Validation of improved TAMANN neural network for operational satellite-derived rainfall estimation in Africa

Real-time rainfall monitoring in Africa is of great practical importance for operational applications in hydrology and agriculture. Satellite data have been used in this context for many years because of the lack of surface observations. This paper describes an improved artificial neural network algorithm for operational applications. The algorithm combines numerical weather model information with the satellite data. Using this algorithm, daily rainfall estimates were derived for 4 yr of the Ethiopian and Zambian main rainy seasons and were compared with two other algorithms-a multiple linear regression making use of the same information as that of the neural network and a satellite-only method. All algorithms were validated against rain gauge data. Overall, the neural network performs best, but the extent to which it does so depends on the calibration/validation protocol. The advantages of the neural network are most evident when calibration data are numerous and close in space and time to the validation data. This result emphasizes the importance of a real-time calibration system.

A hierarchical distributed approach for mining molecular fragments

Recently, two approaches have been introduced that distribute the molecular fragment mining problem. The first approach applies a master/worker topology, the second approach, a completely distributed peer-to-peer system, solves the scalability problem due to the bottleneck at the master node. However, in many real world scenarios the participating computing nodes cannot communicate directly due to administrative policies such as security restrictions. Thus, potential computing power is not accessible to accelerate the mining run. To solve this shortcoming, this work introduces a hierarchical topology of computing resources, which distributes the management over several levels and adapts to the natural structure of those multi-domain architectures. The most important aspect is the load balancing scheme, which has been designed and optimized for the hierarchical structure. The approach allows dynamic aggregation of heterogenous computing resources and is applied to wide area network scenarios.

Distributed mining of molecular fragments

In real world applications sequential algorithms of data mining and data exploration are often unsuitable for datasets with enormous size, high-dimensionality and complex data structure. Grid computing promises unprecedented opportunities for unlimited computing and storage resources. In this context there is the necessity to develop high performance distributed data mining algorithms. However, the computational complexity of the problem and the large amount of data to be explored often make the design of large scale applications particularly challenging. In this paper we present the first distributed formulation of a frequent subgraph mining algorithm for discriminative fragments of molecular compounds. Two distributed approaches have been developed and compared on the well known National Cancer Institute’s HIV-screening dataset. We present experimental results on a small-scale computing environment.

Efficient mining of discriminative molecular fragments

Frequent pattern discovery in structured data is receiving an increasing attention in many application areas of sciences. However, the computational complexity and the large amount of data to be explored often make the sequential algorithms unsuitable. In this context high performance distributed computing becomes a very interesting and promising approach. In this paper we present a parallel formulation of the frequent subgraph mining problem to discover interesting patterns in molecular compounds. The application is characterized by a highly irregular tree-structured computation. No estimation is available for task workloads, which show a power-law distribution in a wide range. The proposed approach allows dynamic resource aggregation and provides fault and latency tolerance. These features make the distributed application suitable for multi-domain heterogeneous environments, such as computational Grids. The distributed application has been evaluated on the well known National Cancer Institute’s HIV-screening dataset.

A potential energy surface for the ground state of formaldehyde, H2CO(1A1)

A model potential energy function for the ground state of H2CO has been derived which covers the whole space of the six internal coordinates. This potential reproduces the experimental energy, geometry and quadratic force field of formaldehyde, and dissociates correctly to all possible atom, diatom and triatom fragments. Thus there are good reasons for believing it to be close to the true potential energy surface except in regions where both hydrogen atoms are close to the oxygen. It leads to the prediction that there should be a metastable singlet hydroxycarbene HCOH which has a planar trans structure and an energy of 2•31 eV above that of equilibrium formaldehyde. The reaction path for dissociation into H2 + CO is predicted to pass through a low symmetry transition state with an activation energy of 4•8 eV. Both of these predictions are in good agreement with recently published ab initio calculations.

Analytical potentials for triatomic molecules VIII. A two valued surface for the lowest 1A¢ surfaces of H2O

Ab initio calculations of the energy have been made at approximately 150 points on the two lowest singlet A' potential energy surfaces of the water molecule, 1A' and 1A', covering structures having D∞h, C∞v, C2v and Cs symmetries. The object was to obtain an ab initio surface of uniform accuracy over the whole three-dimensional coordinate space. Molecular orbitals were constructed from a double zeta plus Rydberg basis, and correlation was introduced by single and double excitations from multiconfiguration states which gave the correct dissociation behaviour. A two-valued analytical potential function has been constructed to fit these ab initio energy calculations. The adiabatic energies are given in our analytical function as the eigenvalues of a 2 2 matrix, whose diagonal elements define two diabatic surfaces. The off-diagonal element goes to zero for those configurations corresponding to surface intersections, so that our adiabatic surface exhibits the correct Σ/II conical intersections for linear configurations, and singlet/triplet intersections of the O + H2 dissociation fragments. The agreement between our analytical surface and experiment has been improved by using empirical diatomic potential curves in place of those derived from ab initio calculations.

