Approximation Analysis of Deep Learning and Related Topics
Hong Kong Principal Investigator: Prof Dingxuan Zhou (City University of Hong Kong)
Mainland Principal Investigator: Prof Zongmin Wu (Fudan University)
Deep learning has been very successful in speech recognition, image classification and many other fields, but its theoretical foundation concerning structured deep neural networks for approximation and generalization has not been well developed yet. It is desirable to have good theoretical understanding of network structures and architectures matching practical applications. In this project we propose to conduct approximation analysis for some deep learning schemes and study some related approximation theory problems and applications. We first plan to carry out generalization analysis for classification with some deep learning schemes including an algorithm with deep convolutional layers of neurons followed by a fully-connected layer. We then plan to study deep learning with functional data to establish generalization error bounds for the produced regression and classification problems. Some problems on function approximation by deep neural networks with structures for deep learning will be studied by approaches from learning theory and wavelet analysis. We also plan to apply deep learning algorithms to some practical applications including readability classification of Chinese text documents.
Design and Mechanism Investigation of High-voltage Solid-state Non-lithium Alkali Metal-ion Batteries
Hong Kong Principal Investigator: Prof Chun-sing Lee (City University of Hong Kong)
Mainland Principal Investigator: Prof Yongbing Tang (Shenzhen Institutes of Advanced Technology (SIAT), CAS)
While lithium ion batteries have been extensively applied, the earth’s lithium reserve is limited. It is thus urgent for the development of alternative battery systems. Sodium and potassium (the sixth and eighth most abundant element on earth), featuring abundant materials-availability, similar physical and chemical properties compared with lithium, have good potential for making practical ion battery systems. Recently, researchers, including the principle investigators, have developed feasible sodium/potassium-based battery systems based on high capacity alloying-type anodes and anion-intercalation-type graphite cathode. While these batteries have merits of high working voltages, low cost and easy recycling, their further commercial applications are limited by: 1) poor cycling stability of the alloying anodes; 2) low capacity of the graphite cathode; 3) possibility of leaking and explosion associated with liquid electrolytes at high working voltages.
In this project, the principles investigators will study and address these issues to design high performance high-voltage solid state sodium and potassium ion batteries. On one hand, high voltage layered cathode materials (modified graphite, transition metal sulfide, nitride, oxide, MXene, etc.), modified alloying anodes (Sn, Ge, Sb, and Bi) and novel solid-state gel polymer electrolytes will be developed. On the other hand, by combining theoretical calculation with in-situ/ex-situ characterizations, we will clearly reveal following problems: 1) pulverization failure mechanisms and mechanical alloying-electrochemical coupling mechanisms of the alloying-type anodes; 2) anion intercalation mechanisms of new layered cathode materials; 3) ion transportation mechanisms and interface stabilities of polymer electrolytes.
Upon success of this project, the PI team will be able to develop a wide range of high-performance sodium and potassium solid state batteries with good safety (with no electrolyte leaking and explosion concerns), high working voltage (> 4.2 V), high reversible capacities (> 120 mAh g-1), high energy density (> 180 Wh kg-1) and good cycling stabilities (capacity retention > 90% after 1000 cycles at 5C). Additionally, the obtained knowledge will be used for design and preparation of other full cells with high capacities, high energy densities, good rate capabilities, and long cycling lives. When these batteries become more matured, companies on knowledge transfer, technology investment and batteries manufacture will be benefited.
Design, Synthesis and Reactivity of new classes of Mn, Co and Fe-based Imido Complexes: Effects of Spin States and Lewis Acids
Hong Kong Principal Investigator: Dr Kai-chung Lau (City University of Hong Kong)
Mainland Principal Investigator: Prof Liang Deng (Shanghai Institute of Organic Chemistry, CAS)
This project is concerned with the development of highly reactive metal imido complexes for selective amination of organic substrates. Amination is the incorporation of nitrogen atoms into organic molecules to give amines, it is an important step in the synthesis of many pharmaceutical products and other useful organic compounds. The search for selective and efficient synthetic routes for amines continues to be a great challenge for chemists. Traditional methods for the synthesis of amines are generally inefficient and with relatively low atom economy. A more viable alternate is to employ a metal imido (M=NR’) species as amination regent, which can be generated by reacting a metal precursor (LM, L represents ligands) with a nitrene (NR’) precursor such as an organic azide (R’N3): M + R’N3 → M=NR’ + N2 and R-H + M=NR’ → R-NH-R’ + M. This approach has a higher atom economy and potentially higher catalytic activity. In addition, the electronic and steric properties of metal imido complexes can be tuned by varying the substituents on the imido group (R’) and other ancillary ligands (L). Although a variety of metal imido complexes are known, most of them are not active enough to undergo amination reaction, especially towards inactivated C-H bonds. In this project we propose to design new classes of highly active metal imido complexes based on earth abundant metals such as Mn, Fe and Co as catalysts for amination reactions. These complexes include M(II) monoimido and M(IV) bisimido complexes (M = Fe, Co) supported by N-heterocyclic carbene (NHC) and α,β-diimine ligands, as well as Mn(V) and Mn(VI) imido complexes supported by macrocyclic tetraamido ligands. Preliminary results show that the spin states of the imido complexes can be tuned by changing the imido substituents, which can be a key factor in determining the reactivity of the imido complexes. We plan to examine in details the effects of spin states of the imido complexes on their amination reactivity.
Frequency-encoded Lithium Niobate Quantum Photonic Integrated Circuit
Hong Kong Principal Investigator: Dr Cheng Wang (City University of Hong Kong)
Mainland Principal Investigator: Dr Xifeng Ren (University of Science and Technology of China)
In October 2019, Google officially announced the achievement of quantum “supremacy” using its latest superconducting quantum processor – Sycamore, solving a specially designed problem (but nothing more) one billion times faster than the best supercomputer on earth. This breakthrough marks a new start of an open contest worldwide to pursue the holy grail of general-purpose quantum computing, which is expected to become the game changer for a variety of industries including cybersecurity and drug discovery. Such realizations, however, will require at least 20 times more quantum bits that can operate in parallel and free of errors.
To achieve this ambitious up-scaling goal, integrated photonics has recently emerged as a particularly attractive platform. Photons are excellent quantum information host, featuring long coherence times and high communication speeds. An integrated platform allows large numbers of optical devices to be patterned on a tiny chip, thus providing the much-needed compactness, scalability and stability.
A major challenge in concurrent quantum integrated photonics is the lack of frequency-domain encoding. As a fundamental property of light, frequency is a natural candidate to create high-dimensional entanglement to further increase the quantum computation power. Frequency encoding, however, has not been possible in popular integrated photonic platforms, e.g. silicon photonics, due to difficulties in achieving efficient and low-loss frequency manipulation simultaneously.
The goal of this project is to address this challenge by developing a frequency-encoded quantum photonic platform using lithium niobate (LiNbO3, or LN), which features excellent electro-optic properties and is ideally suited for frequency-domain operations. However, most integrated LN devices to date are tailored for classical applications only. This project will, instead, focus on the critical performance trade-offs from quantum perspectives and develop the key components for a frequency-encoded quantum circuit. This will be achieved leveraging the complementary expertise of the two PIs: PI (HK) has previously demonstrated a range of ultra-high-performance integrated LN photonic devices, which form the device basis of the project; PI (Mainland) has achieved several milestone quantum integrated photonics demonstrations, and will provide strong theoretical and quantum characterization support for this project.
