MGMT activated by Wnt pathway promotes cisplatin tolerance through inducing slow-cycling cells and nonhomologous end joining in colorectal cancer

Chemotherapy resistance plays a pivotal role in the prognosis and therapeutic failure of patients with colorectal cancer(CRC).Cisplatin(DDP)-resistant cells exhibit an inherent ability to evade the toxic chemotherapeutic drug effects which are characterized by the activation of slow-cycle programs and DNA repair.Among the elements that lead to DDP resistance,O^(6)-methylguanine(O^(6)-MG)-DNA-methyltransferase(MGMT),a DNA-repair enzyme,performs a quintessential role.In this study,we clarify the significant involvement of MGMT in conferring DDP resistance in CRC,elucidating the underlying mechanism of the regulatory actions of MGMT.A notable upregulation of MGMT in DDP-resistant cancer cells was found in our study,and MGMT repression amplifies the sensitivity of these cells to DDP treatment in vitro and in vivo.Conversely,in cancer cells,MGMT overexpression abolishes their sensitivity to DDP treatment.Mechanistically,the interaction between MGMT and cyclin dependent kinase 1(CDK1)inducing slow-cycling cells is attainted via the promotion of ubiquitination degradation of CDK1.Meanwhile,to achieve nonhomologous end joining,MGMT interacts with XRCC6 to resist chemotherapy drugs.Our transcriptome data from samples of 88 patients with CRC suggest that MGMT expression is co-related with the Wnt signaling pathway activation,and several Wnt inhibitors can repress drug-resistant cells.In summary,our results point out that MGMT is a potential therapeutic target and predictive marker of chemoresistance in CRC.

Effect of hydrogen fluoride and magnesium oxide on AZ31 Mg alloy/carbon fiber-reinforced plastic composite by thermal laser joining technique

Although hydrofluoric acid(HF)surface treatment is known to enhance the joining of metals with polymers,there is limited information on its effect on the joining of AZ31 alloy and carbon-fiber-reinforced plastics(CFRPs)through laser-assisted metal and plastic direct joining(LAMP).This study uses the LAMP technique to produce AZ31-CFRP joints.The joining process involves as-received AZ31,HFpretreated AZ31,and thermally oxidized HF-pretreated AZ31 alloy sheets.Furthermore,the bonding strength of joints prepared with thermally oxidized AZ31 alloy sheets is examined to ascertain the combined effect of HF treatment and thermal oxidation on bonding strength.The microstructures,surface chemical interactions,and mechanical performances of joints are investigated under tensile shear loading.Various factors,such as bubble formation,CFRP resin decomposition,and mechanical interlocking considerably affect joint strength.Additionally,surface chemical interactions between the active species on metal parts and the polar amide along with carbonyl groups of polymer play a significant role in improving joint strength.Joints prepared with surface-pretreated AZ31 alloy sheets show significant improvements in bonding strength.

Study on Key Joining Technology and Test Method of Steel/Al Hybrid Structure Body-in-White

Green and low carbon promote the application and development of light-weight materials in body-in-white. Large-scale die-casting Al alloy (DCAA) and high-strength thermo-formed steel sheet (TFSS) have put forward higher requirements for the application of joining technology of high-strength steel/Al dissimilar materials. Taking the new die-casting Al alloy body as an example, this paper systematically studies the progress of the latest joining methods of steel/Al dissimilar material with combination of two-layer plate and three-layer plate. By analyzing the joining technologies such as FSPR, RES, FDS and SPR, the technology and process characteristics of steel/Al dissimilar material joining are studied, and the joining technical feasibility and realization means of different material combination of the body are analyzed. The conditions of material combination, material thickness, material strength, flange height, preformed holes and joint spacing for achieving high-quality joining are given. The FSPR joining technology is developed and tested in order to meet with the joining of parts with DCAA and TFSS, especially for the joining of three-layer plates with them. It finds the method and technical basis for the realization of high quality joining of dissimilar materials, provides the early conditions for the application of large DCAA and TFSS parts in body-in-white, and meets the design requirements of new energy body. .

