Search

Background: Drug-Target interactions are vital for drug design and drug repositioning. However, traditional lab experiments are both expensive and time-consuming. Various computational methods which applied machine learning techniques performed efficiently and effectively in the field. Results: The machine learning methods can be divided into three categories basically: Supervised methods, Semi-Supervised methods and Unsupervised methods. We reviewed recent representative methods applying machine learning techniques of each category in DTIs and summarized a brief list of databases frequently used in drug discovery. In addition, we compared the advantages and limitations of these methods in each category. Conclusion: Every prediction model has both strengths and weaknesses and should be adopted in proper ways. Three major problems in DTIs prediction including the lack of nonreactive drug-target pairs data sets, over optimistic results due to the biases and the exploiting of regression models on DTIs prediction should be seriously considered.

Summary The recent advancements in evaporating water using renewable energy provide one of the promising pathways for treating water with minimized carbon footprint. Pursuing a high evaporation rate of water has been the central focus of this field, as it is directly related to the throughput of evaporative water treatment. However, in conventional designs of evaporators, diffusion of vapor into the atmospheric environment has been the limiting step for evaporation. In this work, we demonstrate that a carbon foam with a three-dimensional interconnected porous structure enables sufficient diffusion of vapor with a convective flow and therefore realizes an evaporation rate as high as 10.9 kg m−2 h−1 and outdoor evaporation of 42.0 kg m−2 for continuous 13 h. With such a high evaporation rate achieved, it shows great promise toward a high-throughput, around-the-clock, and eco-friendly technology of evaporative wastewater disposal.

Dynamic transparent-opaque transition behavior endows the stimuli-chromic materials with the solar modulation capability. However, these materials commonly involve the high manufacturing cost and complexity, the additional consumption of electric energy for solar modulation, or the weak effectiveness of light management. Herein, we develop a low-cost yet broadband light management sodium carboxymethyl cellulose-caging-poly(N-isopropylacrylamide) thermochromic composite (i.e., CMC/PNIPAM), where the nanoscale-skeleton CMC molecules well cage the PNIPAM molecules, which enables the homogeneous dispersion and sufficient distribution of the PNIPAM nanogels in the system. The CMC/PNIPAM features the excellent solar-modulation capability (including optical transmittance modulation of 68.17% and infrared transmittance modulation of 48.50%) and a low phase temperature of 30 °C, as well as the long-term stability of dynamic transparent-opaque transition. Such merits of the broadband light management, low cost, simply fabrication as well as scaling up, make the CMC/PNIPAM function as a promising candidate for the energy-saving buildings and construction.

Complete separation of water and solute is the ultimate goal of water treatment, for maximized resource recycling. However, commercialized approaches such as evaporative crystallizers consume a large amount of electricity with a significant carbon footprint, leading to calls for alternative energy-efficient and eco-friendly strategies. Here, inspired by schooling fish, we demonstrate a collective system self-assembled by expanded polystyrene (EPS)-core/graphene oxide (GO)-shell particles, which enables autonomous, efficient and complete water-solute separation powered by sunlight. By taking advantage of surface tension, these tailored particles school together naturally and are bonded as a system to function collectively and coordinatively, to nucleate, grow and output salt crystals continuously and automatically out of even saturated brine, to complete water-solute separation. Solar-vapor conversion efficiency over 90% and salt production rate as high as 0.39 kg m–2 h–1 are achieved under 1-sun illumination for this system. It reduces the carbon footprint of ∼50 kg for treating 1-ton saturated brine compared with the commercialized approaches., It is important, although very challenging, to realize eco-friendly water-solute separation with minimized carbon footprint. This paper demonstrates a scalable fish-school inspired self-assembled particle system, which enables autonomous, efficient and complete water-solute separation powered by sunlight.

Traditional cooling systems consume tremendous amounts of energy and thus aggravate the greenhouse effect1,2. Passive radiative cooling, dissipating an object’s heat through an atmospheric transparency window (8–13 μm) to outer space without any energy consumption, has attracted much attention3–9. The unique feature of radiative cooling lies in the high emissivity in the atmospheric transparency window through which heat can be dissipated to the universe. Therefore, for achieving high cooling performance, the design and fabrication of selective emitters, with emission strongly dominant in the transparency window, is of essential importance, as such spectral selection suppresses parasitic absorption from the surrounding thermal radiation. Recently, various materials and structures with tailored spectrum responses have been investigated to achieve the effect of daytime radiative cooling6–8,10–15. However, most of the radiative cooling materials reported possess broad-band absorption/emission covering the whole mid-infrared wavelength11–15. Here we demonstrate that a hierarchically designed polymer nanofibre-based film, produced by a scalable electrostatic spinning process, enables selective mid-infrared emission, effective sunlight reflection and therefore excellent all-day radiative cooling performance. Specifically, the C–O–C (1,260–1,110 cm−1) and C–OH (1,239–1,030 cm−1) bonding endows the selective emissivity of 78% in 8–13 μm wavelength range, and the design of nanofibres with a controlled diameter allows for a high reflectivity of 96.3% in 0.3–2.5 μm wavelength range. As a result, we observe ~3 °C cooling improvement of this selective thermal emitter as compared to that of a non-selective emitter at night, and 5 °C sub-ambient cooling under sunlight. The impact of this hierarchically designed selective thermal emitter on alleviating global warming and temperature regulating an Earth-like planet is also analysed, with a significant advantage demonstrated. With its excellent cooling performance and a scalable process, this hierarchically designed selective thermal emitter opens a new pathway towards large-scale applications of all-day radiative cooling materials. A hierarchically designed polymer nanofibre-based film produced by a scalable electrospinning process enables selective mid-infrared emission and effective sunlight reflection, and thus realizes an excellent all-day radiative cooling performance.

The paper presents the design of an all-digital QPSK demodulator, a key component of a satellite digital video broadcast receiver system and chip. The demodulator includes 3 sub-components: symbol synchronization loop; carrier frequency recovery loop; carrier phase recovery loop. The demodulator has a good performance at low SNR, a low system loss and reasonable complexity. The paper presents the design of the loops in detail, and gives the simulation results at optimum coefficients. The paper also shows the results of implementation in FPGA.

This paper presents a design of a 2-bit, 4-level digital correlator implemented with Xilinx Vertex-2 FPGAs. A technique of serial-to-parallel conversion of die input streams is adopted to reduce the speed requirement of the correlation circuits. The correlator can be used for auto- and cross-correlation, and is fully cascadable to form more lags with more FPGA chips. It can run at a clock of 250 MHz. and thus can manipulate input signals with bandwidth of 500 MHz thanks to the 1:4 serial-to-parallel converting scheme.

As an important tool in digital circuit design, Boolean matching can recognize equivalence between two completely specified Boolean functions and improve logic synthesis result. The spectral transform is widely used in Boolean matching. In this paper, a method using the Hadamard transform is proposed to speed up Boolean matching. Experimental results show that our algorithm can finish matching in short time, which satisfies practical demands.