发信人: sunnyday (飞鱼), 信区: Biology 标 题: 胡乱选的基因比发表出来的“cancer marker”能更准确的预测癌 发信站: BBS 未名空间站 (Wed Feb 8 14:07:40 2012, 美东) David Venet from the Université Libre de Bruxelles 用胡乱挑选的基因组合 和已经发表的所谓的癌基因的marker 用在未知的病人的数据上，发现胡乱挑选 的基因的准确率比那些所谓的marker 还要好。 文章结尾直接点了几个杂志的名：Cell, Nature Genetics, PNAS, 认为这些杂志 过分鼓吹所谓的 gene marker 的意义。从文章里面的行文来看，估计这些也是 把这个文章据掉的杂志 （lol) 文章最后发表在 PLoS Comput Biol. 2011 Oct;7(10):e1002240. Epub 2011 Oct 20. http://blog.f1000.com/2012/01/10/random-gene-sets-can-predict-b Random gene sets can predict breast cancer survival better than cancer- related signatures And top-tier journals did not want to publish this surprising study. By Ed Yong 10 January 2012 Email Print 1 comment Tumours are bundles of cells that grow and divide uncontrollably, and their genes are deployed in unusual ways. By analysing the genes from different tumour samples, scientists have tried to pin down the chaotic events that lead to cancer. They seem to be making headway. Dozens of papers have reported “gene expression signatures” that predict the risk of dying or surviving from cancer, and new ones come out every month. These signatures purportedly hint at how healthy cells transform into tumours in the first place. If, for example, the genes in question are involved in wound healing, this tells you that the healing process is somehow involved in a tumour’s progression. These collections of genes reveal deeper truths about the disease they’re associated with. This idea sounds reasonable, but David Venet from the Université Libre de Bruxelles has thrown a big spanner into the works. He has shown that completely random sets of genes can predict the odds of surviving breast cancer better than published signatures. Venet found three signatures that are completely unconnected to cancer. Instead, these collections of genes were associated with laughing at jokes after lunch, with the experience of social defeat in mice, and with the positioning of skin cells. All of them were associated with breast cancer outcomes. It got worse. Venet collected 47 breast cancer signatures from published papers and compared them to sets of random genes. The random sets were equally (or more) strongly associated with breast cancer outcomes than 60% of the published ones. In fact, you can randomly select a group of 100 genes or more, and be 90% sure of finding a statistically significant link with breast cancer. Venet wrote, “Investigators are bound to find an association however whimsical their marker is.” Tubular Adenoma of Breast. Image from Flickr, by Ed Uthman Venet’s study was described as a “must-read” by F1000 member Jinfeng Liu from Genentech Inc. The results may seem unbelievable, but there is a simple reason for them. The activities of thousands of genes across a breast cancer cell’s genome are related to how quickly that cell proliferates ( grows and divides). And that is related to a patient’s prognosis. As an analogy, you could find hundreds of things that correlate with a person’s wellbeing and lifespan: the number of Apple products they own, whether they have university degrees, how many cars they have, and so on. But this doesn’t mean that these things improve our health; instead, they reflect how wealthy we are, our lifestyle choices, and our access to good healthcare. Gene signatures may be relatively useless at illuminating the causes of cancer, but the team stresses that they can still help doctors – after all, they’re still related to prognosis. Writing in The Scientist, the study’s lead author Vince Detours says, “Smoke does not drive fire, yet it is powerful indicator of when and where a fire is burning.” Detours also aims a blow at scientific publishers who have let studies of genetic signatures proliferate uncontrollably. He wrote: It took us four years and six rejections to get this work finally published in a computational biology journal – not the most efficient venue to reach the oncology community. Meanwhile, a steady stream of studies confounded by proliferation rates has appeared. He added, This has to be said; one can no longer stay silent about the rather limited self-correction capability of the top tier publishing system (Cell, Nature Genetics, PNAS, etc.), which promoted these studies in the first place -- ※ 修改:·sunnyday 於 Feb 8 14:20:40 2012 修改本文·[FROM: 128.91.] ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.91.]
