Disaggregation & aggregation of time series components: A hybrid forecasting approach using generalized regression neural networks and the theta method

A hybrid forecasting method is proposed which leverages from statistical and neural network techniques to perform multi-step ahead forecasting. The proposed method is based on the disaggregation of time series components, the prediction of each component individually and the reassembling of the extrapolations to obtain an estimation for the global data. The STL decomposition procedure from the literature [5] is implemented to obtain the seasonal, trend and irregular components of the time series, whilst Generalized Regression Neural Networks (GRNN) [12] are used to perform out-of sample extrapolations of the seasonal and residual components. The univariate Theta model is employed for the estimation of the directional component. The application of the GRNN is based on the dynamic calibration of the training process for each of the seasonal and irregular components individually. The proposed hybrid forecasting method is applied to 60 time series from the NN3 competition and 227 time series from the M1 Competition dataset, to obtain 18 out-of sample predictions. The results from the application demonstrate that the proposed method can outperform standard statistical techniques in the literature. One of the main contributions of the current research lies in the investigation of the strengths and weaknesses of the GRNN in extrapolating structural components of time series.     

The dual specificity phosphatases M3/6 and MKP-3 are highly selective for inactivation of distinct mitogen-activated protein kinases

The mitogen-activated protein (MAP) kinase family includes extracellular signal-regulated kinase (ERK), c-Jun NH2-terminal kinase/stress-activated protein kinase (JNK/SAPK) and p38/RK/CSBP (p38) as structurally and functionally distinct enzyme classes. Here we describe two new dual specificity phosphatases of the CL100/MKP-1 family that are selective for inactivating ERK or JNK/SAPK and p38 MAP kinases when expressed in COS-7 cells. M3/6 is the first phosphatase of this family to display highly specific inactivation of JNK/SAPK and p38 MAP kinases. Although stress-induced activation of p54 SAPKbeta, p46 SAPKgamma (JNK1) or p38 MAP kinases is abolished upon co-transfection with increasing amounts of M3/6 plasmid, epidermal growth factor-stimulated ERK1 is remarkably insensitive even to the highest levels of M3/6 expression obtained. In contrast to M3/6, the dual specificity phosphatase MKP-3 is selective for inactivation of ERK family MAP kinases. Low level expression of MKP-3 blocks totally epidermal growth factor-stimulated ERK1, whereas stress-induced activation of p54 SAPKbeta and p38 MAP kinases is inhibited only partially under identical conditions. Selective regulation by M3/6 and MKP-3 was also observed upon chronic MAP kinase activation by constitutive p21(ras) GTPases. Hence, although M3/6 expression effectively blocked p54 SAPKbeta activation by p21(rac) (G12V), ERK1 activated by p21(ras) (G12V) was insensitive to this phosphatase. ERK1 activation by oncogenic p21(ras) was, however, blocked totally by co-expression of MKP-3. This is the first report demonstrating reciprocally selective inhibition of different MAP kinases by two distinct dual specificity phosphatases.     

Cherenkov and Scintillation Light Measurements with Scintillating Glass, SCG1C

We have been able to observe and measure both the direct Cherenkov (C) and the Scintillation (S) light components from scintillating glass, distinctly separated in time. This has important implications for hadron calorimetry, electron/hadron separation and low energy particle identification.     

Gelatin as a promising printable feedstock for microbial fuel cells (MFC)

The microbial fuel cell (MFC) is an energy transducer that can directly produce electricity from bacterial oxidation of organic matter. MFCs consist of two reaction chambers (anode and cathode) separated by a semipermeable membrane. This study describes the work carried out towards the optimization of critical MFC components, with 3D fabricated materials. The response of the optimized fuel cells, which were fed with soft materials such as gelatin, alginate and Nafion?, is also reported. The optimised components were the membrane and the cathode electrode. A conventional Nafion membrane was substituted with a custom made terracotta sheet and the electrode used was a single sheet of carbon veil coated with an activated carbon paste. The results showed that among the soft materials tested within the anodic chamber, gelatin performed the best; it also revealed that even after a 10-day starvation period gelatin demonstrated better longevity. These results show that MFCs have the potential to be 3D-printed monolithically using the EVOBOT platform.     

Information models of layout constraints for product life-cycle management: a solid-modelling approach

Complex products and systems, like an aircraft, ship, or machinery plant, involve a large number of components which are arranged under spatial constraints/relationships in design space. Current practices approximate these relationships with simplified ‘dimensional constraints’ aiming at formulating the design problem as a system of (in)equalities to be solved automatically, e.g. by a geometric-constraint solver. This research proposes informationally-complete models for design-constraints based on an analysis of geometric and non-geometric properties of the related space volumes. Also, an Extended Product Model is proposed describing the system's structure and components as well as related procedures and constraints to be used as a system life-cycle model.     

The mitogen-activated protein kinase phosphatase-3 N-terminal noncatalytic region is responsible for tight substrate binding and enzymatic specificity

We have reported recently that the dual specificity mitogen-activated protein kinase phosphatase-3 (MKP-3) elicits highly selective inactivation of the extracellular signal-regulated kinase (ERK) class of mitogen-activated protein (MAP) kinases (Muda, M., Theodosiou, A., Rodrigues, N., Boschert, U., Camps, M., Gillieron, C., Davies, K., Ashworth, A., and Arkinstall, S. (1996) J. Biol. Chem. 271, 27205-27208). We now show that MKP-3 enzymatic specificity is paralleled by tight binding to both ERK1 and ERK2 while, in contrast, little or no interaction with either c-Jun N-terminal kinase/stress activated protein kinase (JNK/SAPK) or p38 MAP kinases was detected. Further study revealed that the N-terminal noncatalytic domain of MKP-3 (MKP-3DeltaC) binds both ERK1 and ERK2, while the C-terminal MKP-3 catalytic core (MKP-3DeltaN) fails to precipitate either of these MAP kinases. A chimera consisting of the N-terminal half of MKP-3 with the C-terminal catalytic core of M3-6 also bound tightly to ERK1 but not to JNK3/SAPKbeta. Consistent with a role for N-terminal binding in determining MKP-3 specificity, at least 10-fold higher concentrations of purified MKP-3DeltaN than full-length MKP-3 is required to inhibit ERK2 activity. In contrast, both MKP-3DeltaN and full-length MKP-3 inactivate JNK/SAPK and p38 MAP kinases at similarly high concentrations. Also, a chimera of the M3-6 N terminus with the MKP-3 catalytic core which fails to bind ERK elicits non selective inactivation of ERK1 and JNK3/SAPKbeta. Together, these observations suggest that the physiological specificity of MKP-3 for inactivation of ERK family MAP kinases reflects tight substrate binding by its N-terminal domain.