The successful accomplishment of this project will create a quantum integrated photonic circuit that could achieve the generation, frequency manipulation and quantum interference of frequency-encoded multi-photon states all in one chip. Such technology could find numerous opportunities in quantum computation, secure communications and precision metrology, which could benefit the broad academic and industrial communities both in China and across the globe.
Nanometallic Conductive Composite-hydrogel Core-shell Microneedle Skin Patch for Real-time Monitoring Physiological Signals
Hong Kong Principal Investigator: Dr Chenjie Xu (City University of Hong Kong)
Mainland Principal Investigator: Prof Cheng Yang (Tsinghua University)
The microneedle-based skin patch system allows the minimally invasive extraction of skin interstitial fluid, which offers hopes to realize the quantitative and non-invasive diagnosis/monitoring of biological/physiological signals (biomarkers) in the body. However, diagnosis with exiting microneedle devices suffer from the interferences of the complex physiological processes, thousands of different macromolecules and cells within skin tissue. They cannot realize the real-time conversion of biological/physiological signals to electronic signals either.
To address this unmet need, this project is to develop a new generation of microneedle system that permits the interference-free, real-time, and quantitative detection of biological/physiological signals. Specifically, this platform is based on the development of self-filtering hydrogels (City University of Hong Kong) and the enzymatic electrodes (Tsinghua University), which are responsible for minimizing the interference and allowing the real-time conversion of biological signals to electronic signals, respectively. Two components will then be organically integrated into the microneedle skin patch using the microfabrication technology. Power supply and wireless transmitter components will finally be integrated into the same device. The function of this device will be evaluated in the mouse models by taking glucose, uric acid, and alcohol as the model biomarkers.
Towards Edge-accelerated Computing for Autonomous Driving
Hong Kong Principal Investigator: Prof Jianping Wang (City University of Hong Kong)
Mainland Principal Investigator: Prof Bin Liu (Tsinghua University)
In the foreseeable future, it is expected that autonomous driving will soon reach its momentum as a disruptive technology in transportation. Nevertheless, the success of autonomous driving heavily relies on the safer operation of vehicles, which demands a millisecond-scale response time comparable to or faster than human reflexes. This challenge poses incredible demands to advanced computing and fosters the marriage between Mobile Edge Computing (MEC) and autonomous driving.
Essentially, MEC can provide faster computation for safer autonomous driving because (1) MEC hosts can be physically deployed close to autonomous vehicles, providing millisecond-level round-trip propagation delay, (2) MEC hosts can be equipped with advanced hardware and software components to complete many computing tasks faster than doing so on an autonomous vehicle, (3) MEC can enable faster and more accurate computation by exploiting the incoming data and prior computation results for multiple vehicles.
Despite such salient features of MEC, there are also many fundamental challenges. First, safer autonomous driving counts on valid sensing data, but sensors on autonomous vehicles are vulnerable to various non-contacting physical attacks that can generate fake signals and cause fatal traffic accidents. To mitigate this safety issue, it is challenging to design data validation schemes for MEC to exploit the spatial relations and similarities among objects sensed by multiple vehicles. Second, many computing operations for autonomous driving can be parallelized but there is a lack of solid investigation on how to utilize some parallel computing components, such as ternary content-addressable memory (TCAM) and computation-in-memory (Cim). Third, driving decisions in autonomous driving are made through a set of inter-dependent computing tasks, including some proprietary software that must be executed on vehicles and others that can be offloaded to MEC. Therefore, it is challenging to design task allocation schemes to optimally offload computing tasks to MEC hosts with ultra-low processing delay constraints. Finally, due to the high mobility of vehicles, offloading to MEC may incur frequent service migration.
In this project, we aim to systematically address the above research challenges. In particular, we will develop feature-based cross-vehicle sensing data validation methods with high accuracy and low computation and communication costs. We will design TCAM-Cim (ternary content-addressable memory with computation-in-memory) based fast object detection by reusing historical object detection results. We will study task allocation and resource planning at MEC to meet stringent real-time requirements. Last but not the least, we will design tailored migration strategies to reduce migration delay.
Elucidating the Functional Role of Transcription Factor SIX2 and Its Targeting in Neuroendocrine Prostate Cancer
Hong Kong Principal Investigator: Prof Franky Leung Chan (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Zhiqun Shang (The Second Hospital of Tianjin Medical University)
Prostate cancer is a frequently diagnosed male-specific cancer in many countries and regions, including Mainland China and Hong Kong. Most prostate cancers emerge in the form of adenocarcinoma and are generally androgen-dependent. Advances in detection and therapy for prostate cancer have translated to many patients being successfully treated by surgery and/or radiotherapy. Concomitantly, androgen-deprivation therapy (ADT) has resulted in significant survival gains for patients with advanced prostate cancer. However, majority of the tumors will inevitably progress to a lethal hormone-independent stage (described clinically as castration-resistant prostate cancer/CRPC), which is usually accompanied by reactivation of androgen receptor (AR) signaling, or even to a more aggressive variant, which is acquired with transdifferentiation to a neuroendocrine (NE) phenotype termed as neuroendocrine prostate cancer (NEPC). NEPC or mixed CRPC-NEPC is characterized by the increased expression of NE-associated biomarkers and attenuated AR signaling, leading to resistance to contemporary ADT. Recent studies have witnessed a rising prevalence of NEPC, largely due to the widespread clinical integration of highly potent next-generation AR-pathway inhibitors, implying that the treatment-induced NEPC might be tightly coupled to the suppression of AR signature. To date, the molecular events involved in NE transdifferentiation of prostate cancer remained unclear. Our preliminary study has identified a transcription factor (TF) SIX2 to be a potential regulator in NEPC progression, as evidenced by its increased expression in advanced prostate cancer and NEPC; and its positive correlation with poor pathological and clinical outcomes in prostate cancer patients. Our pilot functional and molecular studies also indicated that overexpression of SIX2 in prostate cancer cells could increase the expression of NE-associated biomarkers and fuel their androgen-insensitive growth capacity; antiandrogen-mediated suppression of AR could dramatically up-regulate SIX2 expression; and also identification of the pluripotency TF SOX2 as its direct target. Based on this, we hypothesize that up-regulation of SIX2 as driven by AR suppression during ADT could facilitate NE transdifferentiation in prostate cancer via its transcriptional control of multiple NE-associated biomarkers and key regulators, particularly SOX2. The main aim of this study is to elucidate the regulatory role of SIX2 in NEPC progression and also shed light on its molecular targets responsible for NE transdifferentiation. We expect that through this study, it not only could provide an insight into the molecular determinants underlying NEPC progression but also evaluate the potential translational value of targeting SIX2 as a novel therapeutic approach for better management of advanced prostate cancer, particularly NEPC.
Mechanism Driven Transition Metal-Catalyzed Trifluoromethylation Using Fluoroform
Hong Kong Principal Investigator: Prof Chit Tsui (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Qilong Shen (Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences)
Trifluoromethylated molecules play a privileged role in pharmaceuticals, agrochemicals and specialty materials. Many of the top-selling drugs contain the trifluoromethyl (CF3) group, including the anti-HIV Efavirenz, antacid Lansoprazole and antidepressant Fluoxetine. The search for more efficient and sustainable synthetic methods to introduce CF3 group into organic molecules is a challenging and important goal in both academia and industrial sectors. Fluoroform (CF3H) has recently emerged as a highly attractive CF3 source for developing trifluoromethylation reactions. It is a large-volume industrial byproduct from Teflon manufacturing and a potent greenhouse gas. Compared to other nucleophilic, electrophilic and radical trifluoromethylating reagents, using fluoroform has the advantages of better atom economy, lower toxicity, lower cost and environmental friendliness. Despite the fact that activation of fluoroform by transition metals is known, the catalytic trifluoromethylation using fluoroform has been virtually unexplored.
