Search

Additional file 1.

AutoDock VINA is one of the most-used docking tools in the early stage of modern drug discovery. It uses a Monte-Carlo based iterated search method and multithreading parallelism scheme on multicore machines to improve docking accuracy and speed. However, virtual screening from huge compound databases is common for modern drug discovery, which puts forward a great demand for higher docking speed of AutoDock VINA. Therefore, we propose a fast method VINA-GPU, which expands the Monte-Carlo based docking lanes into thousands of ones coupling with a largely reduced number of search steps in each lane. Furthermore, we develop a heterogeneous OpenCL implementation of VINA-GPU that leverages thousands of computational cores of a GPU, and obtains a maximum of 403-fold acceleration on docking runtime when compared with a quad-threaded AutoDock VINA implementation. In addition, a heuristic function was fitted to determine the proper size of search steps in each lane for a convenient usage. The VINA-GPU code can be freely available at https://github.com/DeltaGroupNJUPT/VINA-GPU for academic usage.

This paper presents an analysis of the comprehensive risks associated with the typhoons in the China coastal main functional area from 1991 to 2010, a prediction of the changes of these comprehensive risks in the future, and mitigation measures to limit the impact of these disasters on the economy and the population. The analysis is based on the two main criteria of vulnerability and danger. The former includes socioeconomic factors of the regions affected by typhoon, namely the population and GDP. The latter reflects the intensity of the typhoon itself. Our analysis reveals that: (1) The impact of typhoons on coastal provinces is highly variable in space, the province of Hainan, and coastal cities in Guangdong, Fujian and Zhejiang provinces being the most severely affected; (2) From 1991 to 2010, the risk associated with typhoons showed an upward trend in China's coastal cities, with the vulnerability increasing gradually from the inland to the coastal areas; (3) Regions comprising both medium and low risk areas, functional zoning should be adjusted according to the distribution of risk in those regions, and in high risk areas, emphasis should be given on improving the current disaster prevention and mitigation measures, especially with regards to the implementation of better early warning system and post-disaster management. This article analyzes and simulates the typhoon risk changes of the four main functional areas in the coastal of China for the first time, and proposes disaster avoidance measures, which can provide reference for relevant policy makers.

At present, automobiles have become a common means of transportation, but with the increase of vehicles, safety issues have gradually emerged. Therefore, the assembly, manufacture and production of vehicles require systematic testing and rigorous inspection. Therefore, defect detection of vehicle parts is particularly important. Vehicle parts defect detection has evolved from manual detection of traditional classification methods to machine vision methods. In this paper, the deep learning method is used to firstly detect the defects of vehicle parts through the training of VGG16 network structure model. The accuracy rate is 94.36. Secondly, the VGG16 network structure model is improved. By introducing the inceptionv3 module, the width of the model is increased on the basis of depth, the image is better recognized with an accuracy of 95.29. However, the accuracy of the traditional HOG+SVM classification method is only 93.88, and the efficiency of both methods is higher than that of the traditional method.

In human memory, forgetting occur rapidly after the remembering and the rate of forgetting slowed down as time went. This is so-called the Ebbinghaus forgetting curve. There are many explanations of how this curve occur based on the properties of the brains. In this article, we use a simple mathematical model to explain the mechanism of forgetting based on rearrangement inequality and get a general formalism for short-term and long-term memory and use it to fit the Ebbinghaus forgetting curve. We also find out that forgetting is not a flaw, instead it is help to improve the efficiency of remembering when human confront different situations by reducing the interference of information and reducing the number of retrievals. Furthurmove, we find that the interference of informations limits the capacity of human memory, which is the "magic number seven"., Comment: 4 pages, 2 figures