Effects of corn silage derived from a genetically modified variety containing two Transgenes on feed intake, milk production, and composition, and the absence of detectable Transgenic deoxyribonucleic acid in milk in holstein dairy cows

The objectives were to compare the chemical composition, nutritive value, feed intake, milk production and composition, and presence in milk of transgenic DNA and the encoded protein Cry1Ab when corn silages containing 2 transgenes (2GM: herbicide tolerance: mepsps and insect resistance: cry1Ab) were fed as part of a standard total mixed ration (TMR) compared with a near isogenic corn silage ( C) to 8 multiparous lactating Holstein dairy cows in a single reversal design study. Cows were fed a TMR ration ad libitum and milked twice daily. Diets contained [ dry matter (DM) basis] 45% corn silage, 10% alfalfa hay, and 45% concentrate (1.66 Mcal of net energy for lactation/kg of DM, 15.8% crude protein, 35% neutral detergent fiber, and 4.1% fat). Each period was 28-d long. During the last 4 d of each period, feed intake and milk production data were recorded and milk samples taken for compositional analysis, including the presence of transgenic DNA and Cry1Ab protein. There was no significant difference in the chemical composition between C and 2GM silages, and both were within the expected range (37.6% DM, 1.51 Mcal of net energy for lactation/kg, 8.6% crude protein, 40% neutral detergent fiber, 19.6% acid detergent fiber, pH 3.76, and 62% in vitro DM digestibility). Cows fed the 2GM silage produced milk with slightly higher protein (3.09 vs. 3.00%), lactose ( 4.83 vs. 4.72%) and solids-not-fat (8.60 vs. 8.40%) compared with C. However, the yield (kg/d) of milk (36.5), 3.5% fat-corrected milk (34.4), fat (1.151), protein (1.106), lactose (1.738), and solids-not-fat ( 3.094), somatic cell count (log(10): 2.11), change in body weight (+ 7.8 kg), and condition score (+ 0.09) were not affected by type of silage, indicating no overall production difference. All milk samples were negative for the presence of transgenic DNA from either trait or the Cry1Ab protein. Results indicate that the 2GM silage modified with 2 transgenes did not affect nutrient composition of the silages and had no effect on animal performance and milk composition. No transgenic DNA and Cry1Ab protein were detected in milk.

In vitro evaluation of fibrolytic enzymes as additives for maize (Zea mays L.) silage - III. Comparison of enzymes derived from psychrophilic, mesophilic or thermophilic sources

A completely randomised study was completed to examine the influence of fibrolytic enzymes derived from psychrophilic, (F), mesophilic, (L) or thermophilic (Ta) sources, applied at ensiling, on the chemical characteristics and in vitro rumen fermentation of maize silage, assessed using the Reading Pressure Technique (RPT). Treatments, all in triplicate, consisted of untreated maize forage or treated with preparations F, L, Ta or a mixture (1: 1, v/v) of F and L (FL), at two levels each, and ensiled for 210 days in plastic mini-silos. Addition of enzymes L decreased (P < 0.05) silage pH relative to the control, whereas enzyme Ta tended (P < 0.10) to reduce it. Preparations F, L and Ta tended to reduce (P < 0.10) the fibre contents of the silages, with effects being attributable to a decrease in the cellulose fraction. Starch contents were reduced (P < 0.05) in the treatments including enzyme F. End-point (96 h) gas production (GP) values did not differ among treatments, suggesting that enzymes did not change the total amount of fermentable substrate. However, consistent with the decrease in starch contents, adding enzyme F reduced (P < 0.05) GP at most incubation times. Addition of enzymes increased (P < 0.05) the initial (6 h) organic matter degradation (OMD) levels in all but one treatment (F), with increases of 14, 19, and 26% for preparations L, Ta, and FL, respectively, averaged across levels. Furthermore, the addition of enzymes increased (P < 0.05) the soluble OM losses, however, these increases did not fully account for the initial increase in OMD. The latter suggests that enzymes increased solubility and also altered silage structure, making it more amenable to degradation by ruminal microorganisms. As a result of the increase in OMD, without a concomitant increase in GP, the fermentation efficiency was greatly increased (P < 0.05) in enzyme treatments. Addition of enzymes to maize at ensiling, particularly those from the mesophilic and thermophilic sources used here, have the potential to increase the initial rate of silage OMD. (C) 2003 Elsevier B.V. All rights reserved.

