for the 3 antennae (placed orthogonally) configuration considering the 4 nearest neighbors wherein a mean error rate of 15% of the maximum possible error was achieved under ideal conditions. We tested the algorithm for different human occlusion scenarios i.e. blocking 1 or 2 wall antennas, standing in random positions and then roaming in the field area randomly. The mean error rate for the standing scenario was measured as 20% of the maximum possible error and 18% in the case of roaming configuration. The error was found to be consistently within our defined maximum error for 100% of the recorded readings. The results obtained were found to be satisfactory for our application where, more than the exact location of the object, knowing whether the object is within a particular region is good enough for the users to know what task is being carried out in the trauma bay. Also the algorithm holds good in an indoor environment having a lot of factors and materials which affect the RF signal disrupting accurate calculation of the location co-ordinates. The algorithm does not require extensive data collection prior to implementation which makes it easily deployable in any environment. Apart from the problems mentioned there are some other factors like materials on which the tags are attached and orientation of tags which were found to be potential hindrances for accurate localization. Acceptable solutions to these problems form a part of our future work.M.S.Includes bibliographical referencesby Akshay Shett
Effect of roughness, microstructure, and chemistry on the environmental durability of structural alloys
No full textUnlike functional materials where one design property is usually optimized, structural materials need to meet several design requirements including but not limited to excellent mechanical properties and environmental tolerance. However, more often than not, the metallurgical principles used to improve mechanical properties result in a decrease in the environmental tolerance. Though it is difficult to design materials that can thrive in extreme environments like high temperature, high pressure, and harsh chemicals, these conditions are often where the biggest scientific and economic opportunities lie. Aeroengines, and oil refineries are two example applications that encounter these issues. This study specifically focusses on improving the environmental tolerance of NiCrAl films a common bond coat used in aeroengines, and ferrous alloys used in petroleum production. The goal was to develop solutions that improve the environmental tolerance while minimally affecting other properties of these alloys.
The oxidation behavior of a model sputtered NiCrAl system was studied in various time and oxygen partial pressure (pO2) regimes. Low pO2 conditions seemed to favor the formation of protective oxides. However, tuning the composition of the base alloy was an effective way to limit the oxidation rate in high pO2 conditions. Thermal destabilization of the sputtered microstructure was found to take place on similar timescales to the transient oxide formation. Thus dilute Y additions were made to temporarily stabilize the sputtered microstructure and manipulate the transient stages of oxidation.
Yttrium addition not only retarded grain growth through nanoclustering and kinetically pinning grains, but also helped nucleate and grow dense, and slow-growing oxides. In some cases, the improvement in oxidation resistance via. Y addition was similar to reducing the pO2 by several orders of magnitude. Robust α-alumina oxides formed on Y doped NiCrAl films at temperatures as low as 900 oC by oxidation in an air environment which is unprecedented and could be of major commercial importance. An attempt was made to understand this anomalous oxidation behavior by using unconventional diffusion-triples comprising of a sputtered NiCr (undoped and Y doped) top layer, Al middle layer, and a sintered (micrograined) NiCr bottom layer. Annealing experiments conducted on the diffusion-triples proved that Al diffusion in sputtered NiCr is more rapid than that in sintered NiCr. Through the use of profile processing techniques, Al was shown to follow type B kinetics for grain boundary diffusion in sputtered NiCr. It also revealed that Y addition to sputtered NiCr further accelerates Al diffusion through a non-Fickian mechanism involving Al clustering.
The baseline fouling and corrosion behavior of ferrous alloys in a high temperature, high pressure, and an asphaltenic environment was also evaluated. Key insights were generated on the interplay between the thermochemical properties of the asphaltene, the environmental conditions, surface preparation of the alloys, and the chemistry of the deposits. X-ray photoelectron spectroscopy (XPS) and transmission electron microscopy allowed for the first time to pinpoint mechanisms for high temperature model asphaltene deposition on ferrous alloys. Improving surface roughness alone was found to be a good strategy to mitigate asphaltenic fouling at lower temperatures where the asphaltene remains intact. However, at temperatures where reactive asphaltene decomposition products become present in solution, surface chemistry control becomes important. Specifically, a protective atomic layer deposition alumina chemistry on steels was found to significantly reduce asphaltenic fouling.
