Typing mineral deposits using their grades and tonnages in an artificial neural network

A test of the ability of a probabilistic neural network to classify deposits into types on the basis of deposit tonnage and average Cu, Mo, Ag, Au, Zn, and Pb grades is conducted. The purpose is to examine whether this type of system might serve as a basis for integrating geoscience information available in large mineral databases to classify sites by deposit type. Benefits of proper classification of many sites in large regions are relatively rapid identification of terranes permissive for deposit types and recognition of specific sites perhaps worthy of exploring further. Total tonnages and average grades of 1,137 well-explored deposits identified in published grade and tonnage models representing 13 deposit types were used to train and test the network. Tonnages were transformed by logarithms and grades by square roots to reduce effects of skewness. All values were scaled by subtracting the variable's mean and dividing by its standard deviation. Half of the deposits were selected randomly to be used in training the probabilistic neural network and the other half were used for independent testing. Tests were performed with a probabilistic neural network employing a Gaussian kernel and separate sigma weights for each class (type) and each variable (grade or tonnage). Deposit types were selected to challenge the neural network. For many types, tonnages or average grades are significantly different from other types, but individual deposits may plot in the grade and tonnage space of more than one type. Porphyry Cu, porphyry Cu-Au, and porphyry Cu-Mo types have similar tonnages and relatively small differences in grades. Redbed Cu deposits typically have tonnages that could be confused with porphyry Cu deposits, also contain Cu and, in some situations, Ag. Cyprus and kuroko massive sulfide types have about the same tonnages. Cu, Zn, Ag, and Au grades. Polymetallic vein, sedimentary exhalative Zn-Pb, and Zn-Pb skarn types contain many of the same metals. Sediment-hosted Au, Comstock Au-Ag, and low-sulfide Au-quartz vein types are principally Au deposits with differing amounts of Ag. Given the intent to test the neural network under the most difficult conditions, an overall 75% agreement between the experts and the neural network is considered excellent. Among the largestclassification errors are skarn Zn-Pb and Cyprus massive sulfide deposits classed by the neuralnetwork as kuroko massive sulfides—24 and 63% error respectively. Other large errors are the classification of 92% of porphyry Cu-Mo as porphyry Cu deposits. Most of the larger classification errors involve 25 or fewer training deposits, suggesting that some errors might be the result of small sample size. About 91% of the gold deposit types were classed properly and 98% of porphyry Cu deposits were classes as some type of porphyry Cu deposit. An experienced economic geologist would not make many of the classification errors that were made by the neural network because the geologic settings of deposits would be used to reduce errors. In a separate test, the probabilistic neural network correctly classed 93% of 336 deposits in eight deposit types when trained with presence or absence of 58 minerals and six generalized rock types. The overall success rate of the probabilistic neural network when trained on tonnage and average grades would probably be more than 90% with additional information on the presence of a few rock types.

Structure-function studies of the plant cyclotides: The role of a circular protein backbone

The traditional idea of proteins as linear chains of amino acids is being challenged with the discovery of miniproteins that contain a circular backbone. The cyclotide family is the largest group of circular proteins and is characterized by an amide-circularized protein backbone and six conserved cysteine residues. These conserved cysteines are paired to form a knotted network of disulfide bonds. The combination of the circular backbone and a cystine knot, known as the cyclic cystine knot (CCK) motif, confers exceptional stability upon the cyclotides. This review discusses the role of the circular backbone based on studies of both the oxidative folding of kalata B1, the prototypical cyclotide, and a comparison of the structure and activity of kalata B1 and its acyclic permutants.