Shancheng Yan1 2 ∗, Zhongdang Xiao3, Yi Shi2, Dong Hu1, Xin Xu1, Jiansheng Wu1, and Minmin Zhou2 1School of Geography and Biological Information, Nanjing University of Posts and Telecommunications, Nanjing 210046, China 2National Laboratory of Solid State Microstructures, Department of Physics, Nanjing University, Nanjing 210093, P. R. China 3State Key Laboratory of Bioelectronics (Chien-Shiung Wu Lab), School of Biological Science and Medical Engineering, Southeast University, Si Pai Lou 2#, Nanjing 210096, P. R. China

Urban ecological security pattern is the focus and emphasis of the landscape ecology research,and identifying the source area is the first link to build the ecological security pattern.This paper summarized the existing researches on the recognition methods of the important patches,overall,the commonly used method is relatively simple,which based on the biodiversity as well as the importance of ecosystem services.However,these methods are biased towards considering the structure and function of the patch itself,while ignoring the importance of patches in the entire matrix structure,as well as the relationship with the surrounding environment.Therefore,this paper attempted to propose a comprehensive recognition method of connectivity analysis,biodiversity service evaluation and habitat quality evaluation to recognize the source area of the ecological security pattern. The research area of this paper is Shenzhen City which as China′s opening up window in the south of Guangdong province.With the unconventional high-speed development,Shenzhen has accumulated a large number of ecological and environmental problems,which have threatened the ecological safety in Shenzhen and has become major restrictive conditions to achieving sustainable urban development.Therefore,the ecological protection is urgent.In this paper,both the functions of the ecological patch itself and the structure importance in the landscape pattern were incorporated when recognizing the important patches.Furthermore,as an integrated approach,the landscape connectivity index,the biodiversity service equivalent and the habitat quality evaluation module of the InVEST model were employed to recognize the important ecological patches.The results are as follows: (1) The total area of the source area of the ecological security pattern is 312.10km2,accounting for 15.9% of the total area of the city.The ecological land was divided into five grades in accordance with the landscape connectivity,biodiversity service and habitat quality,the former three grades constituted the source area,in whichthe most important patches are mainly located in Longgang District,Yantian District and Luohu District.It can be seen from the distribution characteristics of the important patches that they are mainly distributed on the mountains.It is suggested that different intensity of management measures should be taken for different levels of source region patches to maximize the efficiency of protection.(2) Superimpose the recognition results with the present ecological control line to analyse the potential ecological security risk.The results shows that more than 95% of the important patches falls within the ecological control line,in this way,this recognition method is feasible to some extent.The small amount of patches exposed outside the control line is ecological risk areas,which mainly distribute in the Kuichong town and Nan′aotown,Longgang District,where we should strengthen the protection.(3) Contrast to the existing research on the recognition of the source area,the method proposed in this paper has a certain degree of improvement,which considered not only the natural factors but also the human factors.It provided a reference for the theory development of landscape pattern and urban ecological conservation practices.However,this paper has some shortcomings,only the scope of Shenzhen City was considered when assessed the habitat quality,making the evaluation of the patches on the edge inaccurate.In addition,dynamic research can be considered in the future work,through the recognition of the source area for the different years to explore the characteristics and change,thus the recognition will be more credible.At last,the rationality and universality of this method needs more practices to verify and validate.

