|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Photonic matrix multiplication lights up photonicaccelerator and beyond显示文摘Matrix computation,as a fundamental building block of information processing in science and technology,contributes most of the computational overheads in modern signal processing and artificial intelligence algorithms.Photonic accelerators are designed to accelerate specific categories of computing in the optical domain,especially matrix multiplication,to address the growing demand for computing resources and capacity.Photonic matrix multiplication has much potential to expand the domain of telecommunication,and artificial intelligence benefiting from its superior performance.Recent research in photonic matrix multiplication has flourished and may provide opportunities to develop applications that are unachievable at present by conventional electronic processors.In this review,we first introduce the methods of photonic matrix multiplication,mainly including the plane light conversion method,Mach–Zehnder interferometer method and wavelength division multiplexing method.We also summarize the developmental milestones of photonic matrix multiplication and the related applications.Then,we review their detailed advances in applications to optical signal processing and artificial neural networks in recent years.Finally,we comment on the challenges and perspectives of photonic matrix multiplication and photonic acceleration. | Hailong Zhou Jianji Dong Junwei Cheng Wenchan Dong Chaoran Huang Yichen Shen Qiming Zhang Min Gu Chao Qian Hongsheng Chen Zhichao Ruan Xinliang Zhang | 2022 | Light(Science & Applications)2022，11，2： | 4 |
| 2 | PEPT1-mediated prodrug strategy for oral delivery of peramivir显示文摘Peramivir was a novel and highly potent neuraminidase(NA) inhibitor for the treatment of influenza A and B. However, it exhibited a very low oral bioavailability(only 3%) due to the high polarity(log P of-1.4) and the low membrane permeability across the intestine. To utilize the PEPT1-mediated prodrug strategy to improve the oral absorption and develop the oral alternative, seven amino acid ester prodrugs and seven amino acid amide prodrugs have been synthesized. The permeability of these prodrugs across Caco-2 cells were screened. Peramivr-(CH_2)_2-l-Val and Peramivir-l-Ile were of the highest permeability in ester prodrugs and amide prodrugs, respectively, and then they were selected for further studies. Glycylsarcosine(gly-sar) uptake by Caco-2 could be inbihited by Peramivir-(CH_2)_2-l-Val and Peramivir-l-Ile in a concentration-dependent manner, and the IC 50 was 1.34 ± 0.31 m M and 1.78 ± 0.48 m M, respectively. The direct uptake of Peramivir-(CH_2)_2-l-Val and Peramivirl-Ile in MDCK-PEPT1 cells were significantly higher than in MDCK mock cells, and could be markedly inhibited by gly-sar. The uptake of Peramivir-(CH_2)_2-l-Val and Peramivir-l-Ile(0.01 to 50 m M) in MDCK-hPEPT1 cells conformed to Michaelis–Menten Equation. The oral bioavailability of peramivir was 65.3% and 37.3% after the oral administration of Peramivir-(CH_2)_2-l-Val and Peramivir-l-Ile to rats, respectively. The oral absorption and bioactivation of Peramivir-(CH_2)_2-l-Val was rapid and extensive, and no Peramivir-(CH_2)_2-l-Val was found in plasma. Because the amide bond was relatively stable, Peramivir-l-Ile could not be totally converted to the parent drug in vivo. Peramivir-(CH_2)_2-l-Val with good oral profiles and rapid bioactivation might be a promising prodrug for the further clinic development. The present study also corroborated the idea that the PEPT1-mediated prodrug approach has enormous promise for improving the oral absorption of poorly absorbed drug. | Yongbing Sun Wei Gan Mingdao Lei Wei Jiang Meng Cheng Junwei He Qi Sun Wan Liu Lvjiang Hu Yi Jin | 2018 | Asian Journal of Pharmaceutical Sciences2018，13，6： | 4 |