发信人: KingofMS (你太有才了), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预测癌 发信站: BBS 未名空间站 (Wed Feb 8 14:16:22 2012, 美东) nice 【 在 sunnyday (飞鱼) 的大作中提到: 】 : David Venet from the Université Libre de Bruxelles 用胡乱挑选的基因组合 : 和已经发表的所谓的癌基因的marker 用在未知的病人的数据上，发现胡乱挑选 : 的基因的准确率比那些所谓的marker 还要好。 : 文章结尾直接点了几个杂志的名：Cell, Nature Genetics, PNAS, 认为这些杂志 : 过分鼓吹所谓的 gene marker 的意义。 : http://blog.f1000.com/2012/01/10/random-gene-sets-can-predict-breast-cancer-survival-better-than-cancer-related-signatures/?emailType=Feb2012MonthlyUpdate&utm_source=Email_Monthly_Feb2012&utm_medium=Email&utm_content=EditorsChoice_4&utm_campaign=F1K_Monthly_Feb2012 : Random gene sets can predict breast cancer survival better than cancer- : related signatures : And top-tier journals did not want to publish this surprising study. : By Ed Yong : ................... -- ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 206.81.]
发信人: shakuras (doskey), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预测 发信站: BBS 未名空间站 (Wed Feb 8 14:24:50 2012, 美东) 他那几个基因怎么来的？有没有general的意义还是碰巧？ 【 在 sunnyday (飞鱼) 的大作中提到: 】 : David Venet from the Université Libre de Bruxelles 用胡乱挑选的基因组合 : 和已经发表的所谓的癌基因的marker 用在未知的病人的数据上，发现胡乱挑选 : 的基因的准确率比那些所谓的marker 还要好。 : 文章结尾直接点了几个杂志的名：Cell, Nature Genetics, PNAS, 认为这些杂志 : 过分鼓吹所谓的 gene marker 的意义。从文章里面的行文来看，估计这些也是 : 把这个文章据掉的杂志 （lol) : 文章最后发表在 : PLoS Comput Biol. 2011 Oct;7(10):e1002240. Epub 2011 Oct 20. : http://blog.f1000.com/2012/01/10/random-gene-sets-can-predict-breast-cancer-survival-better-than-cancer-related-signatures/?emailType=Feb2012MonthlyUpdate&utm_source=Email_Monthly_Feb2012&utm_medium=Email&utm_content=EditorsChoice_4&utm_campaign=F1K_ : Random gene sets can predict breast cancer survival better than cancer- : ................... -- ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.135.]
发信人: peoplem (我爱我家), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预测 发信站: BBS 未名空间站 (Wed Feb 8 14:27:37 2012, 美东) Venet found three signatures that are completely unconnected to cancer. Instead, these collections of genes were associated with laughing at jokes after lunch, with the experience of social defeat in mice, and with the positioning of skin cells. 哈哈哈哈哈 【 在 sunnyday (飞鱼) 的大作中提到: 】 : David Venet from the Université Libre de Bruxelles 用胡乱挑选的基因组合 : 和已经发表的所谓的癌基因的marker 用在未知的病人的数据上，发现胡乱挑选 : 的基因的准确率比那些所谓的marker 还要好。 : 文章结尾直接点了几个杂志的名：Cell, Nature Genetics, PNAS, 认为这些杂志 : 过分鼓吹所谓的 gene marker 的意义。从文章里面的行文来看，估计这些也是 : 把这个文章据掉的杂志 （lol) : 文章最后发表在 : PLoS Comput Biol. 2011 Oct;7(10):e1002240. Epub 2011 Oct 20. : http://blog.f1000.com/2012/01/10/random-gene-sets-can-predict-breast-cancer-survival-better-than-cancer-related-signatures/?emailType=Feb2012MonthlyUpdate&utm_source=Email_Monthly_Feb2012&utm_medium=Email&utm_content=EditorsChoice_4&utm_campaign=F1K_ : Random gene sets can predict breast cancer survival better than cancer- : ................... -- ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.32.]
