Modeling of dynamical systems through deep learning

This review presents a modern perspective on dynamical systems in the context of current goals and open challenges. In particular, our review focuses on the key challenges of discovering dynamics from data and finding data-driven representations that make nonlinear systems amenable to linear analysis. We explore various challenges in modern dynamical systems, along with emerging techniques in data science and machine learning to tackle them. The two chief challenges are (1) nonlinear dynamics and (2) unknown or partially known dynamics. Machine learning is providing new and powerful techniques for both challenges. Dimensionality reduction methods are used for projecting dynamical methods in reduced form, and these methods perform computational efficiency on real-world data. Data-driven models drive to discover the governing equations and give laws of physics. The identification of dynamical systems through deep learning techniques succeeds in inferring physical systems. Machine learning provides advanced new and powerful algorithms for nonlinear dynamics. Advanced deep learning methods like autoencoders, recurrent neural networks, convolutional neural networks, and reinforcement learning are used in modeling of dynamical systems.     

Obtaining maximum muscle excitation for normalizing shoulder electromyography in dynamic contractions

Muscle specific maximal voluntary isometric contractions (MVIC) are commonly used to elicit reference amplitudes to normalize electromyographic signals (EMG). It has been questioned whether this is appropriate for normalizing EMG from dynamic contractions. This study compares EMG amplitude when shoulder muscle activity from dynamic contractions is normalized to isometric and isokinetic maximal excitation as well as a hybrid approach currently used in our laboratory. Anterior, middle and posterior deltoid, upper and lower trapezius, pectoralis major, latissimus dorsi and infraspinatus were monitored during (1) manually resisted MVICs, and (2) maximum voluntary dynamic concentric contractions (MVDC) on an isokinetic dynamometer. Dynamic contractions were performed (a) at 30°/s about the longitudinal, frontal and sagittal axes of the shoulder, and (b) during manual bi-rotation of a tilted wheel at 120°/s. EMG from the wheel task was normalized to the maximum excitation from (i) the muscle specific MVIC, (ii) from any MVIC (MVIC_ALL), (iii) for any MVDC, (iv) from any exertion (maximum experimental excitation, MEE). Mean EMG from the wheel task was up to 45% greater when normalized to muscle specific isometric contractions (method i) than when normalized to MEE (method iv). Seventy-five percent of MEE’s occurred during MVDCs. This study presents an 20 useful and effective process for obtaining the greatest excitation from the shoulder muscles when normalizing dynamic efforts.     

马铃薯立枯丝核菌拮抗菌QHZ11的实时荧光定量PCR快速检测与应用

【背景】植物根际促生菌(plant growth-promoting rhizobacteria,PGPR)在根际的定殖是其发挥作用的基础,直观有效的跟踪技术和定量方法是研究PGPR在根际原位分布规律的重要工具。【目的】建立一种马铃薯黑痣病病原菌——立枯丝核菌拮抗菌QHZ11的实时荧光定量PCR快速检测体系,并检测拮抗菌QHZ11在马铃薯根际的动态变化。【方法】根据GenBank中登录的类芽孢杆菌及近源菌株gyrB基因序列差异筛选特异性引物,优化反应条件;通过盆栽试验对马铃薯根际拮抗菌进行快速检测。盆栽试验设3个处理,T1：对照(无菌水,CK);T2：QHZ11菌悬液灌土(QHZ11);T3：将功能菌在有机肥中进行二次固体发酵制成生物有机肥(BOF11)。【结果】筛选出拮抗菌QHZ11的专用引物为gyrB-F/gyrB-R;建立的拮抗菌QHZ11实时荧光定量PCR检测方法特异性好、灵敏度高且重复性较好,线性相关系数为0.999 8,检测组内变异系数均在1%以内,扩增效率为0.9,可检测出1×103?1×1010 copies/g-soil的拮抗菌,具有检出限低和扩增效率高的特点。盆栽试验结果发现,T3处理马铃薯根际拮抗菌QHZ11的数量从接种的第10天即高出T2处理一个数量级,并于马铃薯盛花期(接种后第60天)达到峰值,说明二次固体发酵增加了拮抗菌在根际土壤中的存活和繁殖。【结论】建立的实时荧光定量PCR快速检测体系灵敏、高效,可为研究拮抗菌在马铃薯根际原位分布以及与病原菌的互作方面提供便捷有效的方法。     

Fitting dynamic growth models of biological phenomena from sample observations through Gaussian diffusion processes

This paper addresses the building of stochastic models that adequately describe dynamic phenomena and, in particular, those that occur in the Biosciences. In this context, the empirical fitting of a Gaussian diffusion process from sample data of a dynamic growth phenomenon is considered. In order to do this, a methodology based on approximations to its mean and variance functions is presented. Finally, several applications based on simulated and real data have been carried out.     

Physicochemical study of the protein–liposome interactions: influence of liposome composition and concentration on protein binding

Abstract

The aim of the present study is to investigate the interactions between liposomes and proteins and to evaluate the role of liposomal lipid composition and concentration in the formation of protein corona. Liposomes composed of 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) or hydrogenated soybean phosphatidylcholine (HSPC) with 1,2-dipalmitoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (sodium salt) (DPPG), 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-3000] (DPPE-PEG 3000), cholesterol (CH) or mixtures of these lipids, were prepared at different concentrations by the thin-film hydration method. After liposomes were dispersed in HPLC-grade water and foetal bovine serum (FBS), their physicochemical characteristics, such as size, size distribution, and ζ-potential, were determined using dynamic and electrophoretic light scattering. Aggregation of DPPC, HSPC, DPPC:CH (9:1 molar ratio), and HSPC:CH (9:1 molar ratio) in FBS was observed. On the contrary, liposomes incorporating DPPG lipids and CH both in a molar ratio of 11% were found to be stable over time, while their size did not alter dramatically in biological medium. Liposomes containing CH and PEGylated lipids retain their size in the presence of serum as well as their physical stability. In addition, our results indicate that the protein binding depends on the presence of polyethylene glycol (PEG), CH, concentration and surface charge. In this paper, we introduce a new parameter, fraction of stealthiness (F_s), for investigating the extent of protein binding to liposomes. This parameter depends on the changes in size of liposomes after serum incubation, while liposomes have stealth properties when F_s is close to 1. Thus, we conclude that lipid composition and concentration affect the adsorption of proteins and the liposomal stabilization.     