In this connection, we propose the current joint project by combining the expertise of the Hong Kong and mainland teams to develop the first transition metal-catalyzed trifluoromethylation using fluoroform as the CF3 source, guided by mechanistic understandings. Systematic studies of the active reaction intermediates in each elementary step will provide key mechanistic insights to guide the design of catalytic cycles and optimization of reaction conditions. Implementation of the developed catalytic processes will lead to useful, practical and scalable syntheses of value-added trifluoromethylated compounds with diverse structures. The outcome of the proposed project should have a significant influence on both fundamental research and industrial applications.
Molecular Mechanism Study of the Reciprocal Regulation of SnRK1 Kinase and Plant Autophagy
Hong Kong Principal Investigator: Prof Xiaohong Zhuang (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Caiji Gao (South China Normal University)
Being sessile organisms, plants have to obtain nutrition in a relatively fixed location. To compensate their immobility, plants have involved sophisticate mechanisms to sense energy and nutrient availability. The sucrose non-fermenting 1 (SNF1)-related protein kinase 1(SnRK1) complex, a subfamily of serine/threonine kinases, functions as a master sensor of nutrient status in plants by phosphorylating diverse downstream proteins. Multiple cellular processes, including lipid/sucrose metabolism and autophagy, will be activated to coordinate the plant energy/nutrient balance and finally to modulate plant growth and environmental adaptation. Key scientific questions in this research area include what these specific substrates of SnRK1 in response to different environmental signals are, and through what kinds of positive and negative feedback regulatory mechanisms to precisely modulate the signal input and output of SnRK1-regulated signaling pathway. Although the important functions of SnRK1 in plant stress responses have been demonstrated via the phenotypic analysis of snrk1 related plant mutants, our current understanding of the feedback regulation mechanisms of SnRK1 kinase activity, as well as the molecular mechanism of the reciprocal relationship between SnRK1 kinase and autophagy are still largely unexplored in plant cells. In our preliminary studies, we identified two SnRK1 substrate candidates, including the plant-unique autophagic regulator SH3P2 and one plant-specific protein SIP1(SnRK1-interacting protein 1) with unknown function. Therefore, in this research project we firstly aim to characterize the functional role of SnRK1-SH3P2 interaction in the regulation of plant autophagy. Moreover, we will investigate the molecular mechanism of SnRK1 kinase activity regulated by the plant specific protein SIP1. Last but not least, we will also study the role of selective autophagy for SIP1 degradation in response to the changes of external and cellular energy levels like the sucrose concentration, as well as the novel function of a SIP1-dependent positive feedback loop in the regulation of SnRK1 activity during plant autophagy. We anticipate that this research project will uncover novel insights into the feedback control mechanism of SnRK1 signals in plant autophagy pathway, and unveil novel interaction network of the SnRK1-dependent autophagy and sugar signaling pathways in the regulation of plant energy and nutrient homeostasis.
Biomimetic Platform of Perfusable Organoid-chip for Studying the Molecular Pathogenesis of Retinitis Pigmentosa
Hong Kong Principal Investigator: Prof Hon-fai Chan (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Jiansu Chen (Jinan University)
Retinitis pigmentosa (RP) represents a group of inherited retinopathies with a prevalence of 1:3500 in the world and 1:4000 in China. Currently, 81 causative gene loci associated with RP have been identified. While RP can cause night blindness, visual field defect and potentially total blindness, the pathogenic mechanism is still unclear and no effective cure is available. Traditional approach of studying the disease in mice model is not ideal since late onset of photoreceptor degeneration is observed in mice and there are evolutionary differences between human and mice models. Compared with animal models, retinal organoids (RO) have been shown to better mimic human ocular physiology and pathology, and are amenable to high-throughput screenings. Nevertheless, RO has limited lifespan in vitro (< 300 days) because vasculature in RO is normally missing to supply nutrient. In addition, without the proper interaction with the photoreceptors and retinal epithelial cells (RPE), the current configuration of RO is simplistic and semi-physiological only. Consequently, the potential of applying RO in investigating the pathogenic mechanism of RP, which targets photoreceptor and RPE primarily, has not been fully realized.
To elucidate the pathogenic mechanism of RP caused by various mutations, we have successfully established multiple RP patient-specific iPSC cell lines with various mutations, followed by RO construction from the iPSCs. We also reported the similarities and differences in the developmental features between RO and native retina based on transcriptomic analysis. In order to further improve RO configuration, we will adopt microfluidic organ-on-a-chip technology to integrate the culture of RO and other cell types (e.g. RPE) in a single perfusion device. Moreover, we will also construct a “nichoid” platform as photoreceptor scaffold using two-photon polymerization that can recapitulate the structure of human retinal outer nuclear layer.
In this project, we will explore the use of a perfumable retinal organoid chip (ROC) to model and study RP with an aim to providing new treatment options for RP. We hypothesize that the combination of perfusion culture and biomimetic scaffold should foster the development and co-culture of RO, photoreceptor and RPE. The overall goals include elucidating the molecular mechanism of RP onset and progression, exploring the correlation among RP patient genotype, RP clinical characteristics, and the phenotype of RO/ROC derived from RP patients (genotype-phenotype-organoid), and identifying potential therapeutics for RP treatment. This will contribute greatly to our knowledge of RP disease development and offer novel insights on new treatment strategies.
Combined Effects of Ozone Pollution and Nitrogen Deposition on Forest Primary Productivity and Water Use in China
Hong Kong Principal Investigator: Prof Amos Pui-kuen Tai (The Chinese University of Hong Kong)
Mainland Principal Investigator: Prof Zhaozhong Feng (Nanjing University of Information Science & Technology)
The rising demands for food and energy from a rapidly growing world population have substantially increased the quantities of reactive nitrogen (N) including nitrogen oxides (NOx) and ammonia (NH3) released into the atmosphere, mostly via fossil fuel combustion and agricultural activities. One of the most pressing consequences is ozone (O3) pollution, for which NOx is a major precursor and which poses significant threats to both human and vegetation health. Another consequence is the increased N deposition onto the land surface, which can enhance plant productivity especially in N-limited forest ecosystems. O3 pollution and N deposition thus have significant effects on terrestrial ecosystems, biogeochemical cycles and climate at large. Most studies to date have examined their effects only separately, but their interactive and combined impacts on forest productivity and water use, as well as the different sensitivities of representative Chinese tree types to O3 and soil N, are largely uncertain but potentially significant.
This proposal thus aims to address this knowledge gap by combining carefully designed open-top chamber (OTC) experiments with state-of-the-art computer model development and regional simulations. At two Chinese sites (in Beijing and Nanjing), a series of OTC experiments with five O3 fumigation levels and four N fertilizer application levels will be conducted for several representative species of major plant types (northern vs. southern; deciduous vs. evergreen). Important plant, soil and meteorological variables will be regularly measured. The experimental results will be used to build statistical models to quantitatively represent the interactive effects of O3 pollution and N deposition on forest primary productivity and water use, which will then be implemented into a high-performance computer modeling framework consisting of one atmospheric chemistry model (GEOS-Chem) and three options of biospheric models (CLM, YIBs and TEMIR), all driven by consistent prescribed meteorology. This modeling framework will be validated against a consolidated database for Chinese meteorology, air quality and vegetation for the recent past, and then used to conduct an ensemble of simulations for the present day (around year 2020) and future (around year 2050 and 2100) following selected Shared Socioeconomic Pathways. The simulated results will help us ascertain the differential and integral roles of O3 pollution and N deposition on forest ecosystem functions and services, with important utilities for Chinese agencies and stakeholders to devise optimal air quality and forest management strategies to simultaneously safeguard human and ecosystem health.