The demand for coal resources is increasing with economic development in China.But,the rich mineral resources along with the intensive mining activities,which have led to a lot of ecological and environmental problems,such as environmental pollution,ecological destruction and geological disasters.Eco-risk assessment is an important tool for effective ecological environment management with evaluations of the likelihood and degree of harm. Thus,this paper applied landscape eco-risk assessment methods to estimate the level of the opencast mine area eco-risk.But also the ESDA(Exploratory spatial data analysis) methods were used to analyze the spatial association characteristic of landscape eco-risk in mining area.It turned out that those high-risk areas is the core of the spatial distribution of Landscape eco-risk and from high-risk to low-risk present ring features.In addition,high-risk is mainly distributed in the mining industry and city life areas,sources of which are the mining,coal gangue tie up land and city construction activities.Medium-risk are mainly distributed in nature,half natural functional areas,the main sources of which are the city life and the agricultural production activities.Low-risk areas are covered by sparse vegetation,such as shrub and grassland.However,in research areas,vegetation degradation,soil and water loss leads to the regional ecological base quality was poor.So the ecosystems and environmental factors are fragile here.Moreover,the frequent production activities in mining areas cause more serious ecological and environment problems.The risk degree of the artificial area is higher than the half natural area,and the half natural area is higher than the natural area.It is conform to the conventional law of ecological risk level difference.At the same time,based on the ESDA of the landscape eco-risk in opencast mine area,the space correlation of the risk is up to 0.7521,which indicate that the agglomeration features of the risk obviously.Different characteristics of agglomeration provide the basis for the ecological and environment management.Space differentiation study of Landscape eco-risk in this opencast mine area shows that interference are the main driving factors in the high-risk and low-risk areas.But in the medium-risk area,fragile degree is the main driving factors.With the increase of human interference,landscape lost diversity and tend to simplification.The results of the study on the mining area provide certain data support and theoretical basis to environment management.

We have prepared the skutterudite-related compounds FeCo3Sb12 and La0.75Fe3CoSb12 with different average grain sizes (about 0.8 and 3.9μm) by hot pressing. Samples were characterized by XRD, EPMA and SEM. The lattice thermal conductivity was investigated in the temperature range from room temperature to 200°C. Based on the Debye model, we analyse the change in lattice thermal conductivity due to various phonon scattering mechanisms by examining the relationship between the weighted phonon relaxation time τ(ω/ωD)2 and the reduced phonon frequency ω/ωD. The effect of grain boundary scattering to phonon is negligible within the range of grain sizes considered in this study. The large reduction in lattice thermal conductivity of FeCo3Sb12 compound contributes to the electron–phonon scattering. As for La0.75Fe3CoSb12 compound, the atoms of La filled into the large voids in the structure of the skutterudite produce more significant electron–phonon scattering as well as more substitute of Fe at Co site at the same time. Moreover, the point-defect scattering appears due to the difference between the atoms of La and the void. In addition, the scattering by the rattling of the rare-earth atoms in the void is another major contribution to the reduced lattice thermal conductivity. Introducing the coupling of the electron–phonon scattering with the point-defect scattering and the scattering by the rattling of the rare-earth atom is an effective method to reduce the lattice thermal conductivity of the skutterudite-related compounds by substitution of Fe for Co and the atoms of La filled in the large voids in the skutterudite structure.

In traditional motion detection, the problem is that light changing and shadow are sensitive for background subtraction and it is difficult to achieve real-time detection due to high dimension and a large number of calculating when suing center-symmetric local binary patterns (CSLBP). A reasonable method of combining both method is proposed and used for motion detection. Using improved background subtraction based on center-symmetric local binary patterns, it solved the time efficiency problem by eliminating the need for handling shadows and high dimension. Experimental results show that detection results are improved obviously and it could meet the requirements of real-time detection.

