Hybrid.Tree(MBCluster.Seq)
Hybrid.Tree()所属R语言包:MBCluster.Seq
Do hybrid-hierarchical clustering for RNA-seq data
做RNA-Seq数据的混合分层聚类
译者:生物统计家园网 机器人LoveR
描述----------Description----------
The hybrid-hierarchical clustering starts from an initial partition of the objects, and merges the small clusters gradually into one tree structure
混合分层聚类分割的对象从最初的开始,并合并成一个树状结构的小聚类逐步
用法----------Usage----------
Hybrid.Tree(data, cluster0, model = "nbinom")
参数----------Arguments----------
参数:data
RAN-seq data standardized by RNASeq.Data()
RAN-Seq数据标准化由RNASeq.Data()
参数:cluster0
A partition of the objects, should be a vector with values ranging from 1 to K0, where K0 is the number of small clusters at the bottom of the hierarchical structure.
一个分区的对象,应该是一个向量,其值的范围从1到K0,其中K0是小聚类的分层结构的底部的数目。
参数:model
The probability models to calculated the distance between to merged clusters
的概率模型来计算的距离之间的合并后的簇
值----------Value----------
a table is returned to keep the information of the tree structure. The table has K rows and 2 columns, where K is the maximum level of the tree, and each row shows the two node being merged in each step
一个表被返回,以保持信息的树结构。该表具有K行和第2列,其中,K是树中的最大电平,并且每行示出了在每个步骤中被合并的两个节点
实例----------Examples----------
###### run the following codes in order[#####运行下面的代码,以]
#[]
# data("Count") ## a sample data set with RNA-seq expressions [数据(“计数”)##的样本数据集与RNA-seq的表达]
# ## for 1000 genes, 4 treatment and 2 replicates[##为1000个基因中,有4治疗,2个重复]
# head(Count)[头(计数)]
# GeneID=1:nrow(Count)[GeneID = 1:NROW(计数)]
# Normalizer=rep(1,ncol(Count))[正规化= REP(1 NCOL(计数))]
# Treatment=rep(1:4,2)[治疗= REP(1:4,2)]
# mydata=RNASeq.Data(Count,Normalize=NULL,Treatment,GeneID) [MYDATA = RNASeq.Data(计数,规范化= NULL,治疗,GeneID)]
# ## standardized RNA-seq data[##标准化RNA-Seq数据]
# c0=KmeansPlus.RNASeq(mydata,nK=10)$centers[C0 = KmeansPlus.RNASeq(MYDATA,NK = 10)$中心]
# ## choose 10 cluster centers to initialize the clustering [##选择10初始化聚类的聚类中心]
# cls=Cluster.RNASeq(data=mydata,model="nbinom",centers=c0,method="EM")$cluster[CLS = MYDATA = Cluster.RNASeq(数据,模型=“nbinom”,中心= C0 =“EM”)$聚类]
# ## use EM algorithm to cluster genes[##使用EM算法簇的基因]
# tr=Hybrid.Tree(data=mydata,cluste=cls,model="nbinom")[TR = Hybrid.Tree(数据= MYDATA,cluste = CLS,模型=“nbinom”)]
# ## bulild a tree structure for the resulting 10 clusters[##bulild的树结构产生的10组]
# plotHybrid.Tree(merge=tr,cluster=cls,logFC=mydata$logFC,tree.title=NULL)[plotHybrid.Tree(合并= TR,聚类=收市,logFC = MYDATA $ logFC,tree.title = NULL)]
# ## plot the tree structure[##图的树状结构]
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。
|