DSpace Community:
http://hdl.handle.net/10722/1033
2024-03-28T04:01:29ZOn Reinforcement Learning in Stabilizability of Probabilistic Boolean Control Networks
http://hdl.handle.net/10722/341761
Title: On Reinforcement Learning in Stabilizability of Probabilistic Boolean Control Networks
Authors: Lin, Lin; Lam, James; Shi, Peng; Ng, Michael K; Lam, Hak-Keung
Abstract: <p>This paper examines the stabilizability probability of probabilistic Boolean control networks (PBCNs) by utilizing reinforcement learning. The bounds of the reachability probability are formulated based on the algebraic state space representation (ASSR) of PBCNs, which leads to the stabilizability criterion. Furthermore, the equivalence between the stabilizability probability and the optimal state-value function in the form of an iteration equation is proved. The Q-learning technique is implemented to estimate the stabilizability probability and obtain the corresponding optimal control sequences. Theoretical results are demonstrated through an apoptosis network to elaborate on the findings.<br></p>2023-11-28T00:00:00ZMetallo-sideromycin as a dual functional complex for combating antimicrobial resistance
http://hdl.handle.net/10722/341724
Title: Metallo-sideromycin as a dual functional complex for combating antimicrobial resistance
Authors: Wang, Chenyuan; Xia, Yushan; Wang, Runming; Li, Jingru; Chan, Chun-Lung; Kao, Richard Yi-Tsun; Toy, Patrick H; Ho, Pak-Leung; Li, Hongyan; Sun, Hongzhe
Abstract: <p>Chromium(III) is extensively used as a supplement for muscle development and the treatment of diabetes mellitus. However, its mode of action, essentiality, and physiological/pharmacological effects have been a subject of scientific debate for over half a century owing to the failure in identifying the molecular targets of Cr(III). Herein, by integrating fluorescence imaging with a proteomic approach, we visualized the Cr(III) proteome being mainly localized in the mitochondria, and subsequently identified and validated eight Cr(III)-binding proteins, which are predominately associated with ATP synthesis. We show that Cr(III) binds to ATP synthase at its beta subunit via the catalytic residues of Thr213/Glu242 and the nucleotide in the active site. Such a binding suppresses ATP synthase activity, leading to the activation of AMPK, improving glucose metabolism, and rescuing mitochondria from hyperglycaemia-induced fragmentation. The mode of action of Cr(III) in cells also holds true in type II diabetic male mice. Through this study, we resolve the long-standing question of how Cr(III) ameliorates hyperglycaemia stress at the molecular level, opening a new horizon for further exploration of the pharmacological effects of Cr(III).<br></p>2023-09-01T00:00:00ZSPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuning
http://hdl.handle.net/10722/341723
Title: SPTNet: An Efficient Alternative Framework for Generalized Category Discovery with Spatial Prompt Tuning
Authors: Wang, Hongjun; Vaze, Sagar; Han, Kai
Abstract: <p>Generalized Category Discovery (GCD) aims to classify unlabelled images from both ‘seen’ and ‘unseen’ classes by transferring knowledge from a set of labelled ‘seen’ class images. A key theme in existing GCD approaches is adapting largescale pretrained models for the GCD task. An alternate perspective, however, is to adapt the data representation itself for better alignment with the pretrained model. As such, in this paper, we introduce a two-stage adaptation approach termed SPTNet, which iteratively optimizes model parameters (i.e., model-finetuning) and data parameters (i.e., prompt learning). Furthermore, we propose a novel spatial prompt tuning method (SPT) which considers the spatial property of image data, enabling the method to better focus on object parts, which can transfer between seen and unseen classes. We thoroughly evaluate our SPTNet on standard benchmarks and demonstrate that our method outperforms existing GCD methods. Notably, we find our method achieving an average accuracy of 61.4% on the SSB, surpassing prior state-of-the-art methods by approximately 10%. The improvement is particularly remarkable as our method yields extra parameters amounting to only 0.039% of those in the backbone architecture.</p>2024-05-17T00:00:00ZGoldfish adiponectin: (I) molecular cloning, tissue distribution, recombinant protein expression, and novel function as a satiety factor in fish model
http://hdl.handle.net/10722/341713
Title: Goldfish adiponectin: (I) molecular cloning, tissue distribution, recombinant protein expression, and novel function as a satiety factor in fish model
Authors: Zheng, Yunhua; Ye, Cheng; He, Mulan; Ko, Wendy K W; Chan, Ying Wai; Wong, Anderson O L
Abstract: <p>Adiponectin (AdipoQ) is an adipokine involved in glucose homeostasis and lipid metabolism. In mammals, its role in appetite control is highly controversial. To shed light on the comparative aspects of AdipoQ in lower vertebrates, goldfish was used as a model to study feeding regulation by AdipoQ in fish species. As a first step, goldfish AdipoQ was cloned and found to be ubiquitously expressed at the tissue level. Using sequence alignment, protein modeling, phylogenetic analysis and comparative synteny, goldfish AdipoQ was shown to be evolutionarily related to its fish counterparts and structurally comparable with AdipoQ in higher vertebrates. In our study, recombinant goldfish AdipoQ was expressed in <em>E. coli</em>, purified by IMAC, and confirmed to be bioactive via activation of AdipoQ receptors expressed in HepG2 cells. Feeding in goldfish revealed that plasma levels of AdipoQ and its transcript expression in the liver and brain areas involved in appetite control including the telencephalon, optic tectum, and hypothalamus could be elevated by food intake. In parallel studies, IP and ICV injection of recombinant goldfish AdipoQ in goldfish was effective in reducing foraging behaviors and food consumption. Meanwhile, transcript expression of orexigenic factors (NPY, AgRP, orexin, and apelin) was suppressed with parallel rises in anorexigenic factors (POMC, CART, CCK, and MCH) in the telencephalon, optic tectum and/or hypothalamus. In these brain areas, transcript signals for leptin receptor were upregulated with concurrent drops in the NPY receptor and ghrelin receptors. In the experiment with IP injection of AdipoQ, transcript expression of leptin was also elevated with a parallel drop in ghrelin mRNA in the liver. These findings suggest that AdipoQ can act as a novel satiety factor in goldfish. In this case, AdipoQ signals (both central and peripheral) can be induced by feeding and act within the brain to inhibit feeding behaviors and food intake via differential regulation of orexigenic/anorexigenic factors and their receptors. The feeding inhibition observed may also involve the hepatic action of AdipoQ by modulation of feeding regulators expressed in the liver.<br></p>2023-10-30T00:00:00Z