Disturbed microtubule function and induction of micronuclei by chelate complexes of mercury(II)

Interactions of mercury(II) with the microtubule network of cells may lead to genotoxicity. Complexation of mercury(II) with EDTA is currently being discussed for its employment in detoxification processes of polluted sites. This prompted us to re-evaluate the effects of such complexing agents on certain aspects of mercury toxicity, by examining the influences of mercury(H) complexes on tubulin assembly and kinesin-driven motility of microtubules. The genotoxic effects were studied using the micronucleus assay in V79 Chinese hamster fibroblasts. Mercury(II) complexes with EDTA and related chelators interfered dose-dependently with tubulin assembly and microtubule motility in vitro. The no-effect-concentration for assembly inhibition was 1muM of complexed Hg(II), and for inhibition of motility it was 0.05 muM, respectively. These findings are supported on the genotoxicity level by the results of the micronucleus assay, with micronuclei being induced dose-dependently starting at concentrations of about 0.05 muM of complexed Hg(II). Generally, the no-effect-concentrations for complexed mercury(II) found in the cell-free systems and in cellular assays (including the micronucleus test) were identical with or similar to results for mercury tested in the absence of chelators. This indicates that mercury(II) has a much higher affinity to sulfhydryls of cytoskeletal proteins than to this type of complexing agents. Therefore, the suitability of EDTA and related compounds for remediation of environmental mercury contamination or for other detoxification purposes involving mercury has to be questioned. (C) 2004 Elsevier B.V. All rights reserved.

The role of disulfide bonds in the structure and function of murine epidermal growth factor (mEGF)

A systematic study using solid phase peptide synthesis has been undertaken to examine the role of the disulfide bonds in the structure and function of mEGF. A combination of one, two and three native disulfide pair analogues of an active truncated (4-48) form of mEGF have been synthesised by replacing specific cysteine residues with isosteric alpha-amino-n-butyric acid (Abu). Oxidation of the peptides was performed using either conventional aerobic oxidation at basic pH, in DMSO under acidic conditions or via selective disulfide formation using orthogonal protection of the cysteine pairs. The contribution of individual, or pairs of, disulfide bonds to EGF structure was evaluated by CD and H-1-NMR spectroscopy. The mitogenic activity of each analogue was determined using Balb/c 3T3 mouse fibroblasts. As we have reported previously (Barnham et al. 1998), the disulfide bond between residues 6 and 20 can be removed with significant retention of biological activity (EC50 20-50 nM). The overall structure of this analogue was similar to that of native mEGF, indicating that the loss of the 6-20 disulfide bridge did not affect the global fold of the molecule. We now show that removal of any other disulfide bond, either singly or in pairs, results in a major disruption of the tertiary structure, and a large loss of activity (EC50>900 nM). Remarkably, the linear analogue appears to have greater activity (EC50 580 nM) than most one and two disulfide bond analogues although it does not have a definable tertiary structure.

Genotoxicity of inorganic lead salts and disturbance of microtubule function

Lead compounds are known genotoxicants, principally affecting the integrity of chromosomes. Lead chloride and lead acetate induced concentration-dependent increases in micronucleus frequency in V79 cells, starting at 1.1 μ M lead chloride and 0.05 μ M lead acetate. The difference between the lead salts, which was expected based on their relative abilities to form complex acetato-cations, was confirmed in an independent experiment. CREST analyses of the micronuclei verified that lead chloride and acetate were predominantly aneugenic (CREST-positive response), which was consistent with the morphology of the micronuclei (larger micronuclei, compared with micronuclei induced by a clastogenic mechanism). The effects of high concentrations of lead salts on the microtubule network of V79 cells were also examined using immunofluorescence staining. The dose effects of these responses were consistent with the cytotoxicity of lead(II), as visualized in the neutral-red uptake assay. In a cell-free system, 20-60 μ M lead salts inhibited tubulin assembly dose-dependently. The no-observed-effect concentration of lead(II) in this assay was 10 μ M. This inhibitory effect was interpreted as a shift of the assembly/disassembly steady-state toward disassembly, e.g., by reducing the concentration of assembly-competent tubulin dimers. The effects of lead salts on microtubule-associated motor-protein functions were studied using a kinesin-gliding assay that mimics intracellular transport processes in vitro by quantifying the movement of paclitaxel-stabilized microtubules across a kinesin-coated glass surface. There was a dose-dependent effect of lead nitrate on microtubule motility. Lead nitrate affected the gliding velocities of microtubules starting at concentrations above 10 μ M and reached half-maximal inhibition of motility at about 50 μ M. The processes reported here point to relevant interactions of lead with tubulin and kinesin at low dose levels. Environ. Mal. Mutagen. 45:346-353, 2005. © 2005 Wiley-Liss, Inc.

