科学网

 找回密码
  注册
科学网 标签 ESS

tag 标签: ESS

相关帖子

版块 作者 回复/查看 最后发表

没有相关内容

相关日志

ABBS: Differentiation of rat iPS cells and ES cells into gra
chshou 2015-11-6 09:21
Differentiation of rat iPS cells and ES cells into granulosa cell-like cells in vitro Juan Zhang, Hui Li, Zhao Wu, XiaoJun Tan, Fengying Liu, Xianghong Huang and Xiaoling Fang Acta Biochim Biophys Sin 2013, 45: 289–295; doi: 10.1093/abbs/gmt008 Department of Obstetrics and Gynecology, Second Xiangya Hospital, Central South University, Changsha 410083, China Premature ovarian failure (POF) is an ovarian defect characterized by the premature depletion of ovarian follicles before 40 years of age, representing one major cause of female infertility. Stem cells provide the possibility of a potential treatment for POF. In this study, rat embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs) were co-cultured with granulosa cells (GCs) to differentiate to GC-like cells. The level of estradiol (E2) analyzed by radioimmunoassay showed that the E2 concentration of the culture supernatant of co-cultured rat iPSCs and ESCs increased in a time-dependent manner, compared with the GCs group that has an opposite trend. The expression of follicle-stimulating hormone receptor (FSHR) was confirmed by immunostaining. These results indicated that rat iPSCs and ESCs were effectively induced to GC-like cells through indirect cell-to-cell contact. Real-time polymerase chain reaction was performed to analyze the expression level of marker genes in POF, including BMP15, FMR1, FSHR, INHA, AMH, NOBOX, FOXO3, EIF2B, FIGLA, and GDF9. The BMP15, FSHR, INHA, AMH, NOBOX, and GDF9 genes were significantly up-regulated in iPSCs and ESCs co-cultured with GCs in comparison with cells that were not co-cultured. Thus, here we demonstrated an available method to differentiate rat iPSCs and ESCs into GC-like cells in vitro for the possible cell therapy of POF. Expression of FSHR in iPSCs and ESCs co-cultured with GCs 全文: http://abbs.oxfordjournals.org/content/45/4/289.full 相关论文: 1 Ovarian adult stem cells: hope or pitfall? 2 Efficient Differentiation of Steroidogenic and Germ-Like Cells from Epigenetically-Related iPSCs Derived from Ovarian Granulosa Cells 3 Generation of Germ Cells In Vitro in the Era of Induced Pluripotent Stem Cells 4 Efficient Induction of Pluripotent Stem Cells from Granulosa Cells by Oct4 and Sox2 5 Optimal vitrification protocol for mouse ovarian tissue cryopreservation: effect of cryoprotective agents and in vitro culture on vitrifiedwarmed ovarian tissue survival 6 Sex-determining region Y-box 2 and GA-binding proteins regulate the transcription of liver receptor homolog-1 in early embryonic cells 7 EDA-Containing Fibronectin Increases Proliferation of Embryonic Stem Cells
个人分类: 期刊新闻|1527 次阅读|0 个评论
凝聚法层次聚类之ward linkage method
热度 1 Songjinghe 2015-9-18 22:22
凝聚法分层聚类中有一堆方法可以用来算两点(pair)之间的距离:欧式,欧式平方,manhattan等,还有一堆方法可以算类(cluster)与类之间的距离,什么single-linkage、complete-linkage、还有这个ward linkage。(即最短最长平均,离差平方和) 其他的好像都挺好理解,就是最后这个有点麻烦。。。 这个方法说白了叫离差平方和(这是个啥?)。是ward写那篇文章时候举的一个特例。这篇文章是说分层凝聚聚类方法的一个通用流程。在选择合并类与类时基于一个object function optimise value,这个object function可以是任何反应研究目的的方程,所以许多标准的方法也被归入了。为了阐明这个过程,ward举了一个例子,用的object function 是error sum of squares(ESS),这个例子就成为ward's method。 找了N多资料,终于把这个算法的过程搞清楚了。首先输入的是一个距离矩阵,知道每两个点之间的距离。然后初始化是每个点做为一个cluster,假设总共N组,此时每个组内的ESS都是0,ESS的公式,如下(从原稿《Hierarchical Grouping To Optimize An Objective Function》上摘的): 我当时还有点蒙ESS是个啥?——我现在知道了,凡是蒙的都是概率没学好(我是说我)……先从wiki上转个公式过来: 这是方差的公式,写的再通俗点,就是: 等号两边同时乘上n,好了,你应该知道ESS是啥了——ESS就是【方差×n】!so easy了~~ 但是等下——这看起来是个一维的公式啊——因为你已经知道ESS是【方差×n】了,那多维的还不会算吗?先求所有点的均值点 ,然后再算所有点到这个均值点(central)的距离(距离公式你得自己定,见开头,但是最后算出来就是一个数),然后把所有距离平方后加起来(此时即为方差乘上n),就得到ESS了。 说了半天光说ESS了,列位看官,人只有一张嘴,故ESS此处按下不表,接着说ward method。ward method是要求每次合并后ESS的增量最小,这怎么讲呢?还是上图吧(图是从youtube上的一个 教程 里截的): 只看最下面ward's method的两个图好了,先看下面的图,合并前红色组和黄色组分别能算各自的ESS,总的ESS是什么呢?很简单,加起来就好了,即: ESS(总-合并前)=ESS(红)+ESS(黄)+ESS(其他没画出来的组) 如果合并这两个组,则可以作为一个新组再算一个ESS,此时 ESS(总-合并后)=ESS(红黄)+ESS( 其他没画出来的组) 你注意这里还没有真的合并,只是算了一下合并红黄两组的“成本”(即:ESS(总-合并后)-ESS(总-合并前),当然这个成本肯定是增加的),如果总共有N个组,必须把每两个组合并的成本都算一遍,也就是算N×(N-1)/2个数出来(是不是感觉运算量很大?不要紧,有 快速算法 ),然后找里面合并后成本最小的两组合并。然后再重复这个过程。 我说清楚了吧!? 嗯,至于画的那个树状图的高度,可以认为是上面说的这个“成本”。 对了,还得说一下这个公式: 啥意思呢,就是说,如果用ward's method来度量两个cluster之间的距离,那么两个cluster之间的距离就是把这两个cluster合并后新cluster的ESS,其中x就表示合并前两个cluster中所有点,而 就是合并后那个新cluster的中心点(均值点), 就表示每个点x到中心点的距离,平方后加起来,就是ESS了。 好了,总结一下,ward's method是凝聚法分层聚类中一种度量cluster之间距离的方法。按照这个方法,任意两个cluster之间的距离就是这两个cluster合并后新cluster的ESS——说了这么多还真是惭愧,估计这些东西随便找一个数据挖掘的视频教程估计都有,我居然搞了这么长时间才弄懂。。。。 PS:附一段代码上来(快速算法): http://my.oschina.net/songjinghe/blog/508553