并行数据库中JOIN运算的并行算法

JOIN运算的并行算法一直是并行数据库领域中研究的热点问题,先后出现了一系列并行JOIN算法或改进算法,但它们都存在着通信效率较低、负载调度频繁等问题。本文针对这些问题,在分析比较前人工作的基础上对SABJ+算法与ABJ+算法加以改进,得到了效率更高的并行JOIN算法ABJ++。

基于Fork/Join的事务日志伴随模式挖掘方法

信息系统产生的大量事务日志数据蕴含着潜在的伴随模式,伴随模式是指在时空上频繁共现的一组对象.由于传统的滑动窗口算法和FP-Growth算法只能调用单一线程进行计算,随着数据规模的扩张,会导致挖掘伴随模式的时间急剧增加.为此本文提出了一种基于Fork/Join并行技术的伴随模式挖掘框架,其能够实现从单线程到多线程的迁移,充分利用多核配置的加速性能.该框架由划定伴随数据集、频繁项集挖掘和关联规则挖掘三部分组成.首先,提出了基于Fork/Join的多核并行滑动窗口算法,以缩短从事务日志中划定伴随数据集的时间;然后,提出基于Fork/Join的多核并行FP-Growth算法,以并行地挖掘伴随数据集中的频繁项集;最后,引入支持度、置信度和提升度3个参数,对伴随模式中各对象间的关联规则进行挖掘.基于门禁刷卡数据的实验结果表明,相比传统算法,本文所提出的框架能够挖掘出更多的伴随模式,同时挖掘效率较高.

并行数据库上的并行CMD-Join算法

并行数据库在多处理机之间的分布方法（简称数据分布方法）对并行数据操作算法的性能影响很大．如果在设计并行数据操作算法时充分利用数据分布方法的特点，可以得到十分有效的并行算法．本文研究如何充分利用数据分布方法的特点，设计并行数据操作算法的问题，提出了基于ＣＭＤ多维数据分布方法的并行ＣＭＤ－Ｊｏｉｎ算法．理论分析和实验结果表明。

An Overview of Self-piercing Riveting Process with Focus on Joint Failures, Corrosion Issues and Optimisation Techniques

Self-piercing riveting(SPR)is a cold forming technique used to fasten together two or more sheets of materials with a rivet without the need to predrill a hole.The application of SPR in the automotive sector has become increasingly popular mainly due to the growing use of lightweight materials in transportation applications.However,SPR joining of these advanced light materials remains a challenge as these materials often lack a good combination of high strength and ductility to resist the large plastic deformation induced by the SPR process.In this paper,SPR joints of advanced materials and their corresponding failure mechanisms are discussed,aiming to provide the foundation for future improvement of SPR joint quality.This paper is divided into three major sections:1)joint failures focusing on joint defects originated from the SPR process and joint failure modes under different mechanical loading conditions,2)joint corrosion issues,and 3)joint optimisation via process parameters and advanced techniques.

NES-Join算法的改进算法

NES-Join算法是一种无需外排序的连接运算算法,其复杂性优于经典的Sort-Merge Join算法及改进后的SDC-Join算法。在NES-Join算法基础上提出一种改进算法,该算法能够有效压缩原算法中未匹配记录暂存块中的空记录信息,从而使NES-Join算法更具实用性。通过实验和分析表明,改进后的NES-Join算法与原算法时间复杂性相当,但显著提高了磁盘空间的使用率。

并行JOIN算法的优化子集合调整方法

子集合调整是克服数据偏斜影响的并行ＨＡＳＨ－ＪＯＩＮ算法的关键组成部分，现有的克服数据偏斜影响的ＨＡＳＨ－ＪＯＩＮ算法都没有很好地解决这个问题。作者深入地研究了子集合调整问题，提出了四种子集合调整算法。理论分析和实验结果都证明，这四种算法都能够很好地解决子集合调整问题，特别是启发式优化算法ＯＰ－ＡＤＪＵＳＴ在多数情况下都能给出优化解。