| 3 | Self-heating Probe Instrument and Method for Measuring High Temperature Melting Volume Change Rate of Material显示文摘The castings defects are affected by the melting volume change rate of material. The change rate has an important effect on running safety of the high temperature thermal storage chamber, too. But the characteristics of existing measuring installations are complex structure, troublesome operation and low precision. In order to measure the melting volume change rate of material accurately and conveniently, a self-designed measuring instrument, self-heating probe instrument, and measuring method are described. Temperature in heating cavity is controlled by PID temperature controller; melting volume change rate υ and molten density are calculated based on the melt volume which is measured by the instrument. Positive and negative υ represent expansion and shrinkage of the sample volume after melting, respectively. Taking eutectic LiF+CaF2 for example, its melting volume change rate and melting density at 1 123 K are -20.6% and 2 651 kg/m-3 measured by this instrument, which is only 0.71% smaller than literature value. Density and melting volume change rate of industry pure aluminum at 973 K and analysis pure NaCl at 1 123 K are detected by the instrument too. The measure results are agreed with report values. Measuring error sources are analyzed and several improving measures are proposed. In theory, the measuring errors of the change rate and molten density which are measured by the self-designed instrument is nearly 1/20-1/50 of that measured by the refitted mandril thermal expansion instrument. The self-designed instrument and method have the advantages of simple structure, being easy to operate, extensive applicability for material, relatively high accuracy, and most importantly, temperature and sample vapor pressure have little effect on the measurement accuracy. The presented instrument and method solve the problems of complicated structure and procedures, and large measuring errors for the samples with high vapor pressure by existing installations. | WANG Junwei WANG Zhiping LU Yang CHENG Bo | 2013 | Chinese Journal of Mechanical Engineering2013，26，2： | 2 |
| 4 | Hybrid Polymer Membrane Functionalized PBO Fibers/Cyanate Esters Wave‑Transparent Laminated Composites显示文摘Hybrid polymer membrane(TA-APTES),synthesized by tannic acid(TA)and aminopropyl trethoxysilane(APTES)based on the Schif’s base and Michael addition reaction,is deposited on the surface of poly(p-phenylene-2,6-benzobisoxazole)(PBO)fbers,and then grafted with epoxy-terminated polysesquisiloxane(POSS)to obtain POSS-g-PBO@TA-APTES fbers.The POSS-g-PBO@TA-APTES fbers reinforced bisphenol A dicyanate ester(BADCy)resins(POSS-g-PBO@TAAPTES fbers/BADCy)wave-transparent laminated composites are prepared.The interlaminar shear strength and fexural strength of POSS-g-PBO@TA-APTES fbers/BADCy composites are respectively enhanced from 36.7 and 587.4 MPa to 42.8 and 645.8 MPa,increased by 16.6%and 9.9%compared with those of PBO fbers/BADCy composites.At 1 MHz,the corresponding dielectric constant and dielectric loss are reduced to 2.85 and 0.0047,respectively,lower than those of PBO fbers/BADCy(3.06 and 0.006)composites.Meanwhile,the simulated wave transmittance rate of POSS-g-PBO@TA-APTES fbers/BADCy composites with the thicknesses of 1.5–3.5 mm is higher than 86.2%at 0.3–40 GHz.The volume resistivity and breakdown strength of POSS-g-PBO@TA-APTES fbers/BADCy composites are 2.8×10^(15)Ω·cm and 19.80 kV/mm,higher than PBO fbers/BADCy composites(2.2×10^(15)Ω·cm and 17.69 kV/mm),respectively.And the corresponding heat resistant index is 221.5℃,lower than PBO fbers/BADCy composites(229.6℃). | Zheng Liu Xiaoli Fan Lei Cheng Junliang Zhang Lin Tang Yusheng Tang Jie Kong Junwei Gu | 2022 | Advanced Fiber Materials2022，4，3： | 2 |
| 5 | Class attention network for image recognition显示文摘Visual attention has become a popular and widely used component for image recognition.Although various attention-based methods have been proposed and achieved relatively competitive results,it is observed that the semantic features of each class are likely to entangle with each other,and few studies focus on explicitly extracting category-aware features so far.To address this issue,this paper presents an attention-based image recognition method by using class-specific dictionary learning to disentangle the neural network’s outputs into class-dependent features,thus boosting their discrimination abilities.Specifically,we develop a class attention network(CANet)via integrating a simple yet effective class-specific attention encoding(CAE)module on the top of convolutional layers.Given the feature maps of the convolutional neural networks(CNNs),the CAE module learns a class-specific dictionary,which is leveraged to encode