期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
Isolation, Identification and Pathogenicity Analysis of Streptococcus suis Type 2 被引量:4
1
作者 zicheng ma Yan LI +6 位作者 Jinyuan GU Tao PENG Zhaohu LIU Hongyu WANG Fanliang MENG Fangkun WANG Sidang LIU 《Agricultural Biotechnology》 CAS 2019年第4期64-68,共5页
[Objectives]This study aimed to investigate the pathogenicity,growth characteristics and drug resistance of Streptococcus suis type 2.[Methods]Bacterial isolation and identification,biochemical experiments,determinati... [Objectives]This study aimed to investigate the pathogenicity,growth characteristics and drug resistance of Streptococcus suis type 2.[Methods]Bacterial isolation and identification,biochemical experiments,determination of growth curve and correlation curve between OD 600 values and viable counts,drug susceptibility tests,pathogenicity analysis,and histopathological observations were carried out.[Results]The Streptococcus strain isolated from infected pigs was identified as Streptococcus suis type 2,which was named TA01 strain.TA01 strain reached the growth peak at 6-8 h post-incubation,and viable counts gradually declined after 8 h of incubation.The correlation equation between OD 600 values and viable counts is y=24.659 x-1.076 1,R^2=0.996 7.TA01 strain was sensitive to penicillin,erythromycin,florfenicol and oxacillin,and resistant to ciprofloxacin,polymyxin B and clindamycin.According to the results of pathogenicity analysis,all the mice in 3.6×10^9 cfu/mouse group died within 48,and these dead mice exhibited acute pyaemia septica.Based on the Reed-Muench formula,it was calculated that LD 50 of TA01 strain was 1.137×10^8 cfu/mouse.Pathological examination showed obvious blue-stained bacteria clusters,accompanied by neutrophil infiltration.[Conclusions]TA01 strain was a virulent strain of Streptococcus suis type 2.Compared with Streptococcus strains which were isolated and reported in China,TA01 strain exhibited strong virulence and rapid proliferation. 展开更多
关键词 STREPTOCOCCUS SUIS TYPE 2 ISOLATION and IDENTIFICATION Growth curve Drug sensitivity test PATHOGENICITY LD 50 determination
在线阅读 下载PDF
基于物理启发式的深度学习方法对蛋白质三维结构的修复与扩展采样
2
作者 陈振宇 林潇涵 +3 位作者 李彦衡 马子程 张骏 高毅勤 《科学通报》 北大核心 2025年第34期5911-5922,共12页
AlphaFold2作为首个达到实验精度的深度学习蛋白质结构预测模型,已成为结构生物学领域的核心工具.与此同时,由此衍生的深度学习方法在蛋白质构象系综的高效采样及蛋白序列设计、配体扩散生成等一系列从蛋白质结构出发的下游结构生物学... AlphaFold2作为首个达到实验精度的深度学习蛋白质结构预测模型,已成为结构生物学领域的核心工具.与此同时,由此衍生的深度学习方法在蛋白质构象系综的高效采样及蛋白序列设计、配体扩散生成等一系列从蛋白质结构出发的下游结构生物学方向中也取得了长足进步.在多构象采样方面,传统分子动力学和蒙特卡罗方法 因计算成本和采样效率限制,难以在较大体系中广泛应用,现有生成式深度学习方法能快速得到蛋白质多构象结构,但难以加入适当约束以控制构象采样的范围.此外,蛋白质初始结构的鲁棒性对基于蛋白质结构的下游模型表现和分子动力学模拟的效率至关重要,但传统的蛋白质结构修复工具在修复蛋白结构缺失区域方面效果有限.针对这些挑战,本研究提出FoldCopilot套件,基于AlphaFold2将目标序列和多序列比对结果作为Evoformer模块的提示词输入,结合同源序列扰动机制,无需重新训练即可生成多样化的局部构象.此外, FoldCopilot利用模板与结构初始化帧,通过结构预测模块的迭代过程实现了对模型生成结构的演化与控制.实验结果表明,该方法在蛋白质结构修复和局部多构象生成等任务中表现出色,为蛋白质-分子相互作用预测和结构建模提供了新的思路,并为后续的分子动力学模拟和生物学研究提供了强有力的支持. 展开更多
关键词 AF2 提示词工程 多构象采样 动力系统控制 同源序列扰动
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部