無監(jiān)督分析下鑒定簇集數(shù)及成員
Wilkerson, D. M, Hayes, Neil D (2010). “ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking.” Bioinformatics, 26(12), 1572-1573. http://bioinformatics.oxfordjournals.org/content/26/12/1572.abstract.
1. 關(guān)于 ConsensusClusterPlus
Consensus Clustering 是一種可用于鑒定數(shù)據(jù)集（比如 microarray 基因表達）中的簇集 (clusters) 成員及其數(shù)量的算法。
ConsensusClusterPlus則將 Consensus Clustering 在 R 中實現(xiàn)了。Jimmy大神說這是他見過最簡單的包┑(￣Д ￣)┍
library(ConsensusClusterPlus)
ls("package:ConsensusClusterPlus")
# [1] "calcICL" "ConsensusClusterPlus"
ConsensusClusterPlusfunction for determing cluster number and class membership by stability evidence.
calcICLfunction for calculating cluster-consensus and item-consensus.
2. 好像真的很簡單 只是操作簡單
使用 ConsensusClusterPlus 的主要三個步驟：
- 準備輸入數(shù)據(jù)
- 跑程序
- 計算聚類一致性 (cluster-consensus) 和樣品一致性 (item-consensus)
3. 準備輸入數(shù)據(jù)
首先收集用于聚類分析的數(shù)據(jù)，比如 mRNA 表達微陣列或免疫組織化學(xué)染色強度的實驗結(jié)果數(shù)據(jù)。輸入數(shù)據(jù)的格式應(yīng)為矩陣。下面以 ALL 基因表達數(shù)據(jù)為例進行操作。
library(ALL)
data(ALL)
dataset <- exprs(ALL)
dataset[1:5,1:5]
# 01005 01010 03002 04006 04007
# 1000_at 7.597323 7.479445 7.567593 7.384684 7.905312
# 1001_at 5.046194 4.932537 4.799294 4.922627 4.844565
# 1002_f_at 3.900466 4.208155 3.886169 4.206798 3.416923
# 1003_s_at 5.903856 6.169024 5.860459 6.116890 5.687997
# 1004_at 5.925260 5.912780 5.893209 6.170245 5.615210
取矩陣中 MAD 值 top 5000 的數(shù)據(jù)：
mads <- apply(dataset, 1, mad)
dataset <- dataset[rev(order(mads))[1:5000],]
dim(dataset)
# [1] 5000 128
4. 運行 ConsensusClusterPlus
先設(shè)定幾個參數(shù)：
- pItem (item resampling, proportion of items to sample) : 80%
- pFeature (gene resampling, proportion of features to sample) : 80%
- maxK (a maximum evalulated k, maximum cluster number to evaluate) : 6
- reps (resamplings, number of subsamples) : 50
- clusterAlg (agglomerative heirarchical clustering algorithm) : 'hc' (hclust)
- distance : 'pearson' (1 - Pearson correlation)
# title <- tempdir() ## 雖說是“當前文件夾”，但似乎結(jié)果會輸出到包的安裝路徑...
## 所以還是??
title <- “YOUR PATH”
results <- ConsensusClusterPlus(dataset, maxK = 6,
reps = 50, pItem = 0.8,
pFeature = 0.8,
clusterAlg = "hc",
distance = "pearson",
title = title,
plot = "png")
## 作者這里是pFeature = 1，和前文不符，于是我依然是按0.8輸入計算的
這時工作路徑的文件夾會出現(xiàn)9張圖。
查看一下結(jié)果：
results[[2]][["consensusMatrix"]][1:5,1:5]
# [,1] [,2] [,3] [,4] [,5]
# [1,] 1.00000 0.9375000 1.0000000 0.90625 1.0000000
# [2,] 0.93750 1.0000000 0.9677419 1.00000 0.9393939
# [3,] 1.00000 0.9677419 1.0000000 0.93750 1.0000000
# [4,] 0.90625 1.0000000 0.9375000 1.00000 0.9062500
# [5,] 1.00000 0.9393939 1.0000000 0.90625 1.0000000
results[[2]][["consensusTree"]]
# Call:
# hclust(d = as.dist(1 - fm), method = finalLinkage)
#
# Cluster method : average
# Number of objects: 128
results[[2]][["consensusClass"]][1:5]
# 01005 01010 03002 04006 04007
# 1 1 1 1 1
4.1 一致性矩陣
分別為圖例、k = 2, 3, 4, 5 時的矩陣熱圖。
4.2 一致性累積分布函數(shù)圖
This figure allows a user to determine at what number of clusters, k, the CDF
reaches an approximate maximum, thus consensus and cluster con dence is at
a maximum at this k.
4.3 Delta Area Plot
The delta area score (y-axis) indicates the relative increase in cluster stability.
4.4 Tracking Plot
This plot provides a view of item cluster membership across different k and enables a user to track the history of clusters relative to earlier clusters.
5. 計算聚類一致性 (cluster-consensus) 和樣品一致性 (item-consensus)
icl <- calcICL(results, title = title,
plot = "png")
## 返回了具有兩個元素的list，然后分別查看一下
dim(icl[["clusterConsensus"]])
# [1] 20 3
icl[["clusterConsensus"]]
# k cluster clusterConsensus
# [1,] 2 1 0.9402982
# [2,] 2 2 0.9062500
# [3,] 3 1 0.8504193
# [4,] 3 2 0.9062500
# [5,] 3 3 0.9869781
# [6,] 4 1 0.9652282
# [7,] 4 2 0.9045058
# [8,] 4 3 0.9062500
# [9,] 4 4 0.9728043
# [10,] 5 1 0.9216686
# [11,] 5 2 0.9145987
# [12,] 5 3 0.9062500
# [13,] 5 4 0.9874950
# [14,] 5 5 NaN
# [15,] 6 1 0.9307379
# [16,] 6 2 0.8897721
# [17,] 6 3 0.7474747
# [18,] 6 4 0.8750000
# [19,] 6 5 0.9885269
# [20,] 6 6 0.6333333
dim(icl[["itemConsensus"]])
# [1] 2560 4
icl[["itemConsensus"]][1:5,]
# k cluster item itemConsensus
# 1 2 1 28032 0.9523526
# 2 2 1 28024 0.9366226
# 3 2 1 03002 0.9686272
# 4 2 1 01005 0.9573623
# 5 2 1 04007 0.9549235
5.1 Cluster-Consensus Plot
5.2 tem-Consensus Plot
References
- ConsensusClusterPlus Tutorial https://bioconductor.org/packages/release/bioc/vignettes/ConsensusClusterPlus/inst/doc/ConsensusClusterPlus.pdf
- Nowicka, Malgorzata, et al. "CyTOF workflow: differential discovery in high-throughput high-dimensional cytometry datasets." F1000Research 6 (2017).
最后，向大家隆重推薦生信技能樹的一系列干貨！
- 生信技能樹全球公益巡講：https://mp.weixin.qq.com/s/E9ykuIbc-2Ja9HOY0bn_6g
- B站公益74小時生信工程師教學(xué)視頻合輯：https://mp.weixin.qq.com/s/IyFK7l_WBAiUgqQi8O7Hxw
- 招學(xué)徒：https://mp.weixin.qq.com/s/KgbilzXnFjbKKunuw7NVfw