attention maps for each category.Then these attention maps are multiplied by the input features for classwise adaptive feature refinement.Extensive experiments on the PASCAL VOC 2007,PASCAL VOC 2012,MS COCO,and CUB-200-2011 datasets demonstrate the fabulous performance of our method on multiple visual recognition tasks,including multi-label image classification and fine-grained visual classification.In addition,the visualization results testify that CNNs can explicitly learn class-wise feature representations by introducing class-specific dictionary learning. | Gong CHENG Pujian LAI Decheng GAO Junwei HAN | 2023 | Science China(Information Sciences)2023，66，3： | 2 |
| 6 | Space Charge Suppression of Polyethylene Induced by Blending with Ethylene-butyl Acrylate Copolymer显示文摘High-voltage direct current(HVDC)power cables have been paid more attention due to the big issue concerning the main insulation materials.The residues in the industrial production process will cause space charge generation and accumulation in the materials,which has become one bottleneck to limit the development of power cables up to higher voltage levels.In this paper,the LDPE matrix was modified by three types of ethylene-butyl acrylate(EBA)copolymers(1.0 wt%)with different polarities(BA content)via melt blending to optimize the space charge behavior of the LDPE insulation.The micromorphology and structure of the blends are examined by polarized light microscope and differential scanning calorimetry.The space charge distribution is tested by the pulsed electro-acoustic method.The trap level is studied by the thermally stimulated current.The results show that EBA introduces more deep traps and decreases the shallow traps.The medium-polar EBA(16%BA content)can effectively suppress charge accumula-tion,and have the same suppressed effect on XLPE/EBA blends.This study will provide important insights into the design and development of advanced HVDC power cable applications. | Jintao Zhai Weikang Li Junwei Zha Qi Cheng Xingming Bian Zhimin Dang | 2020 | CSEE Journal of Power and Energy Systems2020，6，1： | 1 |
| 7 | Task-wise attention guided part complementary learning for few-shot image classification显示文摘A general framework to tackle the problem of few-shot learning is meta-learning,which aims to train a well-generalized meta-learner(or backbone network)to learn a base-learner for each future task with small training data.Although a lot of work has produced relatively good results,there are still some challenges for few-shot image classification.First,meta-learning is a learning problem over a collection of tasks and the meta-learner is usually shared among all tasks.To achieve image classification of novel classes in different tasks,it is needed to learn a base-learner for each task.Under the circumstances,how to make the base-learner specialized,and thus respond to different inputs in an extremely task-wise manner for different tasks is a big challenge at present.Second,classification network usually inclines to identify local regions from the most discriminative object parts rather than the whole objects for recognition,thereby resulting in incomplete feature representations.To address the first challenge,we propose a task-wise attention(TWA)module to guide the base-learner to extract task-specific image features.To address the second challenge,under the guidance of TWA,we propose a part complementary learning(PCL)module to extract and fuse the features of multiple complementary parts of target objects,and thus we can obtain more specific and complete information.In addition,the proposed TWA module and PCL module can be embedded into a unified network for end-to-end training.Extensive experiments on two commonly-used benchmark datasets and comparison with state-of-the-art methods demonstrate the effectiveness of our proposed method. | Gong CHENG Ruimin LI Chunbo LANG Junwei HAN | 2021 | Science China(Information Sciences)2021，64，2： | 1 |