The role of animal nutrition in improving the nutritive value of animal-derived foods in relation to chronic disease

Foods derived from animals are an important source of nutrients in the diet; for example, milk and meat together provide about 60 and 55% of the dietary intake of Ca and protein respectively in the UK. However, certain aspects of some animal-derived foods, particularly their fat and saturated fatty acid (SFA) contents, have led to concerns that these foods substantially contribute to the risk of CVD, the metabolic syndrome and other chronic diseases. In most parts of Europe dairy products are the greatest single dietary source of SFA. The fatty acid composition of various animal-derived foods is, however, not constant and can, in many cases, be enhanced by animal nutrition. In particular, milk fat with reduced concentrations of the C12-16 SFA and an increased concentration of 18:1 MUFA is achievable, although enrichment with very-long-chain n-3 PUFA is much less efficient. However, there is now evidence that some animal-derived foods (notably milk products) contain compounds that may actively promote long-term health, and research is urgently required to fully characterise the benefits associated with the consumption of these compounds and to understand how the levels in natural foods can be enhanced. It is also vital that the beneficial effects are not inadvertently destroyed in the process of reducing the concentrations of SFA. In the future the role of animal nutrition in creating foods closer to the optimum composition for long-term human health is likely to become increasingly important, but production of such foods on a scale that will substantially affect national diets will require political and financial incentives and great changes in the animal production industry.

Very long chain n-3 polyunsaturated fatty acids in the food chain in the UK and the potential of animal-derived foods to increase intake

The very long chain (VLC) n-3 polyunsaturated fatty acids (PUFA), particularly eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), are widely recognised to have beneficial effects on human health. However, recommended intakes of VLC n-3 PUFA (450 mg/day) are not being met by the diet in the majority of the population mainly because of low consumption of oil-rich fish. Current mean intake of VLC n-3 PUFA by adults is estimated to be about 282 mg/day with EPA and DHA contributing about 244 mg/day. Furthermore, the fact that only about 27% of adults eat any oil-rich fish (excluding canned tuna) and knowledge of the poor conversion of α-linolenic acid to EPA and DHA in vivo, particularly in men, leads to the need to review current dietary sources of these fatty acids. Animal-derived foods are likely to have an important function in increasing intake and studies have shown that feeding fish oils to animals can increase the EPA and DHA content of the resulting food products. This paper highlights the importance of examining current and projected consumption trends of meat and other animal products when exploring the potential impact of enriched foods by means of altering animal diets. When related to current food consumption data, potential dietary intakes of EPA+DHA from foods derived from animals fed enriched diets are calculated to be about 231 mg/day. If widely consumed, such foods could have a significant impact on progression of conditions such as cardiovascular disease. Consideration is also given to the sources of VLC n-3 PUFA in animal diets, with the sustainability of fish oil being questioned and the need to investigate the use of alternative dietary sources such as those of algal origin.

Current intakes of EPA and DHA in European populations and the potential of animal-derived foods to increase them

The beneficial effects of long-chain (C chain >= 20) n-3 PUFA are well documented and, overall, increased intake reduces risk of CVD. Recent evidence also points to a role in reducing, age-related decline in cognitive function. The two key fatty acids are EPA (20:5) and DHA (22:6), with current UK recommendation for adults being 450 mg EPA + DHA/d. Whilst some EPA and DHA can be synthesised in vivo from alpha-linolenic acid, recent data indicate this source to be very limited, Suggesting that EPA and DHA should be classified as dietary essentials. In many parts of Europe the daily intake of EPA + DHA by adults and especially young adults (18-24 years) is < 100 mg/d, since many never eat oily fish. Poultry meat contributes small but worthwhile amounts of EPA+DHA. Studies to enrich the EPA+DHA content of animal-derived foods mainly use fish oil in the diet of the animal. Recent work has shown that such enrichment has the potential to provide to the UK adult diet a daily intake of EPA+DHA of about 230 mg, with poultry meat providing the largest amount (74 mg). There are. however. concerns that the Continued and possibly increased use of fish oils in animals diets is not Sustainable and alternative approaches are being examined, including the genetic modification of certain plants to allow them to synthesise EPA and DHA from shorter-chain precursors.