发信人: albertsmwk (.)(.), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预测 发信站: BBS 未名空间站 (Wed Feb 8 14:42:59 2012, 美东) 扫了一眼 大意应该是随便抓100个以上的基因，（包括那些发表的看了joke表达不同的，social 有问题的，皮肤细胞问题的基因）90%的可能cancer和正常细胞的表达谱就会有区别。 作者抓了几十篇paper里面报道的所谓在breast cancer里面表达有区别的基因，和他随 便抓的基因比较，只要抓到100个基因以上，两个结果就相差不显著了。 然后他看了一下发表的那些所谓marker，大部分都和cell cycle/PCNA有关，如果拿 PCNA的表达做normalization，世界清静了，cancer marker基本失效。 结论就是cancer marker就是看细胞分裂的，没啥奇怪。 【 在 sunnyday (飞鱼) 的大作中提到: 】 : David Venet from the Université Libre de Bruxelles 用胡乱挑选的基因组合 : 和已经发表的所谓的癌基因的marker 用在未知的病人的数据上，发现胡乱挑选 : 的基因的准确率比那些所谓的marker 还要好。 : 文章结尾直接点了几个杂志的名：Cell, Nature Genetics, PNAS, 认为这些杂志 : 过分鼓吹所谓的 gene marker 的意义。从文章里面的行文来看，估计这些也是 : 把这个文章据掉的杂志 （lol) : 文章最后发表在 : PLoS Comput Biol. 2011 Oct;7(10):e1002240. Epub 2011 Oct 20. : http://blog.f1000.com/2012/01/10/random-gene-sets-can-predict-breast-cancer-survival-better-than-cancer-related-signatures/?emailType=Feb2012MonthlyUpdate&utm_source=Email_Monthly_Feb2012&utm_medium=Email&utm_content=EditorsChoice_4&utm_campaign=F1K_ : Random gene sets can predict breast cancer survival better than cancer- : ................... -- 如果你启动了电脑之后不知道应该干什么，那么最好先不要用电脑，因为你可能有更重要 的事情需要做。 当老板听完你的proposal想也不想的说“Good, go ahead”，中文翻译成“牛逼，你去 死吧”，请三思而后行。 ※ 修改:·albertsmwk 于 Feb 8 14:47:11 2012 修改本文·[FROM: 137.53.] ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 137.53.]
发信人: shakuras (doskey), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预测 发信站: BBS 未名空间站 (Wed Feb 8 14:46:48 2012, 美东) 抓100个基因有区别没啥问题啊！统计里面你做panel data (longitudinal) 对一堆random的东西regression, 总是抓住一堆dummy variable的 【 在 albertsmwk (.)(.) 的大作中提到: 】 : 扫了一眼 : 大意应该是随便抓100个以上的基因，90%的可能cancer和正常细胞的表达谱就会有区 别。 : 作者抓了几十篇paper里面报道的所谓在breast cancer里面表达有区别的基因，和他随 : 便抓的基因比较，只要抓到100个基因以上，两个结果就相差不显著了。 : 然后他看了一下发表的那些所谓marker，大部分都和cell cycle/PCNA有关，如果拿 : PCNA的表达做normalization，世界清静了，cancer marker基本失效。 : 结论就是cancer marker就是看细胞分裂的，没啥奇怪。 -- ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.135.]