| 8 | Hybrid Recommendation Based on Graph Embedding显示文摘In recent years,online reservation systems of country hotel have become increasingly popular in rural areas.How to accurately recommend the houses of country hotel to the users is an urgent problem to be solved.Aiming at the problem of cold start and data sparseness in recommendation,a Hybrid Recommendation method based on Graph Embedding(HRGE)is proposed.First,three types of network are built,including user-user network based on user tag,househouse network based on house tag,and user-user network based on user behavior.Then,by using the method of graph embedding,three types of network are respectively embedded into low-dimensional vectors to obtain the characterization vectors of nodes.Finally,these characterization vectors are used to make a hybrid recommendation.The datasets in this paper are derived from the Country Hotel Reservation System in Guizhou Province.The experimental results show that,compared with traditional recommendation algorithms,the comprehensive evaluation index(F1)of the HRGE is improved by 20% and the Mean Average Precision(MAP)is increased by 11%. | Cheng Zeng Haifeng Zhang Junwei Ren Chaodong Wen Peng He | 2021 | China Communications2021，18，11： | 1 |
| 9 | A small microring array that performs large complex-valued matrix-vector multiplication显示文摘As an important computing operation,photonic matrix-vector multiplication is widely used in photonic neutral networks and signal processing.However,conventional incoherent matrix-vector multiplication focuses on real-valued operations,which cannot work well in complex-valued neural networks and discrete Fourier transform.In this paper,we propose a systematic solution to extend the matrix computation of microring arrays from the real-valued field to the complex-valued field,and from small-scale(i.e.,4×4)to large-scale matrix computation(i.e.,16×16).Combining matrix decomposition and matrix partition,our photonic complex matrix-vector multiplier chip can support arbitrary large-scale and complex-valued matrix computation.We further demonstrate Walsh-Hardmard transform,discrete cosine transform,discrete Fourier transform,and image convolutional processing.Our scheme provides a path towards breaking the limits of complex-valued computing accelerator in conventional incoherent optical architecture.More importantly,our results reveal that an integrated photonic platform is of huge potential for large-scale,complex-valued,artificial intelligence computing and signal processing. | Junwei Cheng Yuhe Zhao Wenkai Zhang Hailong Zhou Dongmei Huang Qing Zhu Yuhao Guo Bo Xu Jianji Dong Xinliang Zhang | 2022 | Frontiers of Optoelectronics2022，15，2： | 1 |
| 10 | Analysis of the DC bias phenomenon by the Harmonic Balance Finite-Element meth od显示文摘 | Xiaojun Zhao Junwei Lu Lin Li Zhiguang Cheng | 2011 | IEEE Transactions on Power Delivery2011，26，1： | 1 |
| 11 | Conservation laws of some lattice equations显示文摘 | Junwei CHENG Dajun ZHANG | 2013 | Frontiers of Mathematics in China2013，8，5： | 1 |
| 12 | Analysis of DC bias phenomenon by the harmonic balance finite-element method显示文摘 | Zhao Xiaojun Lu Junwei Li Lin Cheng Zhiguang | 2011 | IEEE Transactions on Power Delivery2011，26，1： | 1 |
| 13 | Integrated photonic convolution acceleration core for wearable devices显示文摘With the advancement of deep learning and neural networks,the computational demands for applications in wearable devices have grown exponentially.However,wearable devices also have strict requirements for long battery life,low power consumption,and compact size.In this work,we propose a scalable optoelectronic computing system based on an integrated optical convolution acceleration core.This system enables high-precision computation at the speed of light,achieving 7-bit accuracy while maintaining extremely low power consumption.It also demonstrates peak throughput of 3.2 TOPS(tera operations per second)in parallel processing.We have successfully demonstrated image convolution and the typical application of an interactive first-person perspective gesture recognition application based on depth information.The system achieves a comparable recognition accuracy to traditional electronic computation in all blind tests. | Baiheng Zhao Junwei Cheng Bo Wu Dingshan Gao Hailong Zhou Jianji Dong | 2023 | Opto-Electronic Science2023，2，12： | 1 |