Disulfide bond mutagenesis and the structure and function of the head-to-tail macrocyclic trypsin inhibitor SFTI-1

SFTI-1 is a novel 14 amino acid peptide comprised of a circular backbone constrained by three proline residues, a hydrogen-bond network, and a single disulfide bond. It is the smallest and most potent known Bowman-Birk trypsin inhibitor and the only one with a cyclic peptidic backbone. The solution structure of [ABA(3,11)]SFTI-1, a disulfide-deficient analogue of SFTI-1, has been determined by H-1 NMR spectroscopy. The lowest energy structures of native SFTI-1 and [ABA(3,11)]SFTI-1 are similar and superimpose with a root-mean-square deviation over the backbone and heavy atoms of 0.26 +/- 0.09 and 1.10 +/- 0.22 Angstrom, respectively. The disulfide bridge in SFTI-1 was found to be a minor determinant for the overall structure, but its removal resulted in a slightly weakened hydrogen-bonding network. To further investigate the role of the disulfide bridge, NMR chemical shifts for the backbone H-alpha protons of two disulfide-deficient linear analogues of SFTI-1, [ABA(3,11)]SFTI-1[6,5] and [ABA(3,11)]SFTI-1[1,14] were measured. These correspond to analogues of the cleavage product of SFTI-1 and a putative biosynthetic precursor, respectively. In contrast with the cyclic peptide, it was found that the disulfide bridge is essential for maintaining the structure of these open-chain analogues. Overall, the hydrogen-bond network appears to be a crucial determinant of the structure of SFTI-1 analogues.

The molecular basis of temperature compensation in the Arabidopsis circadian clock

Circadian clocks maintain robust and accurate timing over a broad range of physiological temperatures, a characteristic termed temperature compensation. In Arabidopsis thaliana, ambient temperature affects the rhythmic accumulation of transcripts encoding the clock components TIMING OF CAB EXPRESSION1 (TOC1), GIGANTEA (GI), and the partially redundant genes CIRCADIAN CLOCK ASSOCIATED1 (CCA1) and LATE ELONGATED HYPOCOTYL (LHY). The amplitude and peak levels increase for TOC1 and GI RNA rhythms as the temperature increases (from 17 to 27 degrees C), whereas they decrease for LHY. However, as temperatures decrease ( from 17 to 12 degrees C), CCA1 and LHY RNA rhythms increase in amplitude and peak expression level. At 27 degrees C, a dynamic balance between GI and LHY allows temperature compensation in wild-type plants, but circadian function is impaired in Ihy and gi mutant plants. However, at 12 degrees C, CCA1 has more effect on the buffering mechanism than LHY, as the cca1 and gi mutations impair circadian rhythms more than Ihy at the lower temperature. At 17 degrees C, GI is apparently dispensable for free-running circadian rhythms, although partial GI function can affect circadian period. Numerical simulations using the interlocking-loop model show that balancing LHY/CCA1 function against GI and other evening-expressed genes can largely account for temperature compensation in wild-type plants and the temperature-specific phenotypes of gi mutants.

Mixture density network training by computation in parameter space

Training Mixture Density Network (MDN) configurations within the NETLAB framework takes time due to the nature of the computation of the error function and the gradient of the error function. By optimising the computation of these functions, so that gradient information is computed in parameter space, training time is decreased by at least a factor of sixty for the example given. Decreased training time increases the spectrum of problems to which MDNs can be practically applied making the MDN framework an attractive method to the applied problem solver.

Structured neural network modelling of multi-valued functions for wind retrieval from scatterometer measurements

A conventional neural network approach to regression problems approximates the conditional mean of the output vector. For mappings which are multi-valued this approach breaks down, since the average of two solutions is not necessarily a valid solution. In this article mixture density networks, a principled method to model conditional probability density functions, are applied to retrieving Cartesian wind vector components from satellite scatterometer data. A hybrid mixture density network is implemented to incorporate prior knowledge of the predominantly bimodal function branches. An advantage of a fully probabilistic model is that more sophisticated and principled methods can be used to resolve ambiguities.

