使用heatmap.2绘制热图时部分含NA数据集聚类报错问题

用heatmap.2绘制多幅热图时，多数数据集可正常运行，但部分包含NA的数据集会触发报错：

Error in hclustfun(distr) : NA/NaN/Inf in foreign function call (arg 10)

需求是保留NA用于可视化，且已确认所有数据集的列均为数值型，关闭聚类（比如设置dendrogram="none"）时报错数据集可正常绘图。

相关代码 #

heatmap.2(data.matrix(scaled_df), scale="none",trace="none", Rowv=FALSE, dendrogram="column")

可正常运行的数据集 #

structure(c(NA, NA, NA, -0.373055352100063, -0.385706401091696, 
-0.391309347218752, -0.37940898600181, NA, NA, NA, -1.30818865300157, 
-1.28100289342474, -1.27516499611363, -1.29618539092743, NA, 
NA, NA, -0.451429907792099, -0.458270686949654, -0.462138953607995, 
-0.455415387710277, NA, NA, NA, -0.176790480220641, -0.195752707175293, 
-0.203272400570155, -0.186355718130766, NA, NA, NA, 1.50659539820666, 
1.52712982958485, 1.51743516237196, 1.51133642013844, 2.04124145231932, 
2.04124145231932, 2.04124145231932, 0.802868994907711, 0.793602859056535, 
0.81445053513857, 0.806029062631844), .Dim = 7:6, .Dimnames = list(
    c("G1_1", "G1_2", "G1_3", "G2_1", 
        "G2_2", "G2_3", "G2_4"), c("R1", "R2", "R5", 
    "R6", "R4", "R3")))

报错的数据集 #

structure(c(0.261464614279221, 0.255611337873998, 0.726613533871122, 
    0.728606613293338, NA, NA, NA, 0.53883348857398, 0.540410891091193, 
    0.318717521491049, 0.317760075309925, 0.658338893924264, 0.45488158789282, 
    0.454676504730871, 0.55105410441913, 0.552570638827687, 0.326763829307165, 
    0.326621150890079, 0.67112393563164, 0.45683001858747, 0.456604031661036, 
    0.0282153134878549, 0.0288220522147781, 0.3256335271748, 0.326144472474918, 
    0.532476638629399, 0.455536129433743, 0.455571483454085, 0.611222528034844, 
    0.612795125514566, 0.316575462086262, 0.31481866822182, 0.430494870845778, 
    0.369470777918874, 0.369560504987554, NA, NA, NA, NA, -0.432180139707614, 
    0.30045887607367, 0.300788490080856), .Dim = 7:6, .Dimnames = list(
        c("G1_1", "G1_2", "G1_3", "G1_4", "G2_1", 
        "G2_2", "G2_3"), c("R4", "R1", "R2", "R3", 
        "R6", "R5")))

解决方案 #

报错原因是默认的距离计算函数（dist）遇到NA时会返回NA值，导致聚类函数hclust无法处理。要解决这个问题，需要自定义能处理NA的距离计算函数，在计算列间距离时忽略缺失值对，同时保留NA在热图中显示。

方法1：使用vegan包的vegdist函数

# 安装并加载依赖包
install.packages("vegan")
library(vegan)
library(gplots)

# 自定义距离函数：计算列间距离时忽略NA
custom_dist <- function(x) {
  vegdist(t(x), method = "euclidean", na.rm = TRUE)
}

# 调用heatmap.2时指定自定义距离函数
heatmap.2(data.matrix(scaled_df), 
          scale = "none",
          trace = "none", 
          Rowv = FALSE, 
          dendrogram = "column",
          distfun = custom_dist)

方法2：使用stats包的pairwise.complete.obs参数

library(gplots)

# 自定义距离函数：仅使用成对完整的观测值计算距离
custom_dist <- function(x) {
  dist(t(x), method = "euclidean", pairwise.complete.obs = TRUE)
}

# 绘制热图
heatmap.2(data.matrix(scaled_df), 
          scale = "none",
          trace = "none", 
          Rowv = FALSE, 
          dendrogram = "column",
          distfun = custom_dist)

这两种方法都能在保留NA可视化的前提下，正常生成列聚类树，解决报错问题。

内容的提问来源于stack exchange，提问作者potatojj