In order to evaluate whether a protective alumina chemistry could be generated on components with more complex geometries, low temperature pack cementation of ferrous alloys was conducted. Preliminary data did show an improvement in anti-fouling properties with both model asphaltenes in a static environment, as well as with crude oil in a hydrodynamic environment. However, XPS revealed a mixed alumina-hematite oxide on the surface that may be limiting the anti-fouling properties of these surfaces. Finally, new insights gained from developing low temperature pack cementation for ferrous alloys allowed for the modification of low thermal stability, functional metallic structures like Ni inverse opals. It resulted in a thermal stability enhancement by 500 oC, comparable to refractory metals in the same configuration. And also improved both the modulus and hardness of these structures.Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2021-05-01The student, Pralav Shetty, accepted the attached license on 2019-02-20 at 15:57.The student, Pralav Shetty, submitted this Dissertation for approval on 2019-02-20 at 16:17.This Dissertation was approved for publication on 2019-02-22 at 10:53.DSpace SAF Submission Ingestion Package generated from Vireo submission #13398 on 2019-08-22 at 16:19:59Made available in DSpace on 2019-08-23T20:44:32Z (GMT). No. of bitstreams: 3
SHETTY-DISSERTATION-2019.pdf: 7968379 bytes, checksum: ca85cff311d013ddea30fe2cea1479fc (MD5)
LICENSE.txt: 4218 bytes, checksum: df11f963d29bcf46ca57412adc74e761 (MD5)
PROQUEST_LICENSE.txt: 4564 bytes, checksum: f55f0d715958bd9d08662fc3ee35aec3 (MD5)
Previous issue date: 2019-02-22Embargo set by: Seth Robbins for item 112255
Lift date: 2021-08-23T20:44:50Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 112255
Lift date: 2021-08-23T20:46:41Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 112255
Lift date: 2021-08-23T20:47:38Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 112255
Lift date: 2021-08-23T20:48:32Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemLimited Restriction Lifted for Item 112255 on 2021-08-24T09:15:31Z
A4NT: Author Attribute Anonymity by Adversarial Training of Neural Machine Translation
No full textText-based analysis methods enable an adversary to reveal privacy relevant author attributes such as gender, age and can identify the text's author. Such methods can compromise the privacy of an anonymous author even when the author tries to remove privacy sensitive content. In this paper, we propose an automatic method, called the Adversarial Author Attribute Anonymity Neural Translation (), to combat such text-based adversaries. Unlike prior works on obfuscation, we propose a system that is fully automatic and learns to perform obfuscation entirely from the data. This allows us to easily apply the system to obfuscate different author attributes. We propose a sequence-to-sequence language model, inspired by machine translation, and an adversarial training framework to design a system which learns to transform the input text to obfuscate the author attributes without paired data. We also propose and evaluate techniques to impose constraints on our model to preserve the semantics of the input text. learns to make minimal changes to the input to successfully fool author attribute classifiers, while preserving the meaning of the input text. Our experiments on two datasets and three settings show that the proposed method is effective in fooling the attribute classifiers and thus improves the anonymity of authors
Effect of beneficial microorganisms on nutrient profile of Moringa oleifera leaves.
No full textThis Dissertation / Report is the outcome of investigation carried out by the creator(s) / author(s) at the department/division of Central Food Technological Research Institute (CFTRI), Mysore mentioned below in this page
Trienzyme extraction and quantification of total folate from underutilized fruits.
No full textThis Dissertation / Report is the outcome of investigation carried out by the creator(s) / author(s) at the department/division of Central Food Technological Research Institute (CFTRI), Mysore mentioned below in this page
- …