Neural network-based wind vector retrieval from satellite scatterometer data

Obtaining wind vectors over the ocean is important for weather forecasting and ocean modelling. Several satellite systems used operationally by meteorological agencies utilise scatterometers to infer wind vectors over the oceans. In this paper we present the results of using novel neural network based techniques to estimate wind vectors from such data. The problem is partitioned into estimating wind speed and wind direction. Wind speed is modelled using a multi-layer perceptron (MLP) and a sum of squares error function. Wind direction is a periodic variable and a multi-valued function for a given set of inputs; a conventional MLP fails at this task, and so we model the full periodic probability density of direction conditioned on the satellite derived inputs using a Mixture Density Network (MDN) with periodic kernel functions. A committee of the resulting MDNs is shown to improve the results.

Structured neural network modelling of multi-valued functions for wind retrieval from scatterometer measurements

A conventional neural network approach to regression problems approximates the conditional mean of the output vector. For mappings which are multi-valued this approach breaks down, since the average of two solutions is not necessarily a valid solution. In this article mixture density networks, a principled method to model conditional probability density functions, are applied to retrieving Cartesian wind vector components from satellite scatterometer data. A hybrid mixture density network is implemented to incorporate prior knowledge of the predominantly bimodal function branches. An advantage of a fully probabilistic model is that more sophisticated and principled methods can be used to resolve ambiguities.

Neural network-based wind vector retrieval from satellite scatterometer data

Obtaining wind vectors over the ocean is important for weather forecasting and ocean modelling. Several satellite systems used operationally by meteorological agencies utilise scatterometers to infer wind vectors over the oceans. In this paper we present the results of using novel neural network based techniques to estimate wind vectors from such data. The problem is partitioned into estimating wind speed and wind direction. Wind speed is modelled using a multi-layer perceptron (MLP) and a sum of squares error function. Wind direction is a periodic variable and a multi-valued function for a given set of inputs; a conventional MLP fails at this task, and so we model the full periodic probability density of direction conditioned on the satellite derived inputs using a Mixture Density Network (MDN) with periodic kernel functions. A committee of the resulting MDNs is shown to improve the results.

The missing ingredient:inter-municipal cooperation and central-local relations in the UK