发信人: peoplem (我爱我家), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预测 发信站: BBS 未名空间站 (Wed Feb 8 14:47:21 2012, 美东) 是不奇怪 但是这个不奇怪才真的奇怪 因为大家都可以想象cancer cell和normal cell 应该有巨大差异 整体基因表达就应该有巨大差异 居然如此做biomarker的paper是怎么 发出来的呢？ 【 在 albertsmwk (.)(.) 的大作中提到: 】 : 扫了一眼 : 大意应该是随便抓100个以上的基因，90%的可能cancer和正常细胞的表达谱就会有区 别。 : 作者抓了几十篇paper里面报道的所谓在breast cancer里面表达有区别的基因，和他随 : 便抓的基因比较，只要抓到100个基因以上，两个结果就相差不显著了。 : 然后他看了一下发表的那些所谓marker，大部分都和cell cycle/PCNA有关，如果拿 : PCNA的表达做normalization，世界清静了，cancer marker基本失效。 : 结论就是cancer marker就是看细胞分裂的，没啥奇怪。 -- ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.32.]
发信人: sunnyday (飞鱼), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:09:05 2012, 美东) 我觉得他的文章的意思就是大部分基因的表达程度都和细胞的活跃程度有关， 所以随便乱抓几个都能看出区别来，但是和癌症的机理基本上就是打酱油的关系。 总体意思就是目前的主流方法找出来的marker不能作为推测机理的起点。 他们在faculty 1000上抱怨说他们这篇文章一共投了4年，前后被六个 顶级杂志据了（他们直接点名的几家估计是对他们态度最糟糕的，哈哈哈） 这里是他们在PLoS Comp Biol 发表的文章的摘要。 Bridging the gap between animal or in vitro models and human disease is essential in medical research. Researchers often suggest that a biological mechanism is relevant to human cancer from the statistical association of a gene expression marker (a signature) of this mechanism, that was discovered in an experimental system, with disease outcome in humans. We examined this argument for breast cancer. Surprisingly, we found that gene expression signatures-unrelated to cancer-of the effect of postprandial laughter, of mice social defeat and of skin fibroblast localization were all significantly associated with breast cancer outcome. We next compared 47 published breast cancer outcome signatures to signatures made of random genes. Twenty-eight of them (60%) were not significantly better outcome predictors than random signatures of identical size and 11 (23%) were worst predictors than the median random signature. More than 90% of random signatures >100 genes were significant outcome predictors. We next derived a metagene, called meta-PCNA, by selecting the 1% genes most positively correlated with proliferation marker PCNA in a compendium of normal tissues expression. Adjusting breast cancer expression data for meta-PCNA abrogated almost entirely the outcome association of published and random signatures. We also found that, in the absence of adjustment, the hazard ratio of outcome association of a signature strongly correlated with meta-PCNA (R(2)& #8202;= 0.9). This relation also applied to single-gene expression markers. Moreover, >50% of the breast cancer transcriptome was correlated with meta-PCNA. A corollary was that purging cell cycle genes out of a signature failed to rule out the confounding effect of proliferation. Hence, it is questionable to suggest that a mechanism is relevant to human breast cancer from the finding that a gene expression marker for this mechanism predicts human breast cancer outcome, because most markers do. The methods we present help to overcome this problem. 【 在 peoplem (我爱我家) 的大作中提到: 】 : 是不奇怪 但是这个不奇怪才真的奇怪 因为大家都可以想象cancer cell和normal cell : 应该有巨大差异 整体基因表达就应该有巨大差异 居然如此做biomarker的paper是怎么 : 发出来的呢？ : 别。 -- ※ 修改:·sunnyday 於 Feb 8 19:24:36 2012 修改本文·[FROM: 128.91.] ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.91.]
发信人: ire (ire), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:11:35 2012, 美东) 那些做gene signature的文章用的p值都是设定在什么level？0.05？ 【 在 shakuras (doskey) 的大作中提到: 】 : 抓100个基因有区别没啥问题啊！统计里面你做panel data (longitudinal) : 对一堆random的东西regression, 总是抓住一堆dummy variable的 : 别。 -- ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 165.68.]
发信人: sunnyday (飞鱼), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:12:22 2012, 美东) 一般都比这个严格，经常就是 0.000001 【 在 ire (ire) 的大作中提到: 】 : 那些做gene signature的文章用的p值都是设定在什么level？0.05？ -- ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.91.]
