三亚天涯海角潮间带趋磁细菌特性

在三亚天涯海角潮间带沉积物中发现了一定量的海洋趋磁细菌,最大丰度可达100个/cm3。透射电子显微镜(TEM)观察发现该区域趋磁细菌绝大多数为趋磁球菌。磁小体形状单一,皆为棱柱形;磁小体排列方式多样,以链状排列为主,包括单链、双链,也有少数成簇排列。高分辨率透射电子显微镜(HRTEM)与X-射线能谱仪(EDXS)结果表明,磁小体成分为Fe3O4。趋磁细菌的铁含量平均为2.05%(0.70%~7.53%)。通过16S r RNA基因文库的构建与测序,在78个阳性克隆中得到了16个趋磁细菌序列,分属9个OTU。系统发育分析结果表明,其中8个OTU属于α-变形菌纲,1个OTU属于γ-变形菌纲。在全部9个OTU中,有8个OTU与已知趋磁细菌序列的相似性低于97%,有5个OTU与已知趋磁细菌序列的相似性低于93%,说明该区域存在可观的趋磁细菌资源。     

Magnetospira sp. QH-2 MamK蛋白的结构和系统进化分析

Magnetospira sp.QH-2是一株分离自黄海潮间带的海洋趋磁螺菌,该菌磁小体排列不整齐,并且磁小体链内某些区域排列疏松,推测与其磁小体岛上的mam K编码的Mam K蛋白相关。mam K开放阅读框大小为1086 bp,编码361个氨基酸,Mam K蛋白属于NBD_sugar-kinase_HSP70_actin超家族,为非跨膜蛋白;此外,还利用生物信息学工具预测Mam K蛋白二级和三级结构。系统进化分析显示Mam K蛋白的系统进化和16S r RNA基因的系统进化有一定差异,推测磁小体岛的获得可能与菌种的进化是独立的两个过程;淡水与海水中的趋磁螺菌虽然形态相似,但系统进化地位有一定差异,可能是两者不同生境下适应性演化的结果。     

荣成月湖潮间带单细胞趋磁细菌的多样性研究

Marine magnetotactic bacteria were collected from the intertidal sediments of Yuehu Lake(China), where their abundance reached 103–104 ind./cm3. Diverse morphotypes of magnetotactic bacteria were observed, including cocci and oval, vibrio-, spirillum-, rod-, elliptical-, handle- and bar-shaped forms. The magnetococci were the most abundant, and had flagella arranged in parallel within a bundle. The majority of magnetosomes were arranged in one, two or multiple chains, although irregular arrangements were also evident. All the results of high-resolution transmission electron microscopy(HRTEM) analysis show that magnetosome crystals were composed of Fe3O4, and their morphology was specific to particular cell morphotypes. By the 16 S r RNA gene sequence analysis, we found fourteen operational taxonomic units(OTUs) which were related to magnetotactic bacteria. Among these, thirteen belonged to the Alphaproteobacteria and one to the Gammaproteobacteria.Compared with known axenic and uncultured marine magnetotactic bacteria, the 16 S r RNA gene sequences of most magnetotactic bacteria collected from the Yuehu Lake exhibited sequence identities ranging from 90.1% to96.2%(97%). The results indicate that microbial communities containing previously unidentified magnetotactic bacteria occur in the Yuehu Lake.     

融合空间关联性的特高拱坝位移概率性预测模型

基于机器学习语言建模时,传统方法仅以拟合均方误差（mean square error,MSE）最小为单一优化目标,容易引起过拟合问题。对此,基于关联向量机（relevance vector machine,RVM）,建立了融合变形空间关联性的双优化目标约束下的概率性预测模型。利用形状相似度（shape similarity index,SSI）对拱坝变形的空间关联性进行量化,并将单测点MSE和区域变形SSI相融合,共同作为RVM模型的训练优化目标,以期实现MSE尽可能小,而SSI尽可能大。以锦屏一级拱坝为例,预测均方根误差和最大误差的平均降幅分别为31.2%和24.8%,使用多核函数之后,模型预测性能进一步提升;RVM模型的预测置信带宽明显小于多元回归模型,平均降幅为75.1%,这表明双目标RVM模型可有效提升特高拱坝位移预测的精度和稳定性,并降低不确定性。