Deep Learning Based Prediction of Gas Chromatographic Retention Indices for a Wide Variety of Polar and Mid-Polar Liquid Stationary Phases

Prediction of gas chromatographic retention indices based on compound structure is an important task for analytical chemistry. The predicted retention indices can be used as a reference in a mass spectrometry library search despite the fact that their accuracy is worse in comparison with the experimental reference ones. In the last few years, deep learning was applied for this task. The use of deep learning drastically improved the accuracy of retention index prediction for non-polar stationary phases. In this work, we demonstrate for the first time the use of deep learning for retention index prediction on polar (e.g., polyethylene glycol, DB-WAX) and mid-polar (e.g., DB-624, DB-210, DB-1701, OV-17) stationary phases. The achieved accuracy lies in the range of 16–50 in terms of the mean absolute error for several stationary phases and test data sets. We also demonstrate that our approach can be directly applied to the prediction of the second dimension retention times (GC × GC) if a large enough data set is available. The achieved accuracy is considerably better compared with the previous results obtained using linear quantitative structure-retention relationships and ACD ChromGenius software. The source code and pre-trained models are available online.     

Breathing architecture: Conceptual architectural design based on the investigation into the natural ventilation of buildings

This study explores architectural design by examining air, fluid mechanics, and the natural ventilation of buildings. In this context, this research introduces a new way of dealing with the process of architectural synthesis. The proposed way can be used either to create new architectural projects or to rethink existing ones. This study is supported by previous investigation into the natural ventilation of buildings via computational and laboratory simulation (Stavridou, 2011; Stavridou and Prinos, 2013). The investigation into the natural ventilation of buildings provides information and data that affect architectural design through various parameters. The parameters of architectural synthesis that are influenced and discussed in this paper are the following: (i) inspiration and analogical transfer, (ii) initial conception of the main idea using computational fluid dynamics (digital design), (iii) development of the main idea through an investigatory process toward building form optimization, and (iv) form configuration, shape investigation, and other morphogenetic prospects. This study illustrates the effect of natural ventilation research on architectural design and thus produces a new approach to the architectural design process. This approach leads to an innovative kind of architecture called “breathing architecture.”     

Fully coupled heat conduction and deformation analyses of visco-elastic solids

Visco-elastic materials are known for their capability of dissipating energy. This energy is converted into heat and thus changes the temperature of the materials. In addition to the dissipation effect, an external thermal stimulus can also alter the temperature in a visco-elastic body. The rate of stress relaxation (or the rate of creep) and the mechanical and physical properties of visco-elastic materials, such as polymers, vary with temperature. This study aims at understanding the effect of coupling between the thermal and mechanical response that is attributed to the dissipation of energy, heat conduction, and temperature-dependent material parameters on the overall response of visco-elastic solids. The non-linearly visco-elastic constitutive model proposed by Schapery (Further development of a thermodynamic constitutive theory: stress formulation, 1969, Mech. Time-Depend. Mater. 1:209?C240, 1997) is used and modified to incorporate temperature- and stress-dependent material properties. This study also formulates a non-linear energy equation along with a dissipation function based on the Gibbs potential of Schapery (Mech. Time-Depend. Mater. 1:209?C240, 1997). A numerical algorithm is formulated for analyzing a fully coupled thermo-visco-elastic response and implemented it in a general finite-element (FE) code. The non-linear stress- and temperature-dependent material parameters are found to have significant effects on the coupled thermo-visco-elastic response of polymers considered in this study. In order to obtain a realistic temperature field within the polymer visco-elastic bodies undergoing a non-uniform heat generation, the role of heat conduction cannot be ignored.     

Laser-engineered topography: correlation between structure dimensions and cell control

Topographical cues have a significant impact on cell responses and by this means, on the fabrication of innovative implant materials. However, analysis of cell-topography interactions in dependence of the surface feature dimensions is still challenging due to limitations in the fabrication technology. Here, we introduce surface structuring via picosecond laser systems, which enable a fast production of micro-sized topologies. Changes in the processing parameters further control the feature sizes of so-called spikes. Using surfaces with big and small spike-to-spike-distances for comparisons, we focussed on cell adhesion via extracellular matrix adsorption and focal adhesion complexes, morphology, localisation and proliferation of fibroblasts. The observed cell control was dependent on a turnover point related to the structure dimensions: only big spike-to-spike-distances reduced cell behaviour. Therefore, this technology offers a platform to study cell and tissue interactions with a defined microenvironment.     

Common failures of demand response

Demand response (DR) programs have recently become popular means of mitigating excessive claims on regional electricity networks and providing more reliable access to electric power. Many countries have experimented with DR pilot programs and some are beginning to incorporate similar schemes as permanent elements of the electricity sector. However, DR remains an experimental technique in much of the world as programs frequently fail to meet their goals and reach their potential. This paper examines the central structural and behavioral obstacles to success of DR programs and outlines some potential solutions which could greatly improve the functionality and success of such programs in the future.     

Working-memory-triggered dynamic adjustments in cognitive control.

Dynamic adjustments in cognitive control are well documented in conflict tasks, wherein competition from irrelevant stimulus attributes intensifies selection demands and leads to subsequent performance benefits. The current study investigated whether mnemonic demands, in a working memory (WM) task, can drive similar online control modifications. Demand levels (high vs. low) of WM maintenance (memory load of 2 items vs. 1 item) and delay-spanning distractor interference (confusable vs. not confusable with memoranda) were manipulated using a factorial design during a WM delayed-recognition task. Performance was best subsequent to trials in which both maintenance and distractor interference demands were high, followed by trials with high demand in either of these 2 control domains, and worst following trials with low demand in both domains. These results suggest that dynamic adjustments in cognitive control are not triggered exclusively by conflict-specific contexts but are also triggered by WM demands, revealing a putative mechanism by which this system configures itself for successful task performance. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Estimation of environmental effects of photovoltaic generation in North-west China

In estimating emissions reductions brought about by renewables in China, much of existing research assumes that renewables displace coal power. In this paper, this assumption is challenged and the potential environmental effects of photovoltaic (PV) power in North-west China are reevaluated when the marginal generator actually being displaced is taken into account. The annual PV power generation in the North-west Grid is estimated, in this paper, to be as high as 17900 GW·h in 2015, roughly equaling to the output of 1.5 nuclear power plants in the US today. The total associated emission reduction in 2015 will at most be 0.36 percent of SO₂ and 0.25 percent of NO_x emissions from their 2010 levels in China. Further, PV power may render no emissions reduction at all if it displaces hydropower, which is often used to meet peak demand in the North-west Grid in China. These results imply that a more cost-effective area of focus in the short-term may be on desulfurization and denitrification technologies for coal plants.