发信人: ire (ire), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:13:38 2012, 美东) GWAS的结果本来就不应该作为推测机理 它理论依据本来就是基于几个假说 就算E-8的显著性阈值都还是假阳性很多 【 在 sunnyday (飞鱼) 的大作中提到: 】 : 我觉得他的文章的意思就是大部分基因的表达程度都和细胞的活跃程度有关， : 所以随便乱抓几个都能看出区别来，但是和癌症的机理基本上就是打酱油的关系。 : 总体意思就是GWAS 做出来的东西不能作为推测机理的起点。 : 他们在faculty 1000上抱怨说他们这篇文章一共投了4年，前后被六个 : 顶级杂志据了（他们直接点名的几家估计是对他们态度最糟糕的，哈哈哈） : 这里是他们在PLoS Comp Biol 发表的文章的摘要。 : Bridging the gap between animal or in vitro models and human disease is : essential in medical research. Researchers often suggest that a biological : mechanism is relevant to human cancer from the statistical association of a : gene expression marker (a signature) of this mechanism, that was discovered : ................... -- ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 165.68.]
发信人: peoplem (我爱我家), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:15:49 2012, 美东) GWAS的文章很多统计都是胡来的 如果p cutoff选0.000001 一般说明作者乱选的统计方法 【 在 sunnyday (飞鱼) 的大作中提到: 】 : 一般都比这个严格，经常就是 0.000001 -- ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.32.]
发信人: shakuras (doskey), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:15:59 2012, 美东) good point. 不过，这个难道不是大家早就应该知道的么？ 本来也就仅仅能是hint可能的机理。另外，其实细胞活跃程度 也不能说是和癌症完全打酱油，倒是找机理的时候把这部分 effect给control掉确实十个good point 【 在 sunnyday (飞鱼) 的大作中提到: 】 : 我觉得他的文章的意思就是大部分基因的表达程度都和细胞的活跃程度有关， : 所以随便乱抓几个都能看出区别来，但是和癌症的机理基本上就是打酱油的关系。 : 总体意思就是GWAS 做出来的东西不能作为推测机理的起点。 : 他们在faculty 1000上抱怨说他们这篇文章一共投了4年，前后被六个 : 顶级杂志据了（他们直接点名的几家估计是对他们态度最糟糕的，哈哈哈） : 这里是他们在PLoS Comp Biol 发表的文章的摘要。 : Bridging the gap between animal or in vitro models and human disease is : essential in medical research. Researchers often suggest that a biological : mechanism is relevant to human cancer from the statistical association of a : gene expression marker (a signature) of this mechanism, that was discovered : ................... -- ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.135.]
发信人: sunnyday (飞鱼), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:19:44 2012, 美东) 其实能在高引用杂志发表出来的，大多数都是 10E -30 这么狠的P value. 【 在 peoplem (我爱我家) 的大作中提到: 】 : GWAS的文章很多统计都是胡来的 如果p cutoff选0.000001 一般说明作者乱选的统计 方法 -- ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.91.]
发信人: shakuras (doskey), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:25:57 2012, 美东) 其实，我觉得做生物的都至少要去上一门stat 101. 一个是大家天天搞的sd, error bar的问题 另一个是搞清楚association和causality的关系 最后，好好理解一下multitest correction 【 在 sunnyday (飞鱼) 的大作中提到: 】 : 其实能在高引用杂志发表出来的，大多数都是 10E -30 这么狠的P value. : 方法 -- ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.135.]
发信人: peoplem (我爱我家), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:37:23 2012, 美东) 这样的P value本质上就是用错统计方法的结果 【 在 sunnyday (飞鱼) 的大作中提到: 】 : 其实能在高引用杂志发表出来的，大多数都是 10E -30 这么狠的P value. : 方法 -- ※ 来源:·BBS 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.32.]
发信人: lischka (smiling cat), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预测癌 发信站: BBS 未名空间站 (Wed Feb 8 15:43:02 2012, 美东) ls 给讲讲multetest correction吧。真心求教。 -- ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 204.187.]