The majority of cancers undergo metabolic alterations during initiation and progression of disease. Increased de novo lipogenesis is recognized as metabolic mark of cancer cells. The enzymes responsible for de novo fatty acid synthesis are often overexpressed in cancer, indicating that these enzymes are potential and promising targets for novel therapies. Acetyl CoA Carboxylase 1 (ACC1) is a cytosolic enzyme which catalyzes the rate limiting step of de novo fatty acid synthesis through the ATP and biotin-dependent carboxylation of acetyl-CoA to malonyl-CoA. In doing so, it regulates both fatty acid and acetate metabolism. Both pharmacological inhibition and siRNA knockdown of ACC1 have been demonstrated to deny prostate cancer cells of necessary amounts of fatty acids needs to proliferate, resulting in selective toxicity. Although previous research indicates ACC1 and the entire process of fatty acid metabolism to be promising avenues for novel therapy development, much remains to be understood. To address this, we have generated mice with prostate-specific deletion of exon 22 of the Acetyl CoA Carboxylase 1 (ACCL/L) gene by breeding with Probasin-Cre mice. Loss of ACC1 activity does not affect prostate development as survival, prostate weight, histology, and breeding were similar between ACCL/L; Cre+ and ACCL/L Cre- mice. By breeding ACC+/L mice with the established Pten knockout model of prostate cancer (PtenL/L; Cre+), we have found that homozygous inactivation of ACC1 (ACCL/L; PtenL/L; Cre+) is able to reduce tumor burden in mice at 12 weeks of age when compared to ACC+/+; PtenL/L; Cre+ littermates (average total prostate weight of 90mg versus 120mg respectively, normalized to body weight, p0.05). In our efforts to detail why ACC1 inactivation dampens early tumor burden but not later cancer progression, we have found that tumors from both 12 and 24 week old ACCL/L; PtenL/L; Cre+ mice show no changes in lipid profile when compared to tumors arising from age-matched ACC+/+; PtenL/L; Cre+ siblings. Interestingly, through metabolic analysis, we have found that different prostate cancer cell lines utilize fatty acids in unique ways. Under conditions that mimic bioenergetic stress, only androgen insensitive DU145 cells, and not LNCaPs, were able to oxidize exogenous fatty acids. Moreover, treatment with the non-isoform specific ACC inhibitor TOFA (5-Tetradecyloxy-2-furoic acid) abolished fatty acid oxidation under such conditions. These data indicate a potential for metabolic rewiring associated with disease progression and/or metabolic stress, and that blockade of ACC1 may not be sufficient to prevent disease progression. The role of ACC1 in acetate metabolism, redox modulation, fatty acid oxidation, and compensation for ACC1 activity by ACC2 are all potential mechanisms by which ACC1-deficient prostate cancer cells are able to survive and proliferate. Ultimately, these data further the notion that tumors rewire their metabolic circuits to meet the demands associated with cancer biology. Citation Format: Amanda L. Davis, Kristen Scott, Wu Jiansheng, Yong Q. Chen, Steven J. Kridel. ACC1 is required for prostate cancer initiation, but not progression? [abstract]. In: Proceedings of the AACR Special Conference: Metabolism and Cancer; Jun 7-10, 2015; Bellevue, WA. Philadelphia (PA): AACR; Mol Cancer Res 2016;14(1_Suppl):Abstract nr A21.

In order to study the pore water pressure of saturated muck soil under traffic loading of low frequency in Pearl River Delta, the influence of loading frequency, cyclic stress and consolidation on pore water pressure are analyzed. The results show that the final pore water pressure is increased as the frequency decreases, but the increasing rate of pore water pressure is periodical. Moreover, isotropic consolidation can improve the effective stress of soil and slow down the development of pore water pressure, but it is not the key factor. The key factor of influencing the pore water pressure is the cyclic stress, the following is loading frequency, and the consolidation is the last one. In the end, considering cyclic stress, loading frequency and consolidation, a model of pore water pressure based on a math model is built up.

The number of concept in concept lattice is exponential growth, following the growth size of the formal context, we proposed a novel concept: Regular abstraction and researched its existence condition and its resolving method. This paper proposes the method simplifying the concept lattice by regular abstraction of concept lattice.

In this study, Ground penetrating radar is used to detect the quality of back-filled grouting in a shield tunnel. A simulated GPR record shows that the steel bars in lining segments usually generate strong scattering and multiples, thereby obscuring the reflections from objects and making it difficult to locate the objects. Various filtering methods for multiple suppression (i.e., predictive deconvolution, f-k multiple attenuation, and Karhunen-Loeve transformation) were applied to the GPR record of a typical grouting defect model. The ‘wave field prediction and removal method’ is proposed, which involves predicting propagation of the GPR wave field in the lining segment. By subtracting the predicted wave field from the raw GPR record, reflections from objects are precisely abstracted. Application of the method to a synthetic GPR record revealed that the proposed method achieves better results than those obtained using existing methods.

