基于平均距离聚类的NSGA-Ⅱ

拥挤度距离是一种用于度量解集多样性的指标. 然而, 在许多情况下, 该指标无法有效区分多样性较优个体. 其原因为拥挤度距离主要利用每个位置的局部信息. 为解决该问题, 基于整个种群全局位置信息, 本文设计了基于平均距离聚类的多样性度量指标, 并进一步提出了基于平均距离聚类的NSGA-Ⅱ. 该算法利用平均距离将种群划分为若干个大致均匀分布的小种群, 然后分别在各小种群内执行选择、交叉和变异等操作. 实验结果表明, 本文所提算法可以有效地保持种群多样性.     

Effect of Different Crowding Agents on the Architectural Properties of the Bacterial Nucleoid-Associated Protein HU

HU is a nucleoid-associated protein expressed in most eubacteria at a high amount of copies (tens of thousands). The protein is believed to bind across the genome to organize and compact the DNA. Most of the studies on HU have been carried out in a simple in vitro system, and to what extent these observations can be extrapolated to a living cell is unclear. In this study, we investigate the DNA binding properties of HU under conditions approximating physiological ones. We report that these properties are influenced by both macromolecular crowding and salt conditions. We use three different crowding agents (blotting grade blocker (BGB), bovine serum albumin (BSA), and polyethylene glycol 8000 (PEG8000)) as well as two different MgCl₂ conditions to mimic the intracellular environment. Using tethered particle motion (TPM), we show that the transition between two binding regimes, compaction and extension of the HU protein, is strongly affected by crowding agents. Our observations suggest that magnesium ions enhance the compaction of HU–DNA and suppress filamentation, while BGB and BSA increase the local concentration of the HU protein by more than 4-fold. Moreover, BGB and BSA seem to suppress filament formation. On the other hand, PEG8000 is not a good crowding agent for concentrations above 9% (w/v), because it might interact with DNA, the protein, and/or surfaces. Together, these results reveal a complex interplay between the HU protein and the various crowding agents that should be taken into consideration when using crowding agents to mimic an in vivo system.     

多目标进化算法中基于动态聚集距离的分布性保持策略 *

提出了基于动态聚集距离 ( DCD)的分布性保持策略 ,利用个体在不同维目标上聚集距离的差异程度来定义 DCD,并在种群维护中动态地计算 DCD。与目前经典算法 NSGA-II和ε -MOEA进行比较 ,实验结果表明 DCD能在较大程度上提高分布性 ,并得到较好的收敛性。     

基于档案交叉的动态多目标粒子群优化算法

为保证在动态环境中及时跟踪到最新的真实Pareto前沿,保持解集的均匀性,提出一种基于档案交叉的动态多目标粒子群优化算法。着重利用保存在外部档案的最新非劣解,对这些非劣解进行交叉操作以增加种群的多样性,促进档案中个体信息的交流;提出一种高效的欧氏拥挤距离策略,并将其应用于对外部档案的维护;修改粒子群算法模型使之更适用于动态多目标优化。实验结果表明,该算法能适应动态环境,快速跟踪动态Pareto面,解集均匀性良好。     

基于随机黑洞和逐步淘汰策略的多目标粒子群优化算法

提出一种基于随机黑洞粒子群算法(RBH-PSO) 和逐步淘汰策略的多目标粒子群优化(MRBHPSO-SE) 算法. 利用RBH-PSO 全局优化能力强和收敛速度快的优点逼近Pareto 最优解; 为了避免拥挤距离排序策略的缺陷, 提出逐步淘汰策略, 并将其应用到下一代粒子的选择策略中. 同时, 动态选择领导粒子, 运用动态惯性权重系数和变异操作 来增强种群全局寻优能力, 以及避免早熟收敛. 利用具有不同特点的测试函数进行验证, 结果表明, 与同类算法相比, 该算法具有较高的精度并兼顾优化解的多样性.

     

一种改进的小生境多目标粒子群优化算法

提出一种小生境多目标粒子群优化算法。使用环邻域拓扑且无需任何小生境参数,克服常规小生境技术中需确定小生境参数的困难。采用NSGA-II的非支配排序策略和动态加权方法选择最优粒子。基于拥挤度的变异操作引导粒子跳出局部最优,增强算法的全局搜索能力。通过对ZDT1~ZDT4和ZDT6的测试结果表明,与经典的多目标进化算法NSGA-II、PESA-II和MOPSO相比,该算法在最优解集的收敛度与多样性方面具有明显的优势。     

Immigration and location choices of native‐born workers in Canada

This paper investigates a possible crowding‐out effect of immigration in Canadian labour markets and explores how location choices of native‐born workers can be influenced by industry and occupation specific immigration clustering in both the potential destinations and the departure regions. We apply choice‐specific, clustered fixed‐effect response models. The results show that industry‐specific immigration clustering indices have strong and negative effects on the location choices of the native born. When the scaled immigration measures are used, the results confirm the ‘substitution’ hypothesis: native‐born workers who choose lower immigration in their destinations also move from the origins with higher immigration in their industry.     

Diverse Effects of Macromolecular Crowding on the Sequential Glycan‐Processing Pathway Involved in Glycoprotein Quality Control

Compared with in vitro conditions, the intracellular environment is highly crowded with biomolecules; this has numerous effects on protein functions, including enzymatic activity. We examined the effects of macromolecular crowding on glycan processing of N‐glycoprotein in the endoplasmic reticulum as a model sequential metabolic pathway. Experiments with synthetic substrates of physiological glycan structure clearly showed that the first half of the pathway (glucose trimming) was accelerated, whereas the second (mannose trimming) was decelerated under molecular crowding conditions. Furthermore, calreticulin, a lectin‐like molecular chaperone, bound more strongly to a glycan‐processing intermediate under these conditions. This study demonstrates the diverse effects of molecular crowding on sequential enzymatic processing, and the importance of the effects of macromolecular crowding on in vitro assays for understanding sequential metabolic pathways.     

大分子拥挤下介孔中木瓜蛋白酶的微波辅助固定化

为了增强酶的固定化效果,通过添加大分子试剂,在微波辐射作用下将木瓜蛋白酶固定在介孔泡沫硅的孔道中. 结果表明,在加酶量为400 mg/g时,微波辐射下木瓜蛋白酶与牛血清白蛋白(BSA)共固定化制得的固定化酶催化效果最好. 当BSA含量为加酶量的5%(w)时,固定化酶表观活力高达419.1 U/mg,相对活力和酶活回收率分别为126.0%和119.1%. 影响固定化酶活力的主要因素依次为加酶量、BSA含量、微波功率和固定化pH. 该固定化酶的最适反应pH为7.0,最适反应温度为75℃,热稳定性优于游离酶和未加入BSA的固定化酶,80℃下热处理3 h,剩余活力仍为初始活力的88.2%.