发信人: neverthink (nevernetbug), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:43:30 2012, 美东) Maybe I am not getting it...several quick points: 1. The paper/study set up the assumptions/targets and then shot down... pretty amusing, isn't it? well, they might have a point that there is component of poor scientific publishing... 2. the study went to where it might deserve...I don't see why it shall fair better...that said, it is agreed that some of those NCS are a step away from junk:)))... 3. P value in biomedicine sucks and purely a joke---esp with so many American biologists and beyond whose math intelligence is virtually zero but always insistent on a P value... 4. Who cares if biomarkers have no bearing in mechanisms...I think this is inevitably the trend/true for many complex diseases...otherwise it wouldn't be called biomarkers... 【 在 sunnyday (飞鱼) 的大作中提到: 】 : David Venet from the Université Libre de Bruxelles 用胡乱挑选的基因组合 : 和已经发表的所谓的癌基因的marker 用在未知的病人的数据上，发现胡乱挑选 : 的基因的准确率比那些所谓的marker 还要好。 : 文章结尾直接点了几个杂志的名：Cell, Nature Genetics, PNAS, 认为这些杂志 : 过分鼓吹所谓的 gene marker 的意义。从文章里面的行文来看，估计这些也是 : 把这个文章据掉的杂志 （lol) : 文章最后发表在 : PLoS Comput Biol. 2011 Oct;7(10):e1002240. Epub 2011 Oct 20. : http://blog.f1000.com/2012/01/10/random-gene-sets-can-predict-breast-cancer-survival-better-than-cancer-related-signatures/?emailType=Feb2012MonthlyUpdate&utm_source=Email_Monthly_Feb2012&utm_medium=Email&utm_content=EditorsChoice_4&utm_campaign=F1K_Monthly_Feb2012 : Random gene sets can predict breast cancer survival better than cancer- : ................... -- ※ 修改:·neverthink 於 Feb 8 15:47:16 2012 修改本文·[FROM: 129.112.] ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 129.112.]
发信人: neverthink (nevernetbug), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:46:29 2012, 美东) no hope, you have to go back to the preschool and educate the american and many more---the basic math...even it is possible, it would be a generation away...I am amazed by many biologists who has no math intelligence whatsoever but are solely possessed by P-value when reviewing data/MSs... 【 在 shakuras (doskey) 的大作中提到: 】 : 其实，我觉得做生物的都至少要去上一门stat 101. : 一个是大家天天搞的sd, error bar的问题 : 另一个是搞清楚association和causality的关系 : 最后，好好理解一下multitest correction -- ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 129.112.]
发信人: sunnyday (飞鱼), 信区: Biology 标 题: Re: 胡乱选的基因比发表出来的“cancer marker”能更准确的预 发信站: BBS 未名空间站 (Wed Feb 8 15:50:04 2012, 美东) 【 在 neverthink (nevernetbug) 的大作中提到: 】 : Maybe I am not getting it...several quick points: : 1. The paper/study set up the assumptions/targets and then shot down... : pretty amusing, isn't it? well, they have a point that there is component of : poor scientific publishing... 你说的那个assumptions是因为你在直觉／理论上觉得那个是assumptions。 但是在这个文章之前没有人认真去检查过这个问题。 : 2. the study went to where it might deserve...I don't see why it shall fair : better...that said, it is agreed that some of those NCS are a step away from : junk:)))... : 3. P value in biomedicine sucks and purely a joke---esp with so many : American biologists and beyond whose math intelligence is virtually zero but : always insistent on a P value... -- ※ 修改:·sunnyday 於 Feb 8 19:37:13 2012 修改本文·[FROM: 128.91.] ※ 来源:·WWW 未名空间站 海外: mitbbs.com 中国: mitbbs.cn·[FROM: 128.91.]
Site Map - Contact Us - Terms and Conditions - Privacy Policy 版权所有，未名空间（mitbbs.com），since 1996