N台并联Fork-Join排队网络的弱收敛与强逼近

本文借助于概率测度弱收敛与概率论强逼近理论,较为详细地研究了N台并联Fork-Join排队网络,得到了响应时间、队长、离去过程的弱收敛与强逼近定理。这些结果具有一定的实际意义,并为一般型Fork-Join网络的研究提供了必要的理论基础。

CPU和DRAM加速任务划分方法:大数据处理中Hash Joins的加速实例

硬件加速器能够有效地提高当前计算机系统的能效.然而,传统的硬件加速器(如GPU,FPGA和定制的加速器)和内存是相互分离的,加速器和内存之间的数据移动难以避免,这使得如何降低加速器和内存之间数据移动的开销成为极具挑战性的问题.随着靠近数据的处理技术(near-dataprocessing)和3D堆叠DRAM的出现,我们能够将硬件加速器集成到3D堆叠DRAM中,使得数据移动的开销大大降低.然而,由于3D堆叠DRAM对面积、功耗和散热具有严格的限制,所以不可能将一个功能复杂的硬件加速器完整地集成到DRAM中.因此,在设计内存端的硬件加速器时,应该考虑将加速任务在CPU和加速器之间合理地进行划分.以加速大数据系统中的一个关键操作hash joins为例子,阐述了CPU和内存端加速任务划分的设计思想.以减少数据移动为出发点,设计了一个包含内存端定制加速器和处理器端SIMD加速单元的混合加速系统,并对应用进行分析,将加速任务划分到不同的加速器.其中,内存端的加速器用于加速数据移动受限的执行阶段,而处理器端SIMD加速单元则用于加速数据移动开销较低成本的执行阶段.实验结果表明:与英特尔的Haswell处理器和Xeon Phi相比,设计的混合加速系统的能效分别提升了47.52倍和19.81倍.此外,提出的以数据移动为驱动的方法很容易扩展于指导其他应用的加速设计.

Spark查询引擎中Join操作的优化

Spark是基于Map/Reduce计算模型进行大规模数据处理的分布式系统,每个任务都会被分为很多Map处理和Reduce处理在各个节点上并行执行。Shuffle操作是用于连接Map处理和Reduce处理的桥梁。在对两个大表进行Join操作的过程中,如果两表Join列不完全匹配,Spark中现有的Join实现算法会对大量数据进行shuffle操作,严重影响执行效率。提出一种基于Semi Join思想的Join实现算法——Semi Sort Merge Join,通过对左表Join列数据所构建的HashMap对右表数据进行过滤,可以有效减少Shuffle操作过程中所需传输的数据量。算法分析和实验结果表明,对于Join列数据不完全匹配的大表间Join操作,该算法能有效减少Shuffle操作的开销,右表与左表匹配数据量越少,算法优化的效果越明显。

调度Fork-Join任务图的贪心算法

任务调度算法的目标是把组成并行程序的一组任务分配到多个处理器以使得程序的完成时间最短,这是一个NP完全问题。虽然许多算法在任务满足某些条件时能产生最优调度,但大多都忽略了节省处理器个数和最小化程序总的完成时间等问题。Fork-Join结构是一种并行处理的基本结构。因此,专门针对Fork-Join任务图,提出了一个能产生最优调度的新的贪心调度算法,该算法具有高的加速比和总体效率,时间复杂度为2,其中,表示任务集中任务的个数。实验结果表明,相比其它算法,该算法具有较短的调度长度、较短的完成时间,使用的处理器数较少。

一个调度Fork-Join任务图的最优算法(英文)

Fork-Join任务图是一种并行处理的基本结构.虽然许多算法在任务满足某些条件时能产生最优调度,但往往没有考虑节省处理器个数和减少任务集的总完成时间,从而降低算法的加速比和效率.因此,提出一种基于任务复制的平衡调度算法,其时间复杂度为O(vq+vlogv),v和q分别表示任务集中任务的个数和使用的处理器个数.通过分析已用处理器的负载和空闲时间段,把任务尽量分配到已用的处理器上以均衡负载,从而提高其利用率.实验结果表明,该算法的加速比和总体效率优于其他算法.因此,该算法对于高性能应用程序的调度是一个较好的选择.

