State Key Laboratory of Membrane Biology

State Key Laboratory of Membrane Biology


Time filter

Source Type

Su F.,Beihang University | Su F.,State Key Laboratory of Membrane Biology | Su F.,Peking University | Yuan P.,Beihang University | And 4 more authors.
Protein and Cell | Year: 2016

Artificial neural networks (ANNs) are powerful computational tools that are designed to replicate the human brain and adopted to solve a variety of problems in many different fields. Fault tolerance (FT), an important property of ANNs, ensures their reliability when significant portions of a network are lost. In this paper, a fault/noise injection-based (FIB) genetic algorithm (GA) is proposed to construct fault-tolerant ANNs. The FT performance of an FIB-GA was compared with that of a common genetic algorithm, the back-propagation algorithm, and the modification of weights algorithm. The FIB-GA showed a slower fitting speed when solving the exclusive OR (XOR) problem and the overlapping classification problem, but it significantly reduced the errors in cases of single or multiple faults in ANN weights or nodes. Further analysis revealed that the fit weights showed no correlation with the fitting errors in the ANNs constructed with the FIB-GA, suggesting a relatively even distribution of the various fitting parameters. In contrast, the output weights in the training of ANNs implemented with the use the other three algorithms demonstrated a positive correlation with the errors. Our findings therefore indicate that a combination of the fault/noise injection-based method and a GA is capable of introducing FT to ANNs and imply that the distributed ANNs demonstrate superior FT performance. © 2016 The Author(s)


News Article | October 23, 2015
Site: www.nature.com

In this Letter, author Yong-Bin Yan was incorrectly associated with affiliation number 5 (Department of Ophthalmology, Xijing Hospital) instead of affiliation number 4 (State Key Laboratory of Membrane Biology, School of Life Sciences, Tsinghua University, Beijing 100084, China). Also, an additional affiliation has been added to author Kang Zhang (number 15; Institute of Molecular Medicine, Peking University, Beijing 100871, China), and affiliation number 3 has changed from ‘Department of Ophthalmology and Biomaterials and Tissue Engineering Center’ to ‘Shiley Eye Institute and Biomaterials and Tissue Engineering Center’. These have all been corrected in the online versions of the paper.


PubMed | Beihang University and State Key Laboratory of Membrane Biology
Type: Journal Article | Journal: Protein & cell | Year: 2016

Artificial neural networks (ANNs) are powerful computational tools that are designed to replicate the human brain and adopted to solve a variety of problems in many different fields. Fault tolerance (FT), an important property of ANNs, ensures their reliability when significant portions of a network are lost. In this paper, a fault/noise injection-based (FIB) genetic algorithm (GA) is proposed to construct fault-tolerant ANNs. The FT performance of an FIB-GA was compared with that of a common genetic algorithm, the back-propagation algorithm, and the modification of weights algorithm. The FIB-GA showed a slower fitting speed when solving the exclusive OR (XOR) problem and the overlapping classification problem, but it significantly reduced the errors in cases of single or multiple faults in ANN weights or nodes. Further analysis revealed that the fit weights showed no correlation with the fitting errors in the ANNs constructed with the FIB-GA, suggesting a relatively even distribution of the various fitting parameters. In contrast, the output weights in the training of ANNs implemented with the use the other three algorithms demonstrated a positive correlation with the errors. Our findings therefore indicate that a combination of the fault/noise injection-based method and a GA is capable of introducing FT to ANNs and imply that the distributed ANNs demonstrate superior FT performance.


Hua R.,State Key Laboratory of Membrane Biology | Wei M.P.,State Key Laboratory of Membrane Biology | Zhang C.,State Key Laboratory of Membrane Biology
Science China Life Sciences | Year: 2015

Autism spectrum disorders (ASD) are a pervasive neurodevelopmental disease characterized by deficits in social interaction and nonverbal communication, as well as restricted interests and stereotypical behavior. Genetic changes/heritability is one of the major contributing factors, and hundreds to thousands of causative and susceptible genes, copy number variants (CNVs), linkage regions, and microRNAs have been associated with ASD which clearly indicates that ASD is a complex genetic disorder. Here, we will briefly summarize some of the high-confidence genetic changes in ASD and their possible roles in their pathogenesis. © 2015, The Author(s).


PubMed | Johns Hopkins University, Beijing Advanced Innovation Center for Structural Biology and State Key Laboratory of Membrane Biology
Type: Journal Article | Journal: Cell research | Year: 2016

Sterol regulatory element-binding protein (SREBP) transcription factors are master regulators of cellular lipid homeostasis in mammals and oxygen-responsive regulators of hypoxic adaptation in fungi. SREBP C-terminus binds to the WD40 domain of SREBP cleavage-activating protein (SCAP), which confers sterol regulation by controlling the ER-to-Golgi transport of the SREBP-SCAP complex and access to the activating proteases in the Golgi. Here, we biochemically and structurally show that the carboxyl terminal domains (CTD) of Sre1 and Scp1, the fission yeast SREBP and SCAP, form a functional 4:4 oligomer and Sre1-CTD forms a dimer of dimers. The crystal structure of Sre1-CTD at 3.5 and cryo-EM structure of the complex at 5.4 together with in vitro biochemical evidence elucidate three distinct regions in Sre1-CTD required for Scp1 binding, Sre1-CTD dimerization and tetrameric formation. Finally, these structurally identified domains are validated in a cellular context, demonstrating that the proper 4:4 oligomeric complex formation is required for Sre1 activation.

Loading State Key Laboratory of Membrane Biology collaborators
Loading State Key Laboratory of Membrane Biology collaborators