GSNormalize(GSEAlm)
GSNormalize()所属R语言包：GSEAlm
Aggregating and calculating expression statistics by Gene Set
汇总和计算基因表达的统计集
译者：生物统计家园网 机器人LoveR
描述----------Description----------
Provides an interface for producing aggregate gene-set statistics, for gene-set-enrichment analysis (GSEA). The function is best suited for mean or rescaled-mean GSEA approaches, but is hopefully generic enough to enable other approaches as well.
提供生产总基因组的统计，分析基因组富集（GSEA）接口。平均或重新调整的平均GSEA方法最适合的功能，但希望是通用的，足以使其他的方法以及。
用法----------Usage----------
GSNormalize(dataset, incidence, gseaFun = crossprod, fun1 = "/",
fun2 = sqrt, removeShift=FALSE, removeStat=mean, ...)
identity(x)
one(x)
参数----------Arguments----------
参数：dataset
a numeric matrix, typically of some gene-level statistics
数字矩阵，通常的一些基因水平的统计数字
参数：incidence
0/1 incidence matrix indicating genes' membership in gene-sets
0/1的发病率矩阵表明基因在基因组的成员
参数：gseaFun
function name for the type of aggregation to take place, defaults to 'crossprod'. See 'Details'
发生聚合类型的功能名称，默认为“crossprod”。请参阅“详细资料”。
参数：fun1
function name for normalization, defaults to "/". See 'Details'
标准化的函数名，默认为“/”。请参阅“详细资料”。
参数：fun2
function name for scaling, defaults to 'sqrt'. See 'Details'
缩放函数名，默认为“开方”。请参阅“详细资料”。
参数：removeShift
logical: should normalization begin with a column-wise removal of the mean shift?
逻辑：应该标准化的开始与均值漂移列明智去除呢？
参数：removeStat
(if above is TRUE) the column-wise statistic to be swept out of 'dataset'.
（如果上面是TRUE）列明智的统计，“集”被扫地出门。
参数：...
Additional arguments optionally passed on to 'gseaFun'.
额外的参数可选传递以“gseaFun。
参数：x
any numerical value
任何数值
Details
详情----------Details----------
In gene-set-enrichment analysis (GSEA), the core step is aggregating (or calculating) gene-set-level statistics from gene-set statistics. This utility achieves the feat. It is tailored specifically for rescaled-sums of the type suggested by Jiang and Gentleman (2007), but is designed as a generic template that should other GSEA approaches. In such cases, at this moment users should provide their own version of 'gseaFun'.
在基因组富集分析（GSEA），核心步骤是聚集（或计算）基因组的水平，从基因组统计统计。此实用程序实现的壮举。它是专门针对重标和由江泽民主席和绅士（2007）建议的类型，但作为一个通用的模板应该GSEA办法。在这种情况下，在这一刻的用户应提供他们自己的版本“gseaFun”。
The default will generate sums of gene-level values divided by the square-root of the gene-set size (in other words, gene-set means multiplied by the square-root of gene-set size). The arithmetic works like this:
默认情况下会产生分为基因水平的基因组大小的平方根值的总和（换句话说，基因组是指基因组大小的平方根）乘以。算术是这样的：
gene-set stat = gseaFun(t(incidence),dataset),...) 'fun1' fun2(gene-set size).
基因组STAT = gseaFun（T（发病率），数据集），...）“FUN1 fun2（基因组大小）。
In case there is a known (or suspected) overall baseline shift (i.e., the mass of gene-level stats is not centered around zero) it may be scientifically more meaningful to look for gene-set deviating from this baseline rather than from zero. In this case, you can set 'removeShift=TRUE'.
万一有一个已知（或怀疑）整体基线漂移（即统计是不是集中在零附近的基因水平的质量），它可能是寻找偏离这一基准，而不是从零基因组科学更有意义。在这种情况下，你可以设置“removeShift = TRUE”。
Also provided are the 'identity' function (identity = function(x) x), so that leaving 'gseaFun' and 'fun1' at their default and setting 'fun2 = identity' will generate gene-set means – and the 'one' function to neutralize the effect of both 'fun1' and 'fun2' (see note below).
