Mutational analysis of the gene for Schizosaccharomyces pombe RNase MRP RNA,mrp1, using plasmid shuffle by counterselection on canavanine

Reverse genetics in fission yeast is hindered by the lack of a versatile established plasmid shuffle system. In order to screen efficiently and accurately through plasmid-borne mutations in the essential gene for the RNA component of RNase MRP, mrp1, we have developed a system for plasmid shuffling in fission yeast using counterselection on canavanine. The system takes advantage of the ability of the Saccharomyces cerevisiae CAN1 gene to complement a Schizosaccharomyces pombe can1-1 mutation. Two general use plasmids were constructed that allow directional cloning and initial selection for histidine before counterselection by canavanine. The strain constructed for plasmid shuffling carries auxotrophic markers for ade6, leu1, ura4 and his3 along with the can1-1 mutation. Using this system we examined several partial deletions and point mutations in conserved nucleotides of Schizosaccharomyces pombe RNase MRP RNA for their ability to complement a chromosomal deletion of the mrp1 gene. The degree of background canavanine resistance as well as plasmid–plasmid recombination encountered in these experiments was sufficiently low to suggest that the system we have set up for counterselection by canavanine in fission yeast using multicopy plasmids will be widely useful.     

基于Web技术实现图文表混排报告的在线制作

提出了一种基于Web技术实现图文表混排报告在线制作的新方法,制表效率获得显著提升。     

Bond-free DNA language classes

The absence of the ability to form a specified set of bonds in a collection of DNA strands is crucial in the design of experiments encoding algorithmic problems as single- or double-stranded DNA. Recently, the specification of the bonding types to be avoided has been formalized by defining bond-free properties. Bond-free properties generalize several bonding properties which have been studied in the context of DNA computing. In this paper, we consider a bond-free property as defining a class of languages. We study the properties of these classes of languages. We develop new unary operators on languages for characterizing bond-free properties exactly, using familiar code-theoretic equations. These new operators provide a new characterization of maximal bond-free properties as well. We also focus on relationships to classes of languages from the theory of codes. Research supported in part by a grant from NSERC.     

一种计算缩减级型广义洗牌网络平均跳距的算法

简单介绍了广义洗牌网络（ＧＳＮ）的结构和分类,着重分析缩减级型ＧＳＮ的平均跳距性能,并提出一种计算平均跳距的算法     

置乱技术在数字水印中的应用

近年来,数字水印技术开始广泛应用于数字图象的版权保护,大量的数字水印方案被提出了出来,本文对置乱技术在数字水印中的应用进行了研究并提出了置乱程度的标准测度,这个标准测度能够帮助我们判断了达到最好的水印鲁棒性对一幅图象应该置乱到什么程度。最后,本文提出两种在水印系统中应用的置乱技术,实验结果表明置乱技术在抵抗剪切攻击方面是非常有效的。     

自由空间光互连洗牌型神经网络模型

首先介绍一种洗牌型自由空间光互连多层全互连神经网络模型。该模型的高神经元／权重比可以极大地压缩神经网络的互连权矩阵IWM（interconnectweightmatrix）。对于具有N2个神经元的单层二维全互连神经网络的IWM为N2×N2,而洗牌型全互连神经网络的IWM仅为4N2log2N。另外,洗牌型全互连神经网络整齐、简单的结构方便了网络的综合,特别是网络隐单元的综合,并且十分适合于神经网络的光学实现。然后描述了采用数字光技术实现光互连的洗牌型神经网络的系统模型、关键芯片结构以及关键技术。本文提出的模型和方法使实现与人脑神经网络规模（104量级）相当的实用化自适应光电子全互连神经网络成为可能。     

一种基于密钥的水印嵌入位置置乱算法

提出了一种基于密钥的数字水印嵌入位置的置乱算法,该算法利用分组移位和混洗操作将水印嵌入位置进行多次置乱。算法简单,实现方便,嵌入位置随机分散,嵌入位置信息的安全仅依赖于一组密钥,在相当程度上达到了较理想的置乱效果,增加了系统的安全性。