核子类凸包样本选择方法及其SVM应用

提出一种基于核函数方法的类内训练样本选择方法——核子类凸包样本选择法,并将其用于支持向量机。该样本选择方法通过迭代方法,逐一选择了那些经映射后“距离已选样本”,并将其映射、生成“凸包最远的样本”。实验结果表明,该方法选择的少量样本使支持向量机获得了较高的识别比率,减少了存储需求,提高了分类速度。     

基于正则表示的SQL注入过滤模块设计

研究SQL注入攻击行为及语法特征,采用正则表达式对攻击特征进行描述,在此基础上设计Web服务端SQL注入攻击过滤模块,使Http请求被提交至系统模块处理前实现注入攻击检查。测试结果表明,与单纯基于关键字的过滤相比,基于正则表示的过滤具有更高的识别率和较低的误报率,加载了过滤模块的Web服务器能较好地拦截多种SQL注入攻击,并且服务延迟较小。     

水稻密码子优化的cry2A^＊基因在大肠杆菌中的表达及其表达产物的纯化

通过PCR从克隆载体puC18-3Z/Cry2A*上扩增水稻偏爱型密码子优化的抗虫基因cry2A*,经限制性内切酶Nde I和BamH I双酶切定向插入到原核表达载体pET-28a( ),成功构建了在表达蛋白的N端只带有6个组氨酸标签的融合蛋白表达载体pET-28a( )/Cry2A*,并转入人肠杆菌BL21(DE3)中.通过对其表达条件进行优化,发现在IPTG浓度为0.05mmol/L、诱导时问为3h、诱导温度为20℃的表达条件下目的蛋白大部分以可溶形式进行表达.采用Ni-NTA亲和柱纯化得到高纯度目的蛋白,薄层扫描分析蛋白纯度达到95%.     

赤红球菌组氨酸激酶的融合表达和功能分析

对赤红球菌的组氨酸激酶基因进行密码子优化,将优化后的组氨酸激酶基因（rhks）构建重组表达质粒pGEX-4T-2-rhks。将此质粒导入到大肠杆菌BL21（DE3）中进行异源表达。在25 ℃和1 mmol/L异丙基-β-D-硫代吡喃半乳糖苷诱导条件下,组氨酸激酶融合蛋白（GST-RHK）获得成功表达,并具有催化活性。经谷胱甘肽琼脂糖亲和层析纯化,获得电泳纯的GST-RHK,其中纯化倍数为3.1,得率为19.5%。该蛋白大小约为72.75 kDa,Km、Vmax和Kcat值分别为20.92 μmol/L、0.17 μmol/（L·min）和1.4 min－1。野生型赤红球菌、组氨酸激酶基因增强株sdrhkE和组氨酸激酶基因敲减株sdrhkD在分别含有苯酚、甲苯、氯苯、异辛烷4 种有机溶剂的培养基中培养,菌株sdrhkD的生长情况都优于野生型赤红球菌,菌株sdrhkE的生长情况都低于野生型赤红球菌。本研究为进一步揭示赤红球菌SD3中组氨酸激酶涉及的信号转导途径与赤红球菌有机溶剂耐受性的关联机制提供一定参考依据。     

Recombineering reagents for improved inducible expression and selection marker re-use in Schizosaccharomyces pombe

The fission yeast Schizosaccharomyces pombe is an excellent model organism for cell biology. However, its genetic toolbox is less developed than that of Saccharomyces cerevisiae. In the first part of this study we describe an improved inducible expression vector based on tetracycline regulation of the CaMV35S promoter, which is also capable of chromosomal integration and therefore works in minimal and in rich media. We found that anhydrotetracycline is a superior ligand for induction. Maximum expression levels were observed after 12 h in minimal media (EMM) and after 9 h in rich media (YES), which is faster than the nmt1 promoter system. The system was combined with a convenient recombineering-based subcloning strategy for ease of cloning. In the second part we present four template plasmids, pSVEM-bsd, pSVEM-nat, pSVEM-kan and pSVEM-hph, which harbour four recyclable disruption cassettes based on the Cre recombinase lox71/66 strategy for use in PCR targeting methods. Cre-mediated excision leaves a non-functional mutant lox site in the genome, allowing the reiterative usage of these cassettes for multiple targetings. These cassettes are also configured with dual eukaryotic/prokaryotic promoters so that they can be used for recombineering in E. coli. Amongst other purposes, this permits the rapid and convenient creation of targeting constructs with much longer homology arms for difficult and complex targetings in the Sz. pombe genome.     

两种饲养方式下苏尼特羊肉的氧化稳定性

以舍饲、放牧两种饲养方式下的12?月龄苏尼特羊股二头肌为实验材料,分别测定脂质氧化产物含量、硫代巴比妥酸值、抗氧化能力、抗氧化酶活力以及抗氧化酶相关调控基因表达量等指标并进行比较分析,旨在探索两种饲养方式下苏尼特羊肉的氧化稳定性。结果表明：放牧饲养苏尼特羊肉TBA值高度显著低于舍饲饲养（P＜0.001）;羊肉中主要的脂质氧化产物为己醛、庚醛、壬醛、1-辛烯-3-醇以及2,3-辛二酮,其含量均在放牧饲养羊肉中显著较低（P＜0.05）,表明舍饲饲养羊肉的脂质氧化程度相比放牧饲养羊肉更严重。放牧饲养羊肉的总抗氧化能力（P＜0.01）、铜离子还原能力（P＜0.05）和超氧化物歧化酶（superoxide dismutase,SOD）（P＜0.001）、过氧化氢酶（catalase,CAT）（P＜0.05）、谷胱甘肽过氧化物酶（glutathione peroxidases,GPx）（P＜0.05）活力均显著高于舍饲饲养,说明放牧饲养羊肉中的抗氧化酶活力较高,能有效抑制脂质氧化。通过抗氧化酶相关调控基因表达量的分析得出,舍饲饲养羊肉中脂肪氧合酶基因表达量高度显著高于放牧饲养（P＜0.001）,而SOD（P＜0.001）、CAT（P＜0.05）和GPx（P＜0.05）基因表达量均显著低于放牧饲养,从分子水平验证了放牧饲养羊肉的抗氧化能力较舍饲饲养好。     