Ultracompact Diffractive Hyperspectral Array Imaging (DHSAI) based on Tunable Microfluidic Lens Array
Hong Kong Principal Investigator: Dr Xuming Zhang (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Weixing Yu (Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences)
Hyperspectral imaging (HSI) captures the images of an object at many wavelengths and enables to identify the materials and other critical information using the spectral signatures. Recently it has become increasingly popular for remote sensing, medical diagnostics and food processing. However, the traditional systems often employ mechanically moving components for spectral scanning and suffer from large size, low resolution and slow frame rate. To overcome these problems, this project will develop a new type of diffractive hyperspectral array imaging (DHSAI) system. It consists of a diffractive lens array, a tunable microfluidic lens array (MLA) and a flat image sensor (e.g., CCD or CMOS). Every diffractive lens and the corresponding microfluidic lens work together as an imaging unit to capture a small part of the object, like the ommatidium of a compound eye. The spectral scanning is done by adjusting the focal length of microfluidic lens using electric field. The DHSAI system is designed to target at superior specifications: module size < 6 × 5 × 0.3 cm3, spectral range 250 – 780 nm, spectral resolution < 1 nm, and frame rate ~33 fps.
This project will be conducted by the joint efforts of both the Hong Kong team and the Mainland team. The scientific contribution lies mostly in three aspects: (i) the system design of DHSAI is novel and patentable; (ii) the array imaging method mimics the compound eyes and is new to spectral imaging; and (iii) the proposed MLA-based confocal scanning imaging method is original, it provides a non-mechanical, fast spectral sweeping method. We have conducted pilot studied on the tunable microfluidic lenses, obtaining the tuning of focal length over 1.6 – 9.5 mm with a low aberration (about 1/4 of that of a spherical lens). Although preliminary, these results give our great confidence to the feasibility of this project. To our best knowledge, this is the first attempt to miniaturize the HSI system using microfluidics. In a long run, it may provide an HSI module for smartphone and would enable the HSI for portable, real-time, daily applications such as skin self-check, food quality inspection, fruits/vegetables selection and many others. This would benefit a large population of non-professional civilians (e.g., housewives, fruit retailers).
Micro Optimal Design Strategy for High Performance Three-dimensional Negative Poisson’s Ratio Lattice Structures
Hong Kong Principal Investigator: Prof Hong Hu (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Jian Xiong (Harbin Institute of Technology)
Negative Poisson’s ratio (NPR) lattice structures exhibit unusual deformation behavior: they laterally contract, rather than expand, when compressed. The NPR lattice structures, especially those in 3D forms, have attracted considerable attention in recent years because of their intriguing mechanical properties and numerous promising applications, particularly in stringent environments, such as in aerospace and defense fields. The 3D NPR lattice structures exhibit superior dynamic loading and energy absorption capabilities compared to non-NPR materials and structures. However, a remaining barrier to their application is their low load bearing capability. Despite attempts to address this by implementing an embedding enhanced strategy, no satisfactory and systematic design strategy has been developed. In addition, commonly used basic materials for fabricating the 3D NPR lattice structures are single type of materials such as metals and polymers of which effect on improving the mechanical properties of 3D NPR lattice structures is limited.
In this research project, we will develop a relatively universal design strategy by considering both micro deformation mechanism and basic component materials in order to achieve a novel class of 3D NPR lattice structures that simultaneously have high load bearing and energy absorption capabilities. First, a class of novel 2D NPR lattices with high load-bearing capability will be designed based on the stretching-dominated deformation mechanism. Then, the 2D NPR lattices will be extended towards three dimensions and a topology optimization procedure will be developed for the 3D NPR lattices to simultaneously achieve the optimal load-bearing and energy absorption capabilities. Next, high-performance carbon fiber reinforced composites will be used as the basic materials to fabricate these new 3D lattices using an interlocking method. Finally, a combination of experimental, analytical, and numerical approaches will be adopted to systematically investigate the mechanical properties of these new 3D NPR lattices under the quasi-static, local and non-local impact loading conditions. The theoretical models, numerical simulations and experimental results will be compared in order to fully study the load bearing capability, energy absorption capability and the failure mode of these new 3D NPR lattices. The research will advance and guide the design, fabrication, and application of the 3D NPR lattice structures for engineering applications.
Vascular Ageing In Osteoarthritis: from Mechanism Towards Intervention
Hong Kong Principal Investigator: Dr Chunyi Wen (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Baohua Liu (Shenzhen University)
Osteoarthritis (OA) is a leading cause of chronic pain and disability in older adults. The hallmark of OA is loss of articular cartilage. Although aging represents one of the most important risk factors for OA, the major flipside up to this moment is the mechanisms leading to the age-related cartilage loss remain unclear. In particular, early changes that predispose to chondrocyte senescence and cartilage matrix disruption are not well characterized.
Vascular ageing drives systemic ageing process, including joint ageing and OA. Endothelial ET-1/ETBR expression elevated with vascular ageing in the Lmnaf/f::Tie2-Cre progeroid mice. Under the auspices of RGC, we previously discovered that endothelium-derived endothelin-1(ET-1), a potent vasoconstrictor, induces chondrocyte senescence and cartilage degradation through endothelin type B receptor (ETBR) in murine posttraumatic OA. The overexpressed ET-1/ ETBR hence could predispose joint ageing and OA in the progeroid mice. Yet the cellular role of endothelium and molecular machinery remain to be elucidated. Additionally, endothelium targeted Sirt7 replenishment could restore vasodilation function and extend the health span of progeroid mice. This paves the way for the therapeutic value of Sirt7 in vascular ageing-induced OA. Collectively, we hypothesize that vascular ageing contributes to chondrocyte senescence and cartilage damages in OA initiation and progression through ET-1/ETBR and/or Sirt7 signaling pathway.
Current therapies for OA are limited to symptom relief. This project will shed the light on the role of endothelium and novel molecular basis in joint ageing and OA. Our findings will open up a new avenue for mechanism-based disease-modifying OA drug discovery.
