plw(plw)
plw()所属R语言包：plw
Probe level Locally moderated Weighted median-t.
探针的水平，局部缓和加权中位数-T。
译者：生物统计家园网 机器人LoveR
描述----------Description----------
Computes locally moderated weighted median t-test for microarray data.
本地主持加权平均t检验计算芯片数据。
用法----------Usage----------
plw(x,design=rep(1,ncol(x)),contrast=matrix(1),
probenames = unlist(ifelse(class(x) == "AffyBatch",
list(p = probeNames(x)),
list(p = NULL))),
maxIter = 200, epsilon = 1e-06, verbose = TRUE,
nknots = 10, nOut = 2000, nIn = 4000, knots = NULL,
checkRegulation = TRUE)
参数----------Arguments----------
参数：x
Data, log2(PM) intensities or an AffyBatch object, see details
数据，为log2（下午）强度或AffyBatch的对象，查看详细信息
参数：design
design matrix
设计矩阵
参数：contrast
contrast matrix
对比矩阵
参数：probenames
If not null, it is used to group PM probes into probe sets, see details.
如果不为空，它用于为探针套组下午探针，看到的细节。
参数：maxIter
maximum number of iterations
最大迭代次数
参数：epsilon
convergence criteria
趋同标准
参数：verbose
print computation info or not
打印计算信息或不
参数：nknots
Number of knots of spline for v
v样条节数
参数：nOut
Parameter for calculating knots, see getKnots
为计算结参数，getKnots“
参数：nIn
Parameter for calculating knots, see getKnots
为计算结参数，getKnots“
参数：knots
Knots, if not NULL it overrides nknots, nOut and nIn
疙瘩，如果不为NULL，它覆盖nknots，NOUT和NIN
参数：checkRegulation
If TRUE, data is checked for a correct specified contrast (see details)
如果是TRUE，数据指明了正确的对比检查（见详情）
Details
详情----------Details----------
This function computes the Probe level Locally moderated Weighted median-t statistic (PLW) described in Astrand (2007b), specially design for Affymetrix type data, or other microarray data with multiple probes.
此函数计算探针水平局部缓和加权中位数t统计量（规划地政及工程）A股（2007年b），专门设计的Affymetrix公司类型的数据，或与多个探针的微阵列数据。
The data object x should be either a matrix of perfect match (PM) intensities, or an object of class AffyBatch. When x is a matrix of PM intensities, the intensities should be background corrected, normalized, and logged (with base 2). If x is an AffyBatch object, the default background correction and normalization of RMA is applied to x.
数据对象的x应该是一个完美匹配的矩阵（PM）的强度，或类AffyBatch的对象。当x是一个矩阵下午强度，强度应该纠正的背景下，规范化，并记录（基数为2）。如果x是AffyBatch的对象，默认背景校正和标准化的RMA应用到X。
When probenames is not null, it should be a vector of length equal to the number rows in the matrix x, giving the probe-set identity for each PM probe. Use the function probeNames in the affy package to get probenames when x is a matrix of log2(PM) intensities.
当probenames不是空的，它应该是一个向量在矩阵X的行数等于的长度，每个下午探针探针集身份。使用在affy包的功能probeNames，当x是一个矩阵为log2（下午）强度得到probenames。
Inference is done for each PM probe, thus moderated t-statistic, p-value and log2(FC) is calculated for each probe. The median t-statistics for each probe-set is also computed.
推论是做每个下午探针，从而缓和t-统计量，P-值的log2（FC）计算每个探针。每个探针组中位数为t-统计量计算。
Each PM probe g (row of x) is modeled as:
每个下午探针G（x的行）为蓝本：
where v is function of the mean intensity: v(mean(mu_g)), N denotes a multivariate normal distribution, Sigma is a covariance matrix and InvGamma(a,b) is the inverse-gamma distribution with density function
v平均强度的功能：v(mean(mu_g))，N表示多元正态分布，Sigma是协方差矩阵和InvGamma(a,b)是逆伽马分布密度函数
Given the design matrix D, mu_g equals D*gamma_g, and given the contrast matrix C the hypothesis C*gamma_g=0 is tested. C should be a one row matrix of same length as the column vector gamma_g.
由于设计矩阵D，mu_g等于D*gamma_g和对比度矩阵C的假设C*gamma_g=0测试。 C应该是一个相同长度的一个行矩阵列向量gamma_g。
See examples on how to specify the design and contrast matrices.
请参阅如何指定设计和对比度矩阵的例子。
A cubic spline is used to parameterize the smooth function v(x)
用三次样条参数光滑函数v(x)
where H:R->R^(2p-1) is a set B-spline basis functions for a given set of p interior spline-knots, see chapter 5 of Hastie et al. (2001).