Adaptive responses in short-chain fatty acid absorption,gene expression,and bacterial community of the bovine rumen epithelium recovered from a continuous or transient high-grain feeding

The feeding of high-grain diets to dairy cows commonly results in lowered pH and ruminal dysbiosis, characterized by changes in absorption dynamics of short-chain fatty acids (SCFA) across the reticuloruminal wall, epithelial function, and the epithelial bacteria community structure. Therefore, the present study evaluated the effect of high-grain feeding persistence on the absorption kinetics of reticuloruminal SCFA, gene expression in the rumen epithelium, and the associated shifts in the epithelial bacteria in cows recovering from either a long-term continuous high-grain feeding model or a long-term transient high-grain feeding model. In a crossover study design, 8 nonlactating Holstein cows were fed 60% concentrate either continuously for 4 wk (continuous) or with a 1-wk break in the second week of the high-grain feeding (transient). After the high-grain feeding, all animals were fed a diet of 100% forage (recovery) for an additional 8 wk. Rumen papilla biopsies and SCFA absorption measurements were taken at the start of the trial (baseline), after the 4-wk high-grain feeding (49 d), after 2-wk recovery forage feeding (63 d), and after 8-wk recovery forage (105 d). Absorption of SCFA was determined in vivo using the washed and isolated reticulorumen technique. Rumen papillae biopsies were used for adherent bacterial DNA and host RNA extraction. The epithelial bacteria were determined using Illumina MiSeq (Microsynth AG, Balgach, Switzerland) sequencing of the 16S rRNA gene. No significant effects of the high-grain feeding model were seen for bacterial diversity. However, bacterial diversity increased with time spent in the recovery forage feeding period regardless of feeding model. The relative abundance of Acidobacteria phyla and Acetivibrio spp. increased when animals were fed a transient high-grain feeding model. A trend toward increased CLDN4 expression was observed in the continuous model. Furthermore, there were interactions between feeding model and sampling day for gene targets CD14, DRA, NHE2, NHE3, and MCT2. When comparing length of recovery, in the continuous model increased relative absorption of SCFA was sustained at 63 d but dropped to baseline measurements at 105 d. A similar pattern was found with the transient model but it did not reach significance. The only gene target that was found to significantly correlate to relative absorption of SCFA was DRA (correlation coefficient ≤ −0.41). Whereas, genera Alkalibaculum, Anaerorhabdus, Coprococcus, and Dethiobacter all showed positive correlations to gene targets for pH regulation (NHE2 and NHE3) and SCFA uptake (MCT1) but negative correlations to SCFA absorption. We conclude that while the rumen absorption and epithelial bacteria were able to recover to baseline levels after 8 wk of forage feeding, the time needed for re-establishment of homeostasis in host gene expression is longer, especially when high-grain feeding is interrupted.     

大型洞室群稳定性分析与智能动态优化设计的数值仿真研究

 为实现数值仿真计算结果能准确反映大型洞室围岩实际力学行为,结合锦屏二级水电站地下厂房枢纽洞室群稳定性分析,在阐述大型地下洞室群数值仿真计算要点基础上,首先重点从岩体本构模型识别和力学参数识别2个方面详细介绍如何实现数值仿真计算的正确化。采用考虑空间分布协调的多元监测信息的6次岩体力学参数跟踪识别,获得锦屏二级地下厂房岩体的等效力学参数,从而通过参数反分析的方法在一定程度上证明岩体等效力学参数具有相对稳定性和可识别性。同时,结合数值模拟展现出的围岩应力、变形、塑性区等方面计算结果与工程岩体的具体地质条件和洞室群结构特点,论述锦屏二级水电站地下厂房上游高边墙变形较大、下游侧拱围岩与喷混凝土破坏、母线洞环状开裂、交叉洞口局部塌落等洞室围岩的变形与破坏机制。最后,结合锦屏二级地下厂房枢纽洞室群稳定性分析与实践的研究认识,对大型地下洞室修建中诸如结构面密集的高边墙支护问题、围岩应力集中区支护问题、工程区地下水对围岩稳定性影响等问题进行论述,并探讨如何通过数值仿真计算、现场监测与经验丰富的专业人员的有机结合来实现大型地下洞室群稳定性设计的科学化。     

基于多时间尺度复合子带噪声语音识别新方法

本文根据多时间尺度分析与子带方法,提出了一种多时间尺度复合子带的噪声环境下语音识别新方法.新方法在不同的时间尺度下分别进行子带特征提取和全带特征提取,并分别进行识别,然后在识别概率层相结合得到最终识别结果.本方法兼有多时间尺度方法和子带方法的抗噪性能.此外,进一步引入频谱差分方法提高语音特征的抗噪性能.对E-SET在NoiseX92下白噪声的识别实验表明,新方法具有良好的抗噪性能.     

毛针织服装任意衣片的自动识别

为提高毛针织服装CAD系统的智能性,使其能自动识别任意衣片,为后续工艺制作等工序提供计算依据,提出一种新的基于模糊理论的毛针织服装衣片自动识别算法.通过建立基于关键点的衣片数据模型,给出任意衣片与标准衣片的对比方法及基本步骤;通过对任意衣片的等比例变换,构建了基于关键点及基于轮廓点的2种任意衣片与标准衣片的隶属值公式;...