Evolutionary history of the mtDNA 9-bp deletion in Chinese populations and its relevance to the peopling of east and southeast Asia

In total, 1218 Chinese from twelve ethnic groups and nine Han geographic groups were screened for the mtDNA 9-bp deletion motif. The frequency of the 9-bp deletion in all samples was 14.7% but ranged from 0% to 32% in the various ethnic groups. Three individuals had a triplication of the 9-bp segment. Phylogenetic and demographic analyses of the mtDNA hypervariable segment 1 (HVS1) sequences suggest that the 9-bp deletion occurred more than once in China. The majority of the Chinese deletion:haplotypes (about 90%) have a common origin as a mutational event following an initial expansion of modem humans in eastern Asia. Other deletion haplotypes and the three haplotypes with a 9-bp triplication may have arisen independently in the Chinese, presumably by replication error. HVS1 haplotype analysis suggests two possible migration routes of the 9-bp deletion in east and southeast Asia. Both migrations originated in China with one route leading to the Pacific Islands via Taiwan, the other to southeast Asia and possibly the Nicobar Islands. Along both routes of peopling, a decrease in HVS1 diversity of the mtDNA haplotypes is observed. The "Polynesian motif (16217T/C, 16247A/G, and 16261C/T)" and the 16140T/C, 16266C/A, or C/G polymorphisms appear specific to each migration route.

Frequency of the mtDNA 9-bp deletion in Chinese ethnic groups

The 9-bp deletion in the COII/tRNA(Lys) intergenic region (region V) of human mitochondrial DNA was screened in 1521 Chinese from 16 ethnic groups and 9 Hen geographic groups. The highest frequency was found in populations of Miao (32.4%) and Bouyei (30.8

The Acquisition of an Inheritable 50-bp Deletion in the Human mtDNA Control Region Does Not Affect the mtDNA Copy Number in Peripheral Blood Cells

The mitochondrial DNA (mtDNA) control region is believed to play an important biological role in mtDNA replication. Large deletions in this region are rarely found, but when they do occur they might be expected to interfere with the replication of the molecule, thus leading to a reduction of mtDNA copy number. During a survey for mtDNA sequence variations in 5,559 individuals from the general Chinese population and 2,538 individuals with medical disorders, we identified a 50-bp deletion (m.298_347del50) in the mtDNA control region in a member of a healthy Han Chinese family belonging to haplogroup B4c1b2, as suggested by complete mtDNA genome sequencing. This deletion removes the conserved sequence block II (CSBII; region 299-315) and the replication primer location (region 317-321). However, quantification of the mtDNA copy number in this subject showed a value within a range that was observed in 20 healthy subjects without the deletion. The deletion was detected in the hair samples of the maternal relatives of the subject and exhibited variable heteroplasmy. Our current observation, together with a recent report for a benign 154-bp deletion in the mtDNA control region, suggests that the control of mtDNA replication may be more complex than we had thought. Hum Mutat 31:538-543, 2010. (C) 2010 Wiley-Liss, Inc.

基于ICA和BP网络的电流传感器相位差测量方法

提出了一种高精度在线测定电流传感器相位差的方法.在输入端加入测试信号,在输出端采用独立分量分析(ICA)将该测试信号与传感器正常工作的输出信号分离,通过比较该测试信号输出前后相位的变化确定传感器的相位差.但因ICA算法固有的局限性,同一混合信号每次分离结果误差不同,为了提高分离精度,利用BP网络学习分离结果中相位误差与最大似然指标和负熵的关系,提出了一个指示分离结果误差的评价量,以选取多次分离结果中的最优结果.实验结果证明该评价量的有效性.

基于BP神经网络的系统级电源管理算法

在深入研究目前国际上比较流行的几种系统级电源管理（PM）算法的基础上，利用BP神经网络的非线性映射能力，提出基于BP神经网络的、对任务之间相互间隔时间也就是系统空闲时段的长度进行自适应学习的BPPM算法，具有传统回归PM算法不可比拟的优点。仿真实验表明引入神经网络的电源管理算法较之传统PM算法大大降低了系统级功耗。实现了在不需要建立系统模型、无需预先获得负载统计特性的前提下，通过从系统正常工作产生的数据中不断学习，使系统具有自适应、高效的电源管理能力，以达到降低系统功耗、提高器件可靠性、延长工作寿命的目的。

An improved BP algorithm for pattern recognition

An improved BP algorithm for pattern recognition is proposed in this paper. By a function substitution for error measure, it resolves the inconsistency of BP algorithm for pattern recognition problems, i.e. the quadratic error is not sensitive to whether the training pattern is recognized correctly or not. Trained by this new method, the computer simulation result shows that the convergence speed is increased to treble and performance of the network is better than conventional BP algorithm with momentum and adaptive step size.