A Study of the Microstructures and Properties of Laser Additive Manufactured Refractory High Entropy Alloys and Composites
Hong Kong Principal Investigator: Prof Kang Cheung Chan (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Lin Liu (Huazhong University of Science and Technology)
Refractory high-entropy alloys (HEAs), composed of multi-principal refractory elements, are a new branch in the family of high-entropy alloys. Due to their high melting points, excellent strength, superior thermal stability and sluggish atomic diffusion, refractory HEAs, as a new generation of high-temperature alloys, are expected to be utilized at 1000 ℃ or even above. However, the difficulties in fabrication and processing using traditional approaches block the development and application of refractory HEAs. Recently, laser additive manufacturing (also called laser 3D printing) has been proven to be a feasible route to fabricate refractory HEAs. With the extensive research background of the project teams on the synthesis and properties of conventional HEAs and additive manufacturing, the project aims to develop novel refractory HEAs and composites with promising high-temperature properties using laser engineered net shaping (LENS) process. The HfNbTaTiZr system is chosen as the base alloy for its superior combination of high strength and excellent ductility among various refractory HEAs. A mixture of elemental powders serves as the precursor, then the refractory HEAs are prepared via in-situ alloying of the mixed powders, while mechanical parts with customized geometries can be made simultaneously. To improve the ductility at room temperature and the strength, thermal creep resistance and anti-oxidation performance at elevated temperatures, the refractory HEAs composites, reinforced with nano-sized metal-carbides or metal-silicide, are in-situ prepared via the addition of a small amount of C or Si powders into the powder mixture of the base alloy. Because of the fast heating and cooling rate of the laser 3D printing technique, composition segregation can be highly suppressed and microstructures of refractory HEAs could be significantly refined, finally leading to enhanced mechanical properties of the refractory HEAs. The mechanical performance and high-temperature creep and oxidation behaviour are investigated through detailed analysis of the microstructural evolution, and the origins of the enhanced high-temperature properties are further revealed. In addition, effort is made to tune the as-built microstructures and optimize the high-temperature performance of HEAs and composites using the ultrasound-assisted LENS process. The outcomes of this project will not only provide methodologies for the preparation of refractory HEAs and composites but will also contribute a better understanding of the origins of high-temperature properties of refractory HEA composites. This project also suggests a new research direction and will lay a foundation for developing refractory HEAs and their composites for advanced aerospace applications.
Organic Aerosols in Offshore Marine Atmosphere of Eastern and Southern China: Sources, Chemical Ageing, and Climate Consequence
Hong Kong Principal Investigator: Prof Hai Guo (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Likun Xue (Shandong University)
Organic aerosols (OA) constitute increasing proportions of airborne particulate matters (PM) in China, despite reduced emissions of dust, sulphur dioxide and nitrogen oxides. Due to insufficient understanding on their sources and formation mechanisms, OA have been an obstacle to combating air pollution. Offshore marine atmosphere (OMA) connects the continental and marine air, where air pollutants of land and sea origins interact, and the atmospheric oxidative capacity may be enhanced. These lead to complicated sources and formation mechanisms of offshore marine OA (OMA-OA). Under sea breezes and land breezes, OMA-OA can be recirculated to coastal cities and transported to open oceans, respectively, affecting regional air quality and climate.
Only a handful of studies provided some perspectives on OMA-OA in China, most of which reported the chemical compositions and implied origins of OMA-OA. The documented composition of OMA-OA usually lacked completeness or species-resolution, and its implication on the origins of OMA-OA was uncertain or not dynamic enough to adapt to changing OMA. Moreover, there is almost a blank in the formation and ageing mechanisms of OMA-OA. This project will focus on OMA-OA in eastern and southern China in the following aspects: (i) molecular level chemical characterization, and identification of definite and dynamic sources; (ii) cross-regional transport; (iii) formation and ageing mechanisms; and (iv) environmental and climatic impacts.
Field measurements of OMA-OA will be conducted in Qingdao and Hong Kong, coastal cities in eastern and southern China, respectively. A cruise measurement from Yellow Sea, passing East China Sea, to South China Sea will provide a complement to the comprehensive observations of OMA-OA. A novel technique of gas chromatography–mass spectrometry (GC-MS) coupling with online derivatization and thermal desorption, in addition to other online mass spectrometers, will enable highly time-resolved measurement of OA on molecular levels. The comprehensive measurements will further promote the determination of OMA-OA and research on dynamic sources. The cruise measurement links OMA-OA in eastern China to those in southern China, indicating cross-regional transport. To understand formation and ageing mechanisms of OMA-OA, we will perform on-site and in-lab ageing experiments using oxidation flow reactors, as cross-validation of current secondary organic aerosol (SOA) formation mechanisms. The environmental and climatic implications of OMA-OA will be assessed using chemical transport models and integrated data analysis.
This project will enhance current knowledge about OMA-OA, provide a reference for coastal cities to reduce OA, and point out the impact of OMA-OA on climate.
Study of Multi-Physics Coupling Mechanism of Deep Sea Pipeline and Soft Marine Organic Deposits Interaction Considering Time and Temperature Effects
Hong Kong Principal Investigator: Dr Zhenyu Yin (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Fuping Gao (Institute of Mechanics, Chinese Academy of Sciences)
Deep-sea oil and gas in the South China Sea are important strategic resources for both mainland China and Hong Kong. Indeed, more than 25% of the total electricity generation in Hong Kong depends on a gas supply from the South China Sea. Submarine pipelines are effective and only viable means for transporting oil and gas from this area. However, complex marine environmental conditions in the deep sea in conjunction with high pressure and temperatures inside the pipelines create a range of potential hazards, including pipeline instability and failure that may lead to great economic loss annually. The main challenges to safe operation of deep-sea long-distance pipelines are the wide distribution of problematic soils, namely soft marine organic deposits (SMOD), and the complex thermo-hydro-mechanical (THM) loading conditions that these pipelines have to encounter in deep water. It remains poorly understood of the combined influence of SMOD and THM loadings on pipeline–seabed interactions and the resulting failure of the pipeline system. Effective prediction, mitigation and prevention of possible hazards facing the deep-sea pipelines urgently require better understandings of the coupling mechanism of deep-sea pipeline with SMOD seabed while taking into account the effects of time and temperature.
This project is intended to elucidate the mechanism of interactions between deep-sea pipeline and SMOD seabed by taking into account the effects of time and temperature. The micro and macro characteristics of seabed SMOD will be experimentally examined. A multi-scale model will be further developed to properly evaluate related soil behaviour of SMOD, and a series of scaled model tests will be developed to study the pipeline–SMOD seabed interactions, from two-dimensional sectional condition to three-dimensional buckling under THM loading. Innovative multi-scale, multi-physics finite element modelling approach will be developed to evaluate the model tests and to predict the pipeline system performance under complex loading conditions. A macroelement design tool for assessing the THM buckling of deep-sea pipelines will be developed for convenient application in the engineering field.
This project will offer the first systematic multi-scale and multi-physics investigation of interactions between deep-sea pipelines and SMOD seabed. It will produce a novel and reliable coupled THM modelling tool that provides a rational basis for understanding the buckling behaviour of the pipeline–seabed system, along with an innovative practical design tool for pipeline–seabed systems. It will also contribute significantly to the mitigation and prevention of buckling hazards to deep-sea pipelines worldwide.
Virtual-reality Assessment of Social Cognition Impairment and its Related Neural Mechanism in Schizophrenia Patients and At-risk Individuals
Hong Kong Principal Investigator: Prof Ho Keung David Shum (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Raymond Chor-kiu Chan (Institute of Psychology, Chinese Academy of Sciences)
Schizophrenia is a debilitating psychiatric condition that affects one’s thoughts, emotions and behaviours. Clinically, patients with schizophrenia are known to experience difficulties with social cognition (the ability to interpret and use information in a social context). This impairment adds to the employment and socialisation difficulties faced by patients with schizophrenia. However, due to the complex nature of social cognitive skills, there is a lack of ecologically valid tools for measuring social cognition in adults. As a result, the level of impairment and underlying mechanisms of social cognition in patients with schizophrenia remain unclear. As most social assessment tools are still based on paper and pencil and as this type of assessment relies on observation and subjective judgement of rigid and simple parameters, there is a demand for advanced and quantitative assessment approaches based on ecologically valid scenarios. Here, virtual reality (VR) technology offers a promising solution. Using VR, interactive, realistic and subtle virtual environments will be created to capture people’s real-life social interaction styles and abilities. Carefully programmed VR will capture data on user action in clinical and non-clinical populations. In addition, intelligent machine learning algorithms will be embedded to extract the features of the VR-captured data and perform a diagnostic classification. Attributed to the flexibility of VR, a variety of virtual environments will be built based on theory of mind and empathy, two of the main constructs of social cognition.