| 14 | Fewer is more:efficient object detection in large aerial images显示文摘Current mainstream object detection methods for large aerial images usually divide large images into patches and then exhaustively detect the objects of interest on all patches,no matter whether there exist objects or not.This paradigm,although effective,is inefficient because the detectors have to go through all patches,severely hindering the inference speed.This paper presents an objectness activation network(OAN) to help detectors focus on fewer patches but achieve more efficient inference and more accurate results,enabling a simple and effective solution to object detection in large images.In brief,OAN is a light fully-convolutional network for judging whether each patch contains objects or not,which can be easily integrated into many object detectors and jointly trained with them end-to-end.We extensively evaluate our OAN with five advanced detectors.Using OAN,all five detectors acquire more than 30.0% speed-up on three large-scale aerial image datasets,meanwhile with consistent accuracy improvements.On extremely large Gaofen-2 images(29200 × 27620 pixels),our OAN improves the detection speed by 70.5%.Moreover,we extend our OAN to driving-scene object detection and 4K video object detection,boosting the detection speed by 112.1% and 75.0%,respectively,without sacrificing the accuracy. | Xingxing XIE Gong CHENG Qingyang LI Shicheng MIAO Ke LI Junwei HAN | 2024 | Science China(Information Sciences)2024，67，1： | 1 |
| 15 | Selective substitution induced anomalous phonon stiffening within quasi-one-dimensional P–P chains in SiP2显示文摘Light-matter interactions in low-dimensional quantum-confined structures can dominate the optical properties of the materials and lead to optoelectronic applications.In anisotropic layered silicon diphosphide(SiP2)crystal,the embedded quasi-onedimensional(1D)phosphorus–phosphorus(P–P)chains directly result in an unconventional quasi-1D excitonic state,and a special phonon mode vibrating along the P–P chains,establishing a unique 1D quantum-confined system.Alloying SiP2 with the homologous element serves as an effective way to study the properties of these excitons and phonons associated with the quasi-1D P–P chains,as well as the strong interaction between these quasiparticles.However,the experimental observation and the related optical spectral understanding of SiP2 with isoelectronic dopants remain elusive.Herein,with the photoluminescence and Raman spectroscopy measurements,we demonstrate the redshift of the confined excitonic peak and the stiffening of the phonon vibration mode■of a series of Si(P1−xAsx)2 alloys with increasing arsenic(As)compositions.This anomalous stiffening of■is attributed to the selective substitution of As atoms for P atoms within the P–P chains,which is confirmed via our scanning transmission electron microscopy investigation.Such optical spectra evolutions with selective substitution pave a new way to understand the 1D quantum confinement in semiconductors,offering opportunities to explore quasi-1D characteristics in SiP2 and the resulting photonic device application. | Xueting Dai Feng Qin Caiyu Qiu Ling Zhou Junwei Huang Fanghua Cheng Xiangyu Bi Caorong Zhang Zeya Li Ming Tang Shengqiang Wu Xiaoxu Zhao Yangfan Lu Huiyang Gou Hongtao Yuan | 2023 | Nano Research2023，16，1： | 0 |
| 16 | Strain engineering of anisotropic light-matter interactions in onedimensional P-P chain of SiP_(2)显示文摘Strain engineering can serve as a powerful technique for modulating the exotic properties arising from the atomic structure of materials.Examples have been demonstrated that one-dimensional(1D)structure can serve as a great platform for modulating electronic band structure and phonon dispersion via strain control.Particularly,in a van der Waals material silicon diphosphide(SiP_(2)),quasi-1D zigzag phosphorus–phosphorus(P–P)chains are embedded inside the crystal structure,and can show unique phonon vibration modes and realize quasi-1D excitons.Manipulating those optical properties by the atom displacements via strain engineering is of great interest in understanding underlying mechanism of such P–P chains,however,which remains elusive.Herein,we demonstrate the strain engineering of Raman and photoluminescence(PL)spectra in quasi-1D P–P chains and resulting in anisotropic manipulation in SiP_(2).We find that the phonon frequencies of SiP_(2)in Raman spectra linearly evolve with a uniaxial strain along/perpendicular to the quasi-1D P–P chain directions.Interestingly,by applying tensile strain along the P–P chains,the band gap energy of strained SiP_(2)can significantly decrease with a tunable value of~55 meV.Based on arsenic(As)element doping into SiP_(2),the strain-induced redshifts of phonon frequencies decrease,indicating the stiffening of the phonon vibration with the increased arsenic doping level.Such results provide an opportunity for strain engineering of the light–matter interactions in the quasi-1D P–P chains of SiP_(2)crystal for potential optical applications. | Fanghua Cheng Junwei Huang Feng Qin Ling Zhou Xueting Dai Xiangyu Bi Caorong Zhang Zeya Li Ming Tang Caiyu Qiu Yangfan Lu Huiyang Gou Hongtao Yuan | 2022 | Nano Research2022，15，8： | 0 |