并行数据库的改进Hash划分方法及并行Join算法

文中提出了Ｈａｓｈ 划分的改进方法——ＩＨ 划分． ＩＨ划分为结点扩充时数据的重新划分提供了方便．在论述ＩＨ 划分的基础上，给出了基于该数据划分方法的并行Ｊｏｉｎ 算法．利用已有的数据分布，文中提出的并行Ｊｏｉｎ 算法提高了算法的效率．最后，从理论上对以上并行算法的计算复杂性进行了分析．

排序合并Join算法的新结果

Ｊｏｉｎ操作是数据库中最昂贵和最常用的操作．排序合并Ｊｏｉｎ算法是实现Ｊｏｉｎ操作的重要算法，得到了普遍接受并广为应用．在重新研究了排序合并Ｊｏｉｎ算法后发现，同时对两个Ｊｏｉｎ关系进行外排序是不必要的，会带来很大的额外开销．针对这个问题，提出了一种基于单关系外排序的分治Ｊｏｉｎ算法，并在该算法的基础上提出了基于单关系外排序的并行分治Ｊｏｉｎ算法．理论和实验结果证明，基于单关系排序的分治Ｊｏｉｎ算法高于排序合并Ｊｏｉｎ算法．特别是在并行计算环境下，基于单关系排序的并行分治Ｊｏｉｎ算法的效率远远高于排序合并Ｊｏｉｎ算法的并行版本．

WPDL中的JOIN语义问题和分区解决方案

The semantic of JOIN in workflow process definition language WPDL is analyzed and studied. It points outthat there is a problem about AND-JOIN. To solve the problem, the synchronized semantic of AND-JOIN in extend-ed Xinpai-driven model is defined formally. It puts forward the concepts of synchronizedarea and asynchronizedarea. It gives the solution of true-and-false token rules in synchronized area and true token rules in asychronized area,and some other issues arisen from the solution are studied systematically and completely. The constrain issues of con-trol structures in synchronized area and asynchronized area are discussed, and its solution is put forward. Finally, thealgorithm of how to find the focus point and synchronized area is given.

面向Fork/Join框架的软件重构及性能分析

针对目前对于Fork/Join框架应用和性能分析的相关工作还不多的现状,以JGF基准测试程序套件为基础,对其中的series、crypt、sparsematmult和sor等程序使用Fork/Join框架进行重构,并以series程序为例,详细地说明了重构的过程。在实验中,首先,测试了每个程序在不同阈值下使用Fork/Join框架分别递归1、2、3次执行程序的时间,进而选择相对较好的阈值;然后,对每个程序使用Fork/Join框架和使用Thread的执行时间进行了对比;此外,测试了重构后的程序在执行过程中任务窃取的情况。实验结果表明,Fork/Join框架执行时间与多线程执行时间相比,平均降低了14.2%;对于series程序,当数据大小为size C且线程个数为2时,Fork/Join框架执行时间比多线程执行时间降低高达40%,可见,在多核处理器平台上应用Fork/Join框架比使用多线程将获得更好的性能。

基于并行B^+-树的并行Join算法的设计、分析与实现

B＋-树是一种有效的数据库存储结构，被普遍应用于各种关系数据库系统．把B＋-树并行化，使之用于并行数据库系统显然是一项很有意义的重要工作．本文研究了适用于并行数据库的并行B＋-树存储结构，提出两类基于并行B＋-树的并行Join算法．理论和实验结果表明，这些算法效率高于其它并行Join算法．

后期负载调整：一个并行JOIN动态负载平衡算法

本文针对以往的并行ＪＯＩＮ负载平衡策略所存在的缺陷，根据并行ＪＯＩＮ负载平衡的特点和要求，提出一个在ＪＯＩＮ操作后期对负载动态调整的算法。通过性能评估说明，该算法具有预处理开销少、灵活的自适应能力、负载平衡效果理想等特点，同时算法简捷、易于实现。