The proposed project will have the following aims: 1) further develop and refine a VR-based social cognition test that we pioneered, and establish its reliability and validity in the general population; 2) investigate the neural correlates of social cognition in VR environments in a healthy population; and 3) examine and compare the level of social cognitive deficits in patients with a first episode of schizophrenia, at-risk individuals (people with schizotypal traits) and healthy individuals and investigate the neural mechanisms underlying these group differences. Together, the results of our proposed project are expected to improve the assessment of social cognition in healthy and clinical individuals, identify the early diagnostic markers of schizophrenia and better understand the social deficits of the disorder. In addition, the project is expected to provide a solid foundation for better intervention against the disease.
Research and Development on Waste Ash/Slag-based Artificial Aggregates for Applications in Concrete
Hong Kong Principal Investigator: Prof Jianguo Dai (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Biqin Dong (Shenzhen University)
Construction industry plays a vital role in the urban and economy development in Hong Kong as well as the mainland China. Concrete, due to its cost-effectiveness and wide availability of raw materials, has been the most widely used material in construction industry. The aggregate as a structural filler, in addition to the water and cement in concrete, occupies the most of the concrete volume. However, due to the unprecedented scale of modern construction activities, the availability of aggregates, which are usually made from natural stones/crushed rock and river/washed sea sand, is now facing a critical natural resources shortage problem. The large-scale excavation has also severely imposed a threat on the natural environment. On the other hand, with the rapid economic expansion and urbanization, Hong Kong and mainland China have seen a great increase of the industry waste ash/slag arisen from steel, electricity, coal and other industry sectors as well as the incinerator ash discharged from municipal solid waste incineration facilities. For example, approximately more than 3500 tons of municipal waste incineration fly ash (MWIFA) is produced in mainland China every day. These MWIFA usually contains different types of soluble salts and heavy metal and need to be processed to remove these contaminants, which is very costly, before it is reused for construction. Landfill is the last resort for the disposal of these MWIFA. However, landfills need notable land resources, which is one of the most precious assets of big cities. This project aims to develop a ‘one stone two birds’ solution to the above-mentioned dilemma through a geopolymer-based approach to manufacture waste ash/slag-based artificial aggregates (WASAAs) for concrete production. Geopolymer is a Si/Al inorganic polymer that can be formed through the chemical activation of alumina-silicate precursor materials and necessary engineering strength can be thus achieved if an appropriate mix design is made. It is worth noting that the typical waste ash/slag (such as MWIFA and red mud) are composed of aluminum and silicate phases. Another distinct feature of geopolymer is its excellent heavy metal removal capacity, which has been demonstrated by many previous researches. The project has four specific objectives including the mix proportion design of WASAA, the engineering properties and environmental acceptance of WASAA, and the short-term and long-term properties of WASAA concrete. The project completion will lead to an innovative value-added avenue for relaxing the aggregate shortage problem and creation of a significant body of scientific knowledge in geopolymer concrete technology.
Research on Structural Behaviour of Stainless-Clad Bi-Metallic Steel Welded Connections and Joints under Monotonic and Cyclic Actions
Hong Kong Principal Investigator: Prof Kwok Fai Chung (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Dr Huiyong Ban (Tsinghua University)
Stainless-clad (SC) bi-metallic steel is an advanced high-performance steel (HPS) that consists of two different metals, typically i) S235, S355 and S460 carbon-steels, and ii) S316L stainless-steel, being bonded together at metallurgic level through explosive welding. Use of the SC bi-metallic steel sections in maritime structures is highly desirable in terms of structural performance, corrosion resistance, and cost efficiency. While their use is highly beneficial in achieving both structural adequacy and life cycle costs, there is a concern on the structural performance of welded sections and joints among these steel sections under both monotonic and cyclic actions, in particular, integrity along bonding interfaces of the two component metals under seismic actions. In order to promote wide applications of these SC bi-metallic steel sections in construction, it is essential to demonstrate structural integrity of these metallurgically bonded interfaces in the vicinity of welded sections and joints. Scientific understanding and engineering data on the mechanical properties of the interfaces under large deformations will release concern of structural engineers, and facilitate structural design of these SC bi-metallic steel sections in construction.
This project aims to promote effective use of high-performance steels in construction under severe corrosive environments, such as maritime and offshore structures in tropical areas. With the collaborative research on basic mechanical properties and corrosion resistances of these plates at the Tsinghua University, scientific understandings on structural behaviour of these fabricated SC bi-metallic steel sections will be obtained to enable effective design and construction of maritime structures with service lives at 120 years under severe atmospheric corrosion in affordable costs.
This project is proposed in line with the recent national development policies of the Great Bay Area of Guangdong-Hong Kong-Macau under the 13th National Strategic Development of the Chinese Government from 2016 to 2020, and Hong Kong is well positioned to contribute to infrastructure developments in the South China Sea.
Coupling Hydrogen Production by High Temperature Electrolysis with Iron-and-Steel Works to Recover Industrial Waste-Heat for Low-Carbon Metallurgy
Hong Kong Principal Investigator: Prof Meng Ni (The Hong Kong Polytechnic University)
Mainland Principal Investigator: Prof Minfang Han (Tsinghua University)
As a reversible electrochemical device, solid oxide cell (SOC) has high power generation efficiency and electrolysis efficiency in fuel cell mode and electrolysis mode, respectively. This project intends to build an integrated system for efficient hydrogen production and utilization based on reversible SOC. The SOC can recycle the surplus renewable power such as wind power and photovoltaic power and a large amount of waste heat generated in the industrial production process for hydrogen production and energy storage through high-temperature water electrolysis. The hydrogen produced can replace carbon used in iron-and-steel works or fuel cell system. Firstly, the feasibility of the whole system will be studied from the perspectives of energy utilization efficiency, carbon emission and economic efficiency. At the same time, as the core technology of the system, the reversible SOC, will be studied, including the optimization of the performance and stability of the single cell in the alternate operation of the fuel cell mode and electrolysis mode, as well as the multi-physical field simulation and experimental verification of the reversible SOC short stacks. This project will provide a reliable technical route for the efficient recovery and utilization of waste renewable power and industrial heat, and will closely integrate new low-carbon and efficient energy technologies with traditional industrial systems, thus opening up a new way for the establishment and development of low-carbon energy system and society.
