moduleMergeUsingKME(WGCNA)
moduleMergeUsingKME()所属R语言包:WGCNA
Merge modules and reassign genes using kME.
合并模块,并重新分配使用KME的基因。
译者:生物统计家园网 机器人LoveR
描述----------Description----------
This function takes an expression data matrix (and other user-defined parameters), calculates the module membership (kME) values, and adjusts the module assignments, merging modules that are not sufficiently distinct and reassigning modules that were originally assigned suboptimally.
这个函数有一个表达数据矩阵(和其他用户定义的参数),计算模块成员(KME)值,并调整模块分配,合并不够独特和重新分配模块,最初被分配了次优的模块。
用法----------Usage----------
moduleMergeUsingKME(datExpr, colorh, ME = NULL, threshPercent = 50, mergePercent = 25, reassignModules = TRUE, convertGrey = TRUE, omitColors = "grey", reassignScale = 1, threshNumber = NULL)
参数----------Arguments----------
参数:datExpr
An expression data matrix, with samples as rows, genes (or probes) as column.
表达数据矩阵,行,基因(或探针)列样品。
参数:colorh
The color vector (module assignments) corresponding to the columns of datExpr.
点的颜色矢量(模块分配)对应的列datExpr。
参数:ME
Either NULL (default), at which point the module eigengenes will be calculated, or pre-calculated module eigengenes for each of the modules, with samples as rows (corresponding to datExpr), and modules corresponding to columns (column names MUST be module colors or module colors prefixed by "ME" or "PC").
无论是NULL(默认),点模块特征基因将被计算,或预先计算模块的特征基因,每个模块,作为行(对应datExpr)与样品和模块相应的列(列名必须是模块的颜色前缀为“我”或“PC”)或模块的颜色。
参数:threshPercent
Threshold percent of the number of genes in the module that should be included for the various analyses. For example, in a module with 200 genes, if threshPercent=50 (default), then 50 genes will be checked for reassignment and used to test whether two modules should be merged. See also threshNumber.
应包括用于各种分析模块中的基因的数量的阈值百分比。例如,在与200个基因的模块,如果threshPercent = 50(默认值),然后50个基因将被检查用于再分配和用于测试是否应当合并两个模块。请参阅threshNumber“。
参数:mergePercent
If greater than this percent of the assigned genes are above the threshold are in a module other than the assigned module, then these two modules will be merged. For example, if mergePercent=25 (default), and the 70 out of 200 genes in the blue module were more highly correlated with the black module eigengene, then all genes in the blue module would be reassigned to the black module.
如果大于这个%的所分配的基因是在阈值之上的是在所分配的模块的模块以外的,那么这两个模块将被合并。例如,如果mergePercent = 25(默认值),和蓝色的模块的200个基因中的70,满分的黑色模块eigengene的与更高度,然后蓝色的模块中所有的基因将被重新分配到的黑色模块。
参数:reassignModules
If TRUE (default), genes are resassigned to the module with which they have the highest module membership (kME), but only if their kME is above the threshPercent (or threshNumber) threshold of that module.
如果是TRUE(默认值),基因resassigned的模块,与他们有最高的模块成员(KME),但只有当他们KME以上的threshPercent(或者threshNumber)的阈值,该模块。
参数:convertGrey
If TRUE (default), unassigned (grey) genes are assigned as in "reassignModules"
如果为TRUE(默认值),未分配的(灰色)的基因被分配在“reassignModules”
参数:omitColors
These are all of the module assignments which indicate genes that are not assigned to modules (default="grey"). These genes will all be assigned as "grey" by this function.
这是所有的模块分配未分配模块(默认值=“灰色”)这表明基因。通过此功能,这些基因都将被指定为“灰色”。
参数:reassignScale
A value between 0 and 1 (default) which determines how the threshPercent gets scaled for reassigning genes. Smaller values reassign more genes, but does not affect the merging process.