其中，H:R->R^(2p-1)是一家集B样条基函数为给定的p室内花键结，请参阅第5章哈斯蒂等。 （2001年）。
The parameter estimation procedure is based on the assumption that the specified contrast is close to zero for most genes, or at least that the median contrast over all genes is close to zero. A check is run on data to validate this assumptions. If the checking fails, with the error message "warning: most genes appears to be regulated..." and if YOU ARE SURE that the design and contrast is correct, use checkRegulation=FALSE.
参数估计程序是基于假设指定的对比度是接近零大多数基因，或者至少是对所有基因对比的是中位数接近零。支票上运行的数据，以验证这一假设。 ，如果检查失败，错误消息“警告：大部分基因似乎被监管......”如果你确定，设计和对比度是正确的，使用checkRegulation = FALSE。
值----------Value----------
参数：Sigma
Estimated covariance matrix for y=P' x
y=P' x估计协方差矩阵
参数：m
Estimated shape parameter for inverse-gamma prior for probe variances
形状参数估计为逆伽玛探针差异之前
参数：v
Estimated scale parameter curve for inverse-gamma prior for probe variances
预计规模为反γ曲线参数探针差异
参数：converged
T if the EM algorithms converged
T如果EM算法融合
参数：iter
Number of iterations
迭代次数
参数：modS2
Moderated estimator of probe-specific variances
主持估计探针的具体差异
参数：histLogS2
Histogram of log(s2) where s2 is the ordinary variance estimator
直方图的log（S2）其中S2是普通的方差估计
参数：fittedDensityLogS2
The fitted density for log(s2)
log装密度（S2）
参数：logs2
Variance estimators, logged with base 2.
方差估计，碱基2个记录。
参数：medianT
Median moderated t-statistic for each probe-set
中位数缓和，为每个探针设置的t-统计
参数：t
Moderated t-statistic for each PM probe
主持t-统计每个下午探针
参数：coefficients
Estimated contrast for each PM probe
每个下午探针估计对比
参数：p.value
P-value from the moderated t-statistic for each PM probe
每个下午探针放缓的t-统计的P值
参数：dfT
Degrees of freedom of the moderated t-statistic
度放缓的t-统计的自由
参数：weights
Weights for estimating the contrast
重量估计的对比
参数：P
Transformation matrix
变换矩阵
参数：beta
Estimated parameter vector beta of spline for v(x)
估计参数向量样条线betav(x)
参数：knots
The knots used in spline for v(x)
在样条使用v(x)节
参数：x
The input vector covariate vector x
输入向量的协变量向量x
作者（S）----------Author(s)----------
Magnus <i>A</i>strand
参考文献----------References----------
参见----------See Also----------
estimateSigmaMVbeta, lmw
estimateSigmaMVbeta，LMW
举例----------Examples----------
# ------------------------------------------[------------------------------------------]
# Example analyzing the 6 arrays in the [在6阵列为例分析]
# AffySpikeU95Subset data set[AffySpikeU95Subset数据集]
# Loading the data[加载数据]
data(AffySpikeU95Subset)
# Defining design and contrast matrix[定义设计和对比矩阵]
group<-factor(rep(1:2,each=3))
design<-model.matrix(~group-1)
contrast<-matrix(c(1,-1),1,2)
# Analyzing with an AffyBatch object as input[分析作为输入AffyBatch，对象]
model1<-plw(AffySpikeU95Subset,design=design,contrast=contrast,
epsilon=0.01)
## Look at fitted vs observed density for log(s2)[＃看看装与观测密度log（S2）]
varHistPlot(model1)
## Look at fitted curve for scale parameter[＃看看尺度参数的拟合曲线]
scaleParameterPlot(model1)
## Selecting top genes[＃选择顶部的基因]
topRankSummary(model1,nGenes=10)
## Plotting t-statistics and log2FC for top genes[＃顶级基因绘制的t-统计量和log2FC]
par(mfrow=c(1,2))
plotSummaryT(model1,nGenes=20)
plotSummaryLog2FC(model1,nGenes=20)
###---------------------------------------[＃---------------------------------------]
# Analyzing with BG-adjusted and normalized PM data[与的BG调整和规范化的PM数据分析]
pm1<-pm(bg.correct.rma(AffySpikeU95Subset, bgtype = 2))
pm2<-matrix(.C("qnorm_c", as.double(as.vector(pm1)),
as.integer(nrow(pm1)),
as.integer(ncol(pm1)))[[1]],
nrow(pm1),ncol(pm1))
data<-log2(pm2)
probenames<-probeNames(AffySpikeU95Subset)
model2<-plw(data,design=design,contrast=contrast,
probenames=probenames,epsilon=0.01)
###---------------------------------------[＃---------------------------------------]
# Model1 and model2 should give identical result[model1的Model2应给予相同的结果]
# For example identical top ranking:[例如相同的最高排名：]
range(topRankSummary(model1)$t-
topRankSummary(model2)$t,na.rm=TRUE)
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 2012-2-26 11:16:02