Investigating the roles of SORL1 genetic variants in the pathogenesis of Alzheimer’s disease
Hong Kong Principal Investigator: Prof Nancy Y Ip (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Prof Yu Chen (Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences)
Alzheimer’s disease (AD) is a leading cause of mortality in the elderly. The complex pathophysiology of AD has hindered the development of effective diagnostic tools and treatments. Genetics studies provide important information about the pathological mechanisms of AD and genotype–phenotype relationships. However, the pathological roles of most of these disease-associated genes have only just begun to be studied, and it remains largely unclear how noncoding genetic variants contribute to the disease pathogenesis. Our previous works constitute the first whole-genome sequencing study of AD in the Han Chinese population. Accordingly, we identified SORL1 (sortilin-related receptor 1) as an AD-associated gene as well as several AD-associated variants residing in this gene in the Chinese population. In the proposed study, we aim to understand the functions of SORL1 in AD and how the previously identified AD-associated genetic variants contribute to the dysfunction of the gene. The aims of the proposed study are as follows: (1) perform fine-mapping of genetic variants in a key AD-associated locus, SORL1, in our in-house, whole-genome sequencing database of the Chinese AD population; (2) validate these variants in additional cohorts and examine their associations with AD-related endophenotypes; and (3) determine how specific variants regulate neuronal function and AD pathogenesis. The findings of the proposed study will not only advance our understanding of the pathophysiology of AD but will also provide novel insights for the development of patient stratification strategies and therapeutic interventions.
Evolution of landslide hazard chains triggered by strong earthquakes and the associated dynamic risk management
Hong Kong Principal Investigator: Dr Li Min Zhang (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Dr Ming Peng (Tongji University)
Strong earthquakes in mountainous areas often trigger a large number of landslides in along the faults. The loose landslide deposits and deteriorated slopes are prone to be reactivated by subsequent triggers like rainfall and earthquakes, forming geological hazard chains (GHC) including avalanches, landslides, debris flows, landslide dams, dam-breaching floods and so on. The GHCs could produce lasting threats to the human lives and infrastructures in a very large area and impede the construction of large projects such as the Sichuan-Tibet railway and large-scale cascade hydropower stations. This research project aims to reveal the evolution mechanisms of GHCs triggered by strong earthquakes and conduct corresponding dynamic risk assessment based on multi-source information fusion. First, a database of earthquake-induced hazard chains will be compiled and applied to build a machine-learning model for forecasting the distribution and characteristic parameters of landslide hazard chains. Second, shaking table tests will be conducted both onboard a centrifuge and in the laboratory, and mass transport tests will be conducted on a large drum centrifuge to investigate the formation mechanisms and evolution of the GHCs. Third, a numerical model for GHC evaluation will be developed by considering the mass and energy conservation of two-phase fluid. Energy and mass criteria will be established to quantitatively determine the escalating points in a representative earthquake-induced hazard chain. Finally, a dynamic risk assessment and management system will be established by using Bayesian network and time series methods. The research results will provide scientific basis for understanding the evolution and disaster-causing mechanisms of GHCs, and for protecting the safety of human lives, properties and major projects in earthquake prone areas.
Artificial Intelligence Methods for Early Diagnosis of Pancreatic Cancer in CT Imaging
Hong Kong Principal Investigator: Prof Kwang-Ting Cheng (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Prof Xin Yang (Huazhong University of Science and Technology)
Pancreatic cancer (PC) is one of the most malignant tumors with the lowest 5-year survival rate among all cancers. Improving the capability of early diagnosis via CT, which is critical for increasing the 5-year survival rate and an effective way to improve the prognosis of pancreatic cancer, becomes an urgent and important research task. For the task of Intelligently diagnosing PC via CT at an early stage, this proposal will incorporate with a rat pancreatic tumor model to develop robust and effective imaging feature representation in weakly supervised situation, to reveal the correlation between morphological changes of PC and the pathological images, to understand the evolving mechanism of pancreatic cancer. Specifically, we will investigate the following research issues: (1) missing-input invariant methods for fusing multimodal images from dual-energy CT; (2) methods for accurate segmentation of pancreas and adjacent organs via little supervised learning; (3) methods for learning discriminative visual features of early pancreatic tumors from subtle visual signals; (4) an artificial intelligence (AI) system for early pancreatic cancer diagnosis from CT images. We will evaluate the proposed methods in this proposal via the AI system. This proposal can provide some key theoretical and technical support for early diagnosis of pancreatic cancer, and play an important role in improving the 5-year survival rate and prognosis of pancreatic cancer.
Establishing a Novel Platform Tightly Coupling Single-Molecule Experiments with Molecular Dynamics Simulations to Reveal Molecular Mechanisms behind Glycosylases’s Efficient Search for Damaged Nucleotides among an Enormous Pool of Genomic DNAs
Hong Kong Principal Investigator: Prof Xuhui Huang (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Prof Chunlai Chen (Tsinghua University)
Specific DNA recognition by DNA binding proteins plays an important role in many fundamental biological processes including DNA damage repair. DNA glycosylases are a family of DNA binding proteins that can survey genomic DNA to recognize lesion sites and initiate DNA repair. One long-lasting puzzle in the field is how DNA glycosylases can effectively and precisely locate DNA lesions in genomic DNA containing millions or even billions of base pairs. It has been hypothesized that DNA glycosylases undergo an one-dimensional diffusion on DNA to search and locate their targets with high efficiency, where they alternate between a high-speed-low-accuracy mode and a low-speed-high-accuracy mode. However, due to the limitations of current techniques, only the fast speed mode has been characterized, there is at present no work that successfully observed the slow mode.
We propose to establish a novel integrative platform that tightly couples experimental scanning Fluorescence Resonance Energy Transfer - Fluorescence Correlation Spectroscopy (FRET-FCS) with computational Markov State Models (MSMs) to elucidate the molecular mechanisms of two DNA glycosylases: Bacillus
cereus Alkylpurine glycosylase D (AlkD) and Human alkyladenine DNA glycosylase (AAG), both of which translocate on DNA to efficiently search and locate lesion sites. The novelty of our proposed platform is twofold: (1) our newly developed FRET-FCS technique can reach unprecedented temporal (at microsecond) and spatial (at sub-nanometer) resolutions compared to existing single-molecule fluorescence imaging methods; and (2) MSMs constructed from extensive Molecular Dynamic (MD) simulations provide a powerful approach to extend the timescales of MD simulations to that of experimental measured dynamics and help reveal the detailed dynamics of translocation motions.
We will employ the proposed integrative platform to elucidate the molecular mechanisms of the high-speed and the low-speed translocation modes for AlkD and AAG during their target search processes. In addition, we will reveal how the alternation between these two modes is regulated by the structural dynamics of AlkD and AAG. In our preliminary work, we have for the first time successfully characterized the slow-mode using our scanning FRET-FCS, and also elucidated its asymmetric translocation pathway via MSMs. Elucidation of such mechanisms is crucial for understanding how DNA damage is identified and repaired by DNA glycosylases to maintain genome stability.
Role and mechanism for Bora phosphorylation by Aurora A kinase in cell cycle regulation
Hong Kong Principal Investigator: Dr Randy Yat Choi Poon (The Hong Kong University of Science and Technology)
Mainland Principal Investigator: Dr Fangwei Wang (Zhejiang University)
Human cells undergo profound reorganization when they enter mitosis. A protein kinase called Aurora A (AURKA) regulates a variety of mitotic events to ensure that chromosomes are divided accurately between the daughter cells. Abnormality in the expression and activity of AURKA can cause chromosomal instability and promote tumorigenesis. One of the key actions of AURKA requires binding to a protein called Bora. Although Bora has been discovered for a number of years, large gaps in its regulation remain to be established. By leveraging the power of new genome-editing and auxin-inducible degron tools, we aim to investigate the roles of AURKA-mediated phosphorylation of Bora, and the contribution of Bora to DNA damage and mitosis. As AURKA plays a fundamental role in both normal and abnormal cell division and is intricately linked to many anticancer therapies, deciphering its precise scope of functions will have far-reaching implications in heath and disease.