The paper described the basic principle of Naive Bayesian spam filter, and analyzed some major factors which influence Chinese spam filter mostly. On the basis of these, proposed a method to optimize the feature databases. Some experiments were designed to study the filter performance on correctness and efficiency. Some detailed analysis of the experiments also presented in this paper. The experimental results illustrate that the improved method can enhance the precision and efficiency of Bayesian classifier and lower the error rate of legitimate emails.

Objectives The relationship of paraoxonase 1 gene polymorphisms and coronary heart disease has not been confirmed. This study aimed to detect PON-1 genotype (L55M and Q192R) in patients with coronary heart disease and compare with healthy control. Methods Consecutive patients underwent coronary angiography in our catheter laboratory in 2011 were recruited in CHD group or control group in accordance with the results of coronary angiography. L55M and Q192R genotype were detected in both groups by matrix-assisted laser desorption/ionisation time-of-flight mass spectrum (MALDI-TOF-MS). Results A total of 364 cases were recruited in CHD group, 90 cases enrolled in control group. In CHD group, 338 L55L, 26 L55M, 164 Q192R, 46 Q192Q, 154 R192R were detected. In control group, cases of L55L, L55M, Q192R, Q192Q, R192R were 87, 3,49, 13, and 28 respectively. There is no statistical difference between CHD group and control group by χ 2 test. R192R was higher in trend in CHD group (χ 2 measure 3.766, p=0.052). Conclusions Relationship between paraoxonase 1 gene polymorphisms and coronary heart disease has not been confirmed in this study, but a higher trend of R192R was found. Further investigation with larger volume focusing on the biological Significances of PON-1 genotype should be encouraged.

The nanocomposite Nd-Fe-B/FeCo mutilayer films have been prepared by rf sputtering. XRD results and HRTEM observation showed that as-deposited films deposited at substrate temperature of 300K were amorphous. The columnar grains of Nd2Fe14B phase with the c-axis perpendicular to the film plane developed after annealing at 923 K for 15 min. Nd2Fe14B grain size was about 10 nm in average and Nd-O particles of about 7 nm were dispersed within the grains and along the grain boundaries. The magnetization of these samples is found to increase with increasing FeCo layer thickness (tFeCo) and the coercivity decreases with increasing tFeCo. The multilayer films of maximum energy product is 200 kJ/m3 for tFeCo=5 nm at room temperature. The enhancement of the remanence and energy product in nanocomposite films is attributed to the exchange coupling between the magnetically hard and soft phases.

The nanocomposite Nd28Fe66B6/Fe50Co50 bilayer films with Mo underlayer and overlayer have been fabricated on Si substrates. All the samples prepared have perpendicular anisotropy and th e hysteresis loops are simple single loops for fixed Nd-Fe-B layer thickness (10 nm) and different FeCo layer thickness (dFeCo =1—100nm). The shape a nd the behavior of the initial loops and the minor loops suggest that wall pinni ng is the dominant coercivity mechanism in these films. The magnetization of the se samples is found to increase with increasing FeCo layer thickness for a fixed Nd-Fe-B layer thickness and the coercivity of films decrease with increasing Fe Co layer thickness. The sample of maximum energy product is 160×１０３A/m for dFeCo =5nm at room temperature. The enhancement of the rema nence and energy products in nanocomposite bilayer films is attributed to the ex change coupling between the magnetically soft and hard phases.

