Mycosin-1 protease(MycP1)is a serine protease anchored to the inner membrane of Mycobacterium tuberculosis,and is essential in virulence factor secretion through the ESX-1 type VII secretion system(T7SS).Bacterial phy...Mycosin-1 protease(MycP1)is a serine protease anchored to the inner membrane of Mycobacterium tuberculosis,and is essential in virulence factor secretion through the ESX-1 type VII secretion system(T7SS).Bacterial physiology studies demonstrated that MycP1 plays a dual role in the regulation of ESX-1 secretion and virulence,primarily through cleavage of its secretion substrate EspB.MycP1 contains a putative N-terminal inhibitory propeptide and a catalytic triad of Asp-His-Ser,classic hallmarks of a sub-tilase family serine protease.The MycP1 propeptide was previously reported to be initially inactive and activated after prolonged incubation.In this study,we have deter-mined crystal structures of MycP1 with(MycP124-422)and without(MycP1^(63-422))the propeptide,and conducted EspB cleavage assays using the two proteins.Very high struc-tural similarity was observed in the two crystal structures.Interestingly,protease assays demonstrated positive EspB cleavage for both proteins,indicating that the putative propeptide does not inhibit protease activity.Molecu-lar dynamic simulations showed higher rigidity in regions guarding the entrance to the catalytic site in MycP124-422 than in MycP1^(63-422),suggesting that the putative propeptide might contribute to the conformational stability of the active site cleft and surrounding regions.展开更多
为了解决非线性滤波中量测噪声呈厚尾分布且统计特性不确定的问题,提出一种基于Pearson Type VII分布的自适应滤波算法.针对传统鲁棒卡尔曼滤波器因尺度矩阵和自由度参数固定不变而无法自适应调整的问题,以容积卡尔曼滤波器为基础,选择P...为了解决非线性滤波中量测噪声呈厚尾分布且统计特性不确定的问题,提出一种基于Pearson Type VII分布的自适应滤波算法.针对传统鲁棒卡尔曼滤波器因尺度矩阵和自由度参数固定不变而无法自适应调整的问题,以容积卡尔曼滤波器为基础,选择Pearson Type VII分布对厚尾噪声进行建模,将传统鲁棒滤波固定自由度参数的估计转化为Pearson Type VII分布中可自适应调整的双自由度参数的估计,并通过inverse Wishart和Gamma分布描述尺度矩阵、双自由度参数和辅助参数的先验分布,利用遗忘因子对各参数进行时间更新;基于变分贝叶斯理论,对系统状态、尺度矩阵、双自由度参数和辅助参数形成的联合后验概率密度函数进行变分迭代,实现对系统状态和未知厚尾噪声的联合估计.仿真结果表明,在不确定厚尾噪声条件下,本文算法的滤波精度高于传统鲁棒容积卡尔曼滤波.展开更多
基金This work was supported by funds from the National Basic Research Program(973 Program)(Nos.2011CB911104 and 2012CB917202)the National Natural Science Foundation of China(Grant No.31100538)to F.W.,(Grant No.31170817)to C.T.
文摘Mycosin-1 protease(MycP1)is a serine protease anchored to the inner membrane of Mycobacterium tuberculosis,and is essential in virulence factor secretion through the ESX-1 type VII secretion system(T7SS).Bacterial physiology studies demonstrated that MycP1 plays a dual role in the regulation of ESX-1 secretion and virulence,primarily through cleavage of its secretion substrate EspB.MycP1 contains a putative N-terminal inhibitory propeptide and a catalytic triad of Asp-His-Ser,classic hallmarks of a sub-tilase family serine protease.The MycP1 propeptide was previously reported to be initially inactive and activated after prolonged incubation.In this study,we have deter-mined crystal structures of MycP1 with(MycP124-422)and without(MycP1^(63-422))the propeptide,and conducted EspB cleavage assays using the two proteins.Very high struc-tural similarity was observed in the two crystal structures.Interestingly,protease assays demonstrated positive EspB cleavage for both proteins,indicating that the putative propeptide does not inhibit protease activity.Molecu-lar dynamic simulations showed higher rigidity in regions guarding the entrance to the catalytic site in MycP124-422 than in MycP1^(63-422),suggesting that the putative propeptide might contribute to the conformational stability of the active site cleft and surrounding regions.
文摘为了解决非线性滤波中量测噪声呈厚尾分布且统计特性不确定的问题,提出一种基于Pearson Type VII分布的自适应滤波算法.针对传统鲁棒卡尔曼滤波器因尺度矩阵和自由度参数固定不变而无法自适应调整的问题,以容积卡尔曼滤波器为基础,选择Pearson Type VII分布对厚尾噪声进行建模,将传统鲁棒滤波固定自由度参数的估计转化为Pearson Type VII分布中可自适应调整的双自由度参数的估计,并通过inverse Wishart和Gamma分布描述尺度矩阵、双自由度参数和辅助参数的先验分布,利用遗忘因子对各参数进行时间更新;基于变分贝叶斯理论,对系统状态、尺度矩阵、双自由度参数和辅助参数形成的联合后验概率密度函数进行变分迭代,实现对系统状态和未知厚尾噪声的联合估计.仿真结果表明,在不确定厚尾噪声条件下,本文算法的滤波精度高于传统鲁棒容积卡尔曼滤波.