35632 次阅读|1 个评论
A game between an individual and a population: rationale
热度 1 dxd 2012-2-15 19:30
这是本系列的第二篇 (首发于本人 英文博客 ),对上一篇提出的想法作了更细致一点的介绍,并引出后面的话题。下一篇才是最重要的,先在此预告一下~ 2012年4月23日追记:昆明动物所王瑞武老师基于榕小蜂的一系列研究,已经包含了这个思路。参见这两篇论文Wang et al. 2009 ( doi:10.1371/journal.pone.0007802 ), Wang et al. 2011 ( doi: 10.1098/​rsif.2011.0063 )。 看来我还得思考得再远一些,才能写出点东西。 At the first glance, the idea proposed in the last post may seem nothing new if we only consider the residents as the only players of the game. However, we are considering not only the residents but also the government as the opposite player. The overall game here can thus be classified as an Asymmetric Two-players' Game which is played Repeatedly. However, at least one of the two players is actually a population, in which each individual makes choices NOT based on the population benefit, but their own individual benefits. When a single government and its strategy is fixed, the scenario is exactly the same as the traditional multiple players' Prisoners' Dilemma or the Boxed Pigs' Game, where the Tragedy of the Commons is expected to be the outcome under certain assumptions. Now let's look at what happens if there are a plural of governments who are in competition with one another in accumulating their own wealth. The governments who choose the 'conscience' strategy may have a disadvantage at the beginning if only a small part of its residents would resist, because it will lose more money to its residents than other governments who choose to be 'selfish'. However, when generations of the residents passed under such a government, leaving only the obedient residents left because the resistant ones are eliminated naturally, the government will be paying nothing to the residents simply because nobody will claim their money back. In contrast, under a 'selfish' government, the obidient residents will be eliminated after a plural of generations with all the left residents becoming resistent, as the result the government will finally lose the money to its residents's pockets. Finally we see that the selfish governments lose the game and the non-selfish governments win. Notice here that the time span of a generation of the residents must be shorter than that of the governments, which is the prerequisite of the outcome here. This process was introduced briefly in my last post, and in this paragraph with more details. This government-resident model looks perfectly applicable to the interaction between cells and the mitochondia inside them. The mitochondria are the 'residents' living within a cell, the rest part of which is their 'government'. The money is embodied in ATPs, which are allocated between the mitochondria and the rest part (hereafter I will just use 'the cell' for the rest part of it). This allocation, which is actually through a biochemical mechanism, can be seen as the game being played between the mitochondrial genome and the nuclear genome during the macro-scale history of evolution. We know that indeed mitochondria replicate faster than the cell containing them get divided, which is the key prerequisite of the gaming process talked in the above paragraph. Also we know that mitochondria don't 'migrate' between cells in the natural world, which keeps the gaming process simpler than the realistic political issues. Bearing these in mind, we can then go on to talk about how nuclear and mitochondrial genomes (and/or 'epi-genomes') play games in ATPs allocation, and what outcomes are achieved under various conditions (healthy and pathological). To be continued.
个人分类: thoughts|4432 次阅读|2 个评论

Archiver|手机版|科学网 ( 京ICP备07017567号-12 )

GMT+8, 2024-5-1 03:47

Powered by ScienceNet.cn

Copyright © 2007- 中国科学报社

返回顶部