The thrust of the argument presented in this chapter is that inter-municipal cooperation (IMC) in the United Kingdom reflects local government's constitutional position and its exposure to the exigencies of Westminster (elected central government) and Whitehall (centre of the professional civil service that services central government). For the most part councils are without general powers of competence and are restricted in what they can do by Parliament. This suggests that the capacity for locally driven IMC is restricted and operates principally within a framework constructed by central government's policy objectives and legislation and the political expediencies of the governing political party. In practice, however, recent examples of IMC demonstrate that the practices are more complex than this initial analysis suggests. Central government may exert top-down pressures and impose hierarchical directives, but there are important countervailing forces. Constitutional changes in Scotland and Wales have shifted the locus of central- local relations away from Westminster and Whitehall. In England, the seeding of English government regional offices in 1994 has evolved into an important structural arrangement that encourages councils to work together. Within the local government community there is now widespread acknowledgement that to achieve the ambitious targets set by central government, councils are, by necessity, bound to cooperate and work with other agencies. In recent years, the fragmentation of public service delivery has affected the scope of IMC. Elected local government in the UK is now only one piece of a complex jigsaw of agencies that provides services to the public; whether it is with non-elected bodies, such as health authorities, public protection authorities (police and fire), voluntary nonprofit organisations or for-profit bodies, councils are expected to cooperate widely with agencies in their localities. Indeed, for projects such as regeneration and community renewal, councils may act as the coordinating agency but the success of such projects is measured by collaboration and partnership working (Davies 2002). To place these developments in context, IMC is an example of how, in spite of the fragmentation of traditional forms of government, councils work with other public service agencies and other councils through the medium of interagency partnerships, collaboration between organisations and a mixed economy of service providers. Such an analysis suggests that, following changes to the system of local government, contemporary forms of IMC are less dependent on vertical arrangements (top-down direction from central government) as they are replaced by horizontal modes (expansion of networks and partnership arrangements). Evidence suggests, however that central government continues to steer local authorities through the agency of inspectorates and regulatory bodies, and through policy initiatives, such as local strategic partnerships and local area agreements (Kelly 2006), thus questioning whether, in the case of UK local government, the shift from hierarchy to network and market solutions is less differentiated and transformation less complete than some literature suggests. Vertical or horizontal pressures may promote IMC, yet similar drivers may deter collaboration between local authorities. An example of negative vertical pressure was central government's change of the systems of local taxation during the 1980s. The new taxation regime replaced a tax on property with a tax on individual residency. Although the community charge lasted only a few years, it was a highpoint of the then Conservative government policy that encouraged councils to compete with each other on the basis of the level of local taxation. In practice, however, the complexity of local government funding in the UK rendered worthless any meaningful ambition of councils competing with each other, especially as central government granting to local authorities is predicated (however imperfectly) on at least notional equalisation between those areas with lower tax yields and the more prosperous locations. Horizontal pressures comprise factors such as planning decisions. Over the last quarter century, councils have competed on the granting of permission to out-of-town retail and leisure complexes, now recognised as detrimental to neighbouring authorities because economic forces prevail and local, independent shops are unable to compete with multiple companies. These examples illustrate tensions at the core of the UK polity of whether IMC is feasible when competition between local authorities heightened by local differences reduces opportunities for collaboration. An alternative perspective on IMC is to explore whether specific purposes or functions promote or restrict it. Whether in the principle areas of local government responsibilities relating to social welfare, development and maintenance of the local infrastructure or environmental matters, there are examples of IMC. But opportunities have diminished considerably as councils lost responsibility for services provision as a result of privatisation and transfer of powers to new government agencies or to central government. Over the last twenty years councils have lost their role in the provision of further-or higher-education, public transport and water/sewage. Councils have commissioning power but only a limited presence in providing housing needs, social care and waste management. In other words, as a result of central government policy, there are, in practice, currently far fewer opportunities for councils to cooperate. Since 1997, the New Labour government has promoted IMC through vertical drivers and the development; the operation of these policy initiatives is discussed following the framework of the editors. Current examples of IMC are notable for being driven by higher tiers of government, working with subordinate authorities in principal-agent relations. Collaboration between local authorities and intra-interand cross-sectoral partnerships are initiated by central government. In other words, IMC is shaped by hierarchical drivers from higher levels of government but, in practice, is locally varied and determined less by formula than by necessity and function. © 2007 Springer.

Robust automatic mapping algorithms in a network monitoring scenario

Automatically generating maps of a measured variable of interest can be problematic. In this work we focus on the monitoring network context where observations are collected and reported by a network of sensors, and are then transformed into interpolated maps for use in decision making. Using traditional geostatistical methods, estimating the covariance structure of data collected in an emergency situation can be difficult. Variogram determination, whether by method-of-moment estimators or by maximum likelihood, is very sensitive to extreme values. Even when a monitoring network is in a routine mode of operation, sensors can sporadically malfunction and report extreme values. If this extreme data destabilises the model, causing the covariance structure of the observed data to be incorrectly estimated, the generated maps will be of little value, and the uncertainty estimates in particular will be misleading. Marchant and Lark [2007] propose a REML estimator for the covariance, which is shown to work on small data sets with a manual selection of the damping parameter in the robust likelihood. We show how this can be extended to allow treatment of large data sets together with an automated approach to all parameter estimation. The projected process kriging framework of Ingram et al. [2007] is extended to allow the use of robust likelihood functions, including the two component Gaussian and the Huber function. We show how our algorithm is further refined to reduce the computational complexity while at the same time minimising any loss of information. To show the benefits of this method, we use data collected from radiation monitoring networks across Europe. We compare our results to those obtained from traditional kriging methodologies and include comparisons with Box-Cox transformations of the data. We discuss the issue of whether to treat or ignore extreme values, making the distinction between the robust methods which ignore outliers and transformation methods which treat them as part of the (transformed) process. Using a case study, based on an extreme radiological events over a large area, we show how radiation data collected from monitoring networks can be analysed automatically and then used to generate reliable maps to inform decision making. We show the limitations of the methods and discuss potential extensions to remedy these.