A novel strategy for caries prevention: Constructing a dentotropic antimicrobial peptide with pathogen specific and mineralization balanced functions
Hong Kong Principal Investigator: Dr Hai Ming Wong (The University of Hong Kong)
Mainland Principal Investigator: Prof Quan Li Li (Anhui Medical University)
Dental caries is a bacteria-mediated, sugar-driven, multifactorial, dynamic disease that results in the imbalanced mineralization in dental hard tissue. Patients may have no symptoms in the initial stage of caries, but if allowed to progress, a large amount of the tooth structure will be lost, and there is an increased risk of infection, pain, tooth loss, or even serious systemic consequences. In spite of public health education and chemical treatments, such as antibiotics and fluoride, which have been used for decades, dental caries continues to impact nearly half of the world’s population. Conventional restoration procedures including injections and drillings cause discomfort and anxiety in patients and damage to the surrounding healthy tissues. The filling materials are not preventative against caries. In fact, given its properties which are distinct from dental tissue, the long-term performance of the restored tooth falls far short of expectations. We therefore propose to develop a four-pronged strategy to prevent and treat dental caries that aims to (1) have a high affinity for dental hard tissue, or tooth-binding, (2) prevent colonization on the tooth surface by the biofilm-forming bacteria that cause caries, (3) reduce demineralization, or the dissolving of dental hard tissue, while (4) increasing remineralization, or self-repair. This project will design and synthesize a safe and efficient multifunctional antimicrobial peptide complex for caries management. The synthesized complex can be simply applied onto the tooth surface in the form of varnish or gel, which will be an easy home care approach even for children, the elderly, and people with special needs/dental fear.
Reconstruction of Endometrium-like Tissue from Human Endometrial Stem Cells
Hong Kong Principal Investigator: Prof Ernest Hung Yu Ng (The University of Hong Kong)
Mainland Principal Investigator: Prof Songying Zhang (Zhejiang University)
The endometrium is the soil in which the embryo develops into a baby. Thinning of the endometrial lining resulting from damage to the endometrium contributes to infertility or recurrent pregnancy loss. Medical treatments using aspirin and granulocyte colony stimulating factor are insufficient to restore fertility. Therefore, major interests have focused on stem cells therapy using mesenchymal stem cells (MSCs) for treatment of damaged endometrium.
One of the challenges of stem cell transplantation is to retain the stem cells at the site of transplantation. Growing of the stem cells in biomaterials is one of the methods. Collagen is a natural component of extracellular matrix (ECM) widely used in regeneration medicine. We have shown that bone marrow MSCs (BM-MSCs) in collagen scaffold promote endometrial repair in rats. We propose that endometrial mesenchymal stem cells (eMSCs) are better than BM-MSCs in repairing damaged endometrium because they are native cells of the endometrium for the cyclical endometrial regeneration and therefore should perform better in the repair of damaged endometrium. This project hypothesizes that endogenous endometrial stem cells together with appropriate ECM proteins can better promote endometrial regeneration.
The hypothesis will be tested by growing eMSCs on collagen scaffold in the presence and absence of a molecule known to attract MSCs. In addition to eMSCs, endometrial epithelial cells will also be added to the collagen scaffold aiming to reconstitute the endometrial tissue in culture. The interaction of the eMSCs and epithelial cells in the scaffold will be studied. Besides, the natural matrix in normal endometrium will be determined, and the matrix components affecting the biological activities of eMSCs will be identified. The potential application of eMSCs in regenerative medicine for treating women with endometrial proliferation disorders is promising. Outcome from this project will be of major clinical interest and unravel new concepts in the field of endometrial stem cell biology. The findings will be useful and applicable for translational tissue engineering applications – in particular treating women with inadequate endometrium such as Asherman’s syndrome.
Roles of Secretin-like Peptides in Embryogenesis and Neurogenesis of a Cephalochordate, the Amphioxus
Hong Kong Principal Investigator: Prof Billy Kwok-Chong Chow (The University of Hong Kong)
Mainland Principal Investigator: Dr Guang Li (Xiamen University)
Secretin-like peptides and their receptors play a wide range of physiological functions that are essential to our survival. Disturbance of these ligand-receptor systems will lead to pathological conditions. During early vertebrate evolution, there was an expansion of the secretin-like peptide receptors that help driving diversification and evolvement of new species. Amphioxus, as a cephalochordate and being the transitional group between invertebrates and vertebrates, shares a similar body plan as vertebrates by having a notochord, and dorsal nerve cord. Because of its special evolutionary position, simple morphological structure and uncomplicated genome, amphioxus is ideal for evolutionary study that investigate innovative features of vertebrates, as well as to provide insights into some of the complex issues in vertebrate embryo development. In our preliminary studies, we identified all the genes that encode secretin-like peptides and receptors in the amphioxus genome and demonstrated a conserved and distinct expression pattern for these genes in specific time and position during development in the central nervous system of amphioxus embryos. We therefore hypothesize that secretin-like peptides are important to embryo development, particularly to neurogenesis, with reference to their functions in vertebrates. In this proposal, using novel gene editing techniques to knockout individual genes in the genome, we seek to study potential roles of secretin-like peptides and receptors in these processes. Data produced should provide novel insights into answering big questions regarding evolution of the nervous system from invertebrates to vertebrates. In addition, the amphioxus transgenic lines generated are valuable genetic resources for creating a new research direction that potentially be paradigm shifting for future investigation of evolution and development. These novel genetic resources can be developed into a platform for continuous and systematic studies and can be shared with the scientific community with other research groups worldwide sharing the same interests. The PI-HK (Dr. B Chow, HKU) has played a pivotal role in the past to unravel mechanisms in molecular and functional evolution of secretin-like peptides and receptors in vertebrate/invertebrate, while PI-China (Dr G Li, Xiamen University) is an expert in embryology, and was the first to generate gene knockout amphioxus models. This is a genuine collaboration of the two groups to answer important questions in vertebrate/invertebrate evolution.
Organizational Reform and Innovation in Management of Public Hospitals basing on integration of systems through Shenzhen-Hong Kong Collaboration
Hong Kong Principal Investigator: Prof Chung Mau Lo (The University of Hong Kong)
Mainland Principal Investigator: Prof Jun Wang (Renmin University of China)
The Fourth Plenary Session of the 19th Central Committee of the Communist Party of China proposed to promote the modernization of the national governance system and governance capacity. In medical and health field, the governance system and governance capacity need to be improved urgently as there is a huge gap between the high-quality development goals of public hospital reform and its reality. The University of Hong Kong – Shenzhen Hospital (HKU-SZH) has completed the fundamental reform on the management of hospital organization by the integration and innovation of the two management systems. To solve the worldwide problem of public hospital reform, explore a new transregional cooperation model between Shenzhen, Hong Kong and Macao, and provide the new theory and mechanism of public hospital management in both mainland China and Hong Kong, we take the evolution and innovation of the management system from the HKU-SZH as control group, the management system reform of Shenzhen Public General Hospitals, Hunan Provincial Public Hospital Reform Pilot Hospitals, and public hospitals under the jurisdiction of the Hong Kong Hospital Authority as the comparison group, this project develops a new method to explore the conditions and mechanisms for the successful integration of western hospital management systems and modern Chinese hospital management systems and evaluate the effects and effectiveness of the dualistic system on hospital organizational management reform and innovation. This project also tries to figure out the evaluation index system of high-quality development of modern public hospitals and the policy reform path of public hospital reform by policies simulations and policy experiments.