还提供了“身份”功能（身份=（X）X），使留在它们的默认设置fun2 =身份“将产生的基因组手段”gseaFun和FUN1 - “一“功能，以消除双方的影响”FUN1和fun2（见下文附注）。
值----------Value----------
'GSNormalize' returns a matrix with the same number of rows as 'incidence' and the same number of columns as 'dataset' (if 'dataset' is a vector, the output will be a vector as well). The respective row and column names will carry through from 'dataset' and 'incidence' to the output.
“GSNormalize与发病率和集的列相同的号码（如果”数据集“是一个向量，输出将是一个向量，以及）相同数量的行返回一个矩阵。各自的行和列名，将进行到输出从集和发病。
'identity' simply returns x. 'one' returns the number 1.
“身份”只是返回x。 一返回数字1。
注意----------Note----------
If you want to create your own GSEA function for 'gseaFun', note that it should receive the transposed incidence matrix as its first argument, and the gene-level stats as its second argument. In other words, both should have genes as rows. also, you can easily neutralize the effect of 'fun1', 'fun2' by setting "fun2 = one".
如果你想创建你自己的GSEAgseaFun功能，请注意，它应该接受调换的关联矩阵，其第一个参数作为它的第二个参数的基因水平的统计。换句话说，都应该有作为行的基因。 ，你也可以很容易地抵消“FUN1”的效果，通过设置“fun2 =”“fun2”。
作者（S）----------Author(s)----------
Assaf Oron
参考文献----------References----------
参见----------See Also----------
gsealmPerm, which relies heavily on this function. The function applyByCategory from the Category package has similar functionality and is preferable when the applied function is complicated. GSNormalize is better optimized for
的gsealmPerm，这在很大程度上依赖于这个功能。功能applyByCategoryCategory包有类似的功能和复杂的应用功能时是最好的。“ GSNormalize更好地优化
举例----------Examples----------
data(sample.ExpressionSet)
lm1 = lmPerGene(sample.ExpressionSet,~sex+type)
### Generating random pseudo-gene-sets[＃生成随机伪基因组]
fauxGS=matrix(sample(c(0,1),size=50000,replace=TRUE,prob=c(.9,.1)),nrow=100)
### "tau-stats" for gene-SET-level type effect, adjusting for sex[＃＃“牛头统计”基因SET级类型的影响，调整性]
fauxEffects=GSNormalize(lm1$coefficients[3,]/sqrt(lm1$coef.var[3,]),incidence=fauxGS)
qqnorm(fauxEffects)
### diagonal line represents zero-shift null; note that it doesn't fit[＃对角线代表注意，它不适合零移空;]
abline(0,1,col=2)
### a better option may be to run a diagonal through the middle of the[＃＃一个更好的选择可能是贯穿中间的对角线]
### data (nonzero-shift null, i.e. type may have an effect but it is the[＃数据（非零移空，即类型可能有作用，但它是]
### same for all gene-sets); note that if any outlier shows, it is a purely random one![＃所有基因组相同）;注意，如果任何离群显示，它是一个纯粹的随机！]
abline(median(fauxEffects),1,col=4)
#### Now try with baseline-shift removal[＃＃＃现在尝试用去除基线移]
fauxEffects=GSNormalize(lm1$coefficients[3,]/sqrt(lm1$coef.var[3,]),incidence=fauxGS,removeShift=TRUE)
qqnorm(fauxEffects)
abline(0,1,col=2)
转载请注明:出自 生物统计家园网(http://www.biostatistic.net)。
注：
注1：为了方便大家学习，本文档为生物统计家园网机器人LoveR翻译而成，仅供个人R语言学习参考使用，生物统计家园保留版权。
注2：由于是机器人自动翻译，难免有不准确之处，使用时仔细对照中、英文内容进行反复理解，可以帮助R语言的学习。
注3：如遇到不准确之处，请在本贴的后面进行回帖，我们会逐渐进行修订。
|
发表于 2012-2-25 21:18:17