Electrotactile stimulation on the tongue: Intensity perception,discrimination, and cross-modality estimation

Due to its high sensitivity and conductivity, electrotactile stimulation (ETS) on the tongue has proven to be a useful and technically convenient tool to substitute and/or augment sensory capabilities. However, most of its applications have only provided spatial attributes and little is known about (a) the ability of the tongue's sensory system to process electrical stimuli of varying magnitudes and (b) how modulation of ETS intensity affects subjects’ ability to decode stimulus intensity. We addressed these questions by quantifying: (1) the magnitude of the dynamic range (DR; maximal comfortable intensity/perception threshold) and its sensitivity to prolonged exposure; (2) subjects’ ability to perceive intensity changes; and (3) subjects’ ability to associate intensity with angular excursions of a protractor's handle. We found that the average DR (17 dB) was generally large in comparison with other tactile loci and of a relatively constant magnitude among subjects, even after prolonged exposure, despite a slight but significant upward drift (p < 0.001). Additionally, our results showed that as stimulus intensity increased, subjects’ ability to discriminate ETS stimuli of different intensities improved (p < 0.05) while estimation accuracy, in general, slightly decreased (increasing underestimation). These results suggest that higher ETS intensity may increase recruitment of rapidly adapting mechanoreceptor fibers, as these are specialized for coding stimulus differences rather than absolute intensities. Furthermore, our study revealed that the tongue's sensory system can effectively convey electrical stimuli despite minimal practice and when information transfer is limited by memory and DR drift.     

Analytical methods for detecting pesticide switches with evolution of pesticide resistance

After a pest develops resistance to a pesticide, switching between different unrelated pesticides is a common management option, but this raises the following questions: (1) What is the optimal frequency of pesticide use? (2) How do the frequencies of pesticide applications affect the evolution of pesticide resistance? (3) How can the time when the pest population reaches the economic injury level (EIL) be estimated and (4) how can the most efficient frequency of pesticide applications be determined? To address these questions, we have developed a novel pest population growth model incorporating the evolution of pesticide resistance and pulse spraying of pesticides. Moreover, three pesticide switching methods, threshold condition-guided, density-guided and EIL-guided, are modelled, to determine the best choice under different conditions with the overall aim of eradicating the pest or maintaining its population density below the EIL. Furthermore, the pest control outcomes based on those three pesticide switching methods are discussed. Our results suggest that either the density-guided or EIL-guided method is the optimal pesticide switching strategy, depending on the frequency (or period) of pesticide applications.     

Purification and Characterization of Escherichia coli MreB Protein

The actin homolog MreB is required in rod-shaped bacteria for maintenance of cell shape and is intimately connected to the holoenzyme that synthesizes the peptidoglycan layer. The protein has been reported variously to exist in helical loops under the cell surface, to rotate, and to move in patches in both directions around the cell surface. Studies of the Escherichia coli protein in vitro have been hampered by its tendency to aggregate. Here we report the purification and characterization of native E. coli MreB. The protein requires ATP hydrolysis for polymerization, forms bundles with a left-hand twist that can be as long as 4 μm, forms sheets in the presence of calcium, and has a critical concentration for polymerization of 1.5 μm.     

Protein design in systems metabolic engineering for industrial strain development

Accelerating the process of industrial bacterial host strain development, aimed at increasing productivity, generating new bio-products or utilizing alternative feedstocks, requires the integration of complementary approaches to manipulate cellular metabolism and regulatory networks. Systems metabolic engineering extends the concept of classical metabolic engineering to the systems level by incorporating the techniques used in systems biology and synthetic biology, and offers a framework for the development of the next generation of industrial strains. As one of the most useful tools of systems metabolic engineering, protein design allows us to design and optimize cellular metabolism at a molecular level. Here, we review the current strategies of protein design for engineering cellular synthetic pathways, metabolic control systems and signaling pathways, and highlight the challenges of this subfield within the context of systems metabolic engineering.     

锁定加压钢板和动力加压钢板治疗肱骨中下段骨折的临床比较研究

目的:对比锁定加压钢板与动力加压钢板治疗肱骨中下段骨折患者的临床疗效。方法:抽取我院2010年8月~2013年12月收治的肱骨中下段骨折患者76例,按照随机数字表法分为锁定组(锁定加压钢板治疗)与动力组(动力加压钢板治疗),每组38例,随访1年。对比两组患者临床指标,治疗效果及并发症发生率。结果:锁定组患者手术时间、骨折愈合时间及住院时间均小于动力组,差异均有统计学意义(P0.05);锁定组患者优良率为89.47%,显著高于动力组的71.05%,差异有统计学意义(P0.05);锁定组的并发症发生率为10.53%,显著低于动力组的31.58%,差异有统计学意义(P0.05)。结论:锁定加压钢板应用于治疗肱骨中下段骨折患者疗效优于动力加压钢板治疗,它具有创伤小、恢复快的优点,且并发症发生率较低,值得推广。