Firstly,the electronic structure of β-FeSi2 and the density of states (DOS) of subshells of Fe and Si were calculated. The calculations were carried out using the self-consistent full-potential linearized augmented plane wave method within the density functional theory. The general gradient approximation was used to treat the exchange and correlation potential. The DOS of β-FeSi2 is mainly composed of d-shell electrons of Fe and p-shell electrons of Si. Secondly,the stable atomic substitution positions of doping atoms were determined by calculating the total energies for Fe1-xCoxSi2 and Fe(Si1-xAlx)2. The Fe atom at the FeⅡ position and the Si atom at the SiⅠ position were preferentially substituted by Co and Al,respectively. This preferential substitution is in good agreement with results of the existing publications. Finally,the elctronic structures of Fe1-xCoxSi2 and Fe(Si1-xAlx)2 were calculated and analyzed as well. The intrinsic relations between electronic structures of Fe1-xCoxSi2 and Fe(Si1-xAlx)2 and their respective thermoelectric properties including Seebeck coefficient,electric conductivity and thermal conductivity were discussed in detail.

In this paper, by using the linearized augmented plane wave based on the density functional theory of the first principles, the parameters of a unit cell were o ptimized for CoSi firstly. The steady state energy of a many-particle system was -134684297Ry. The corresponding volume and lattice parameter of the unit cell in the steady state were 5899360a.u.3 and 04438nm respectively. Secondly, the calculations of the electronic structures for CoSi and CoSi075A l025 were performed. At the same time, the characteristics of electronic structures of CoSi and CoSi075Al025 were analyzed. The shape of density of states of CoSi in this paper is almost the same as that of the exist ing calculations except for some differences. The Fermi level of CoSi075 Al025 shifted slightly, compared with that of CoSi. Finally, the int rinsic relations between the electronic structures of CoSi and CoSi075Al 025 and their respective thermoelectric properties were discussed in det ail. The increase of material factor B of CoSi075Al025 w as attributed to the doping with Al. Therefore, doping was a practicable means t o improve the thermoelectric property of CoSi.

The effect of density on the thermoelectric properties of the filled skutterudite compound La0.75Fe3CoSb12 was investigated. The La0.75Fe3CoSb12 showed p-type conduction. The mobility increased with an increase in density. The electrical resistivity increased and the thermal conductivity decreased with decreasing density due to the pores scattering. The Seebeck coefficient is independent of density. The portion of electrical resistivity increase is consistent with that of the thermal conductivity reduction. The value of ZT of the specimen with low density is compared to that of specimen with high density. The value of ZT does not depend on density.

Dislocation structures formed during cyclic deformation of niobium single crystals have been studied. Single crystals of both [321] and [110] orientations have been cyclically deformed in tension-compression under strain control. Moreover, the mechanical property data are supplemented by the direct observation of dislocation arrangements by transmission electron microscopy. In the case of [321] orientation single crystals the hardening curve can be divided into three stages, i.e. first, rapid-hardening and saturation stages associated with various dislocation structures. In the first stage the dominan features of the dislocation configurations are high density networks and debris loops. In the rapid-hardening stage the main feature is the formation of dislocation bundles. In the saturation stage the characteristic features are the dislocation bundles and the screw dislocations between bundles. The imposed strain is accomodated mainly by the motion of screw dislocation travelling to and fro between the bundles. For the [110] orientation single crystals the hardening curves are similar to those of the [321] orientation except that dislocation bundles of more than one direction appear during the first few cycles. Three-dimensional cell, two-dimensional cell and bundle structures are summarized as the saturated dislocation structures of b.c.c. metals.

First principles full-potential linearized augmented plane wave method with local density approximation (LDA) was applied to calculate the ground state properties of ReSi1.75 crystal. Optimized results show that the equilibrium lattice constant of ReSi1.75 is smaller than experimental data by about 0.6%. On the basis of LDA results, LDA+U method was used to calculate the electronic structure. When Ueff=U-J=4.4eV, ReSi1.75 has a narrow gap semiconductor band structure with an indirect gap of 0.12 eV and a direct gap of 0.36eV. Effective mass calculation shows the highly anisotropic character of ReSi1.75 crystals. The density of sates changes sharply near Fermi level，thus the thermoelectric properties can be improved by doping.