| 17 | 6-13 Summary of Ion Source Group in 2015显示文摘Ion source as one very unique and critical part of ion accelerator needs consistent research and development work to improve the performance and to meet the future accelerator needs as well. The ion source group at IMP is taking the responsibility of high performance intense beam ion source development, operation and exploration. | Sun Liangting Feng Yucheng Lu Wang Qian Cheng Zhang Wenhui Wu Qi Yang Yao Guo Junwei Zhao Huanyu Cao Yun Zhang Xuezhen Zhao Hongwei | 2015 | IMP & HIRFL Annual Report2015，，1： | 0 |
| 18 | 6-15 Status Report of On-line Ion Sources in 2015显示文摘In 2015, the service time of the two on-line ion sources of HIRFL-CSR accelerator facility, SECRAL and LECR3, is 3 335.5 and 3 168.5 h, respectively, amouting to 6 504 h. Except 26Mg7+, which was required by the accelerator but failed to be produced due the oxidation of the material, 20 kinds of ion beams have been delivered successfull.The failure time is 49 h this year, mainly attributed to the breakdown of the cryostat system. | Feng Yucheng Lu Wang Zhang Wenhui Ma Hongyi Yang Yao Fang Xing Guo Junwei Ma Baohua Wang Hui Qian Cheng Zhang Junjie Li Jibo Jin Qianyu sheng Sifan Liu Pengyan Li Xixia Zhao Huanyu Zhang Xuezhen Sun Liangting | 2015 | IMP & HIRFL Annual Report2015，，1： | 0 |
| 19 | On-chip photonic spatial-temporal descrambler显示文摘As an indispensable part to compensate for the signal crosstalk in fiber communication systems,conventional digital multi-input multi-output(MIMO)signal processor is facing the challenges of high com-putational complexity,high power consumption and relatively low processing speed.The optical MIMOenables the best use of light and has been proposed to remedy this limitation.However,the currently existing optical MIMO methods are all restricted to the spatial di-mension,while the temporal dimension is neglected.Here,an on-chip spatial-temporal descrambler with four channels were devised and its MIMO functions were experimentally verified simultaneously in both spatial and temporal dimensions.The spatial crosstalk of single-channel descrambler and four-channel descrambler is respectively less than-21 dB and-18 dB,and the time delay is simultaneously com-pensated successfully.Moreover,a more universal model extended to mode-dependent loss and gain(MDL)compensation was further de-veloped,which is capable of being cascaded for the real optical trans-mission system.The first attempt at photonic spatial-temporal de-scrambler enriched the varieties of optical MIMO,and the proposed scheme provided a new opportunity for all-optical MIMO signal pro-cessing. | Wenkai Zhang Xueyi Jiang Wentao Gu Junwei Cheng Hailong Zhou Jianji Dong Dongmei Huang Xinliang Zhang | 2023 | Chip2023，2，2： | 0 |
| 20 | 机械力诱导MHC-Ⅰ构象变化增强TCR抗原识别及T细胞活化显示文摘CD8+T细胞主要通过T细胞表面受体(T cell recep tor,TCR)识别并与Ⅰ型主要组织性复合物提呈的抗原(pMHC-Ⅰ)相互作用[1]。TCR对刺激型抗原的识别在CD8+T细胞毒性和适应性免疫中发挥着关键作用。许多证据表明机械力可以延长TCR与刺激性pMHC作用的键合时间(bond lifetime)形成抗原特异性逆锁键(catch bond)[2],并且这种逆锁键对抗原识别非常重要。然而,机械力调控TCR抗原识别的具体结构机制仍不清楚。通过分子动力学模拟、单分子生物膜力学探针、磁镊、T细胞活化实验和动物模型对此问题展开了系统的研究[4]。发现作用在TCR-pMHC-Ⅰ复合体上的拉力可以作用在TCR-pMHC-Ⅰ复合体上的拉力可以打破MHC-I分子内部α1-α2和β2结构域间的相互作用,导致α1-α2结构域旋转并发生构象变化。力诱导的MHC-I构象变化可以进一步别构地调节TCR与刺激性抗原肽及α1-α2结构域的构象及相互作用,诱导产生新的氢键,增强TCR-pMHC-Ⅰ之间的键合时间,但并不能增强TCR与抑制性pMHC-Ⅰ之间的作用。当用点突变阻断这些新形成的氢键,或者α1-α2和β2结构域被二硫键锁住时,最佳力诱导的TCR-pMHC-Ⅰ作用的键合时间明显缩短并且T细胞的活化受到抑制。另外,在人TCR和HLA-A2相互作用中发现了类似机制,并且与肿瘤相关的HLA-A2点突变[3]可以通过限制HLA-A2α1--α2和β2结构域之间的构象打开减弱TCR对肿瘤抗原的识别及T细胞的功能。研究结果表明,机械力诱导的MHC-I构象变化对TCR抗原识别和T细胞活化非常重要,进一步地阐明了机械力调控TCR抗原识别机制,为临床肿瘤的免疫治疗和药物设计提供了新思路和新靶点。 | Peng Wu Tongtong Zhang Baoyu Liu Panyu Fei Lei Cui Rui Qin Huaying Zhu Danmei Yao Ryan Martinez Wei Hu Chenyi An Yong Zhang Junwei Liu Weiwei Yin Jie Sun Chun Zhou Xun Zeng Jianan Wang Brian Evavold Cheng Zhu Jizhong Lou Wei Chen | 2019 | 医用生物力学2019，34，A01： | 0 |
您所在的机构暂未开通此部分的功能，请联系管理员