值介于0和1(默认值),确定被缩放为的threshPercent如何,重新分配基因。较小的值重新分配更多的基因,但不影响合并过程。
参数:threshNumber
Either NULL (default) or, if entered, every module is counted as having exactly threshNumber genes, and threshPercent it ignored. This parameter should have the effect of
无论是NULL(默认),如果输入的,每一个模块计数的为具有完全threshNumber基因,并threshPercent它忽略了。此参数应具有的效果
值----------Value----------
参数:moduleColors
The NEW color vector (module assignments) corresponding to the columns of datExpr, after module merging and reassignments.
新的颜色矢量(模块分配)对应的列datExpr,模块后合并和调动。
参数:mergeLog
A log of the order in which modules were merged, for reference.
的log中的模块顺序进行合并,以供参考。
注意----------Note----------
Note that this function should be considered "experimental" as it has only been beta tested. Please e-mail jeremyinla@gmail.com if you have any issues with the function.
请注意,这个函数应该被认为是“实验性的”,因为它只有beta测试。请e-mail jeremyinla@gmail.com的功能,如果你有任何问题。
(作者)----------Author(s)----------
Jeremy Miller
实例----------Examples----------
## First simulate some data and the resulting network dendrogram[#第一个模拟一些数据,导致网络聚类分析]
set.seed(100)
MEturquoise = sample(1:100,50)
MEblue = sample(1:100,50)
MEbrown = sample(1:100,50)
MEyellow = sample(1:100,50)
MEgreen = c(MEyellow[1:30], sample(1:100,20))
MEred = c(MEbrown [1:20], sample(1:100,30))
#MEblack = c(MEblue [1:25], sample(1:100,25))[MEblack = C(MEblue [1:25],样品(1:100,25))]
ME = data.frame(MEturquoise, MEblue, MEbrown, MEyellow, MEgreen, MEred)#, MEblack)[,MEblack)]
dat1 = simulateDatExpr(ME, 400, c(0.15,0.13,0.12,0.10,0.09,0.09,0.1), signed=TRUE)
TOM1 = TOMsimilarityFromExpr(dat1$datExpr, networkType="signed")
tree1 = flashClust(as.dist(1-TOM1),method="average")
## Here is an example using different mergePercentages, setting an inclusive threshPercent (91)[#下面是一个例子使用不同的mergePercentages,设置一个的包容性threshPercent(91)]
colorh1 <- colorPlot <- labels2colors(dat1$allLabels)
merges = c(65,40,20,5)
for (m in merges) colorPlot = cbind(colorPlot,
moduleMergeUsingKME(dat1$datExpr,colorh1,threshPercent=91, mergePercent=m)$moduleColors)
plotDendroAndColors(tree1, colorPlot, c("ORIG",merges), dendroLabels=FALSE)
## Here is an example using a lower reassignScale (so that more genes get reassigned)[#下面是一个例子,使用较低的reassignScale(让更多的基因得到重新分配)]
colorh1 <- colorPlot <- labels2colors(dat1$allLabels)
merges = c(65,40,20,5)
for (m in merges) colorPlot = cbind(colorPlot,
moduleMergeUsingKME(dat1$datExpr,colorh1,threshPercent=91, reassignScale=0.7, mergePercent=m)$moduleColors)
plotDendroAndColors(tree1, colorPlot, c("ORIG",merges), dendroLabels=FALSE)
## Here is an example using a less-inclusive threshPercent (75), little if anything is merged.[#下面是一个例子使用较少的包容性threshPercent(75),几乎没有什么合并。]
colorh1 <- colorPlot <- labels2colors(dat1$allLabels)
merges = c(65,40,20,5)
for (m in merges) colorPlot = cbind(colorPlot,
moduleMergeUsingKME(dat1$datExpr,colorh1,threshPercent=75, mergePercent=m)$moduleColors)
plotDendroAndColors(tree1, colorPlot, c("ORIG",merges), dendroLabels=FALSE)
# (Note that with real data, the default threshPercent=50 usually results in some modules being merged)[(请注意,与真实的数据,的默认threshPercent = 50通常会导致在